INDEX TO  RECORD OF  TECHNICAL INFORMATION/DOCUMENTS
      SUPPORTING THE PROPOSED  RULE  TO AMEND 40  CFR  PART 136
                      TO INCORPORATE METHOD 1613

                                                                      Page
                                                             Volume    No.
I.    Method  1613  and Method Performance Studies

     A.   Method  1613: Tetra- through Octa-                      I-A      001
         Chlorinated Dioxins and Furans by  Isotope
         Dilution HRGC/HRMS, USEPA, Office  of Water
         Regulations and Standards  (OWRS),
         Industrial Technology Division (ITD),
         Revision A, April 1990.

     B.   Performance Evaluation of  Method 1613,                 I-B      045
         USEPA, OWRS, ITD, March 1990.

     C.   Study Plan for the Evaluation  of Method                I-D      202
         1613, USEPA, OWRS, ITD, May 1990.

     D.   Summary Report, USEPA ITD, Method  Detection            I-C      231
         Limit Study for Method 1613 Determination
         of 2,3,7,8-TCDD and 2,3,7,8-TCDF,  USEPA,
         OWRS, ITD, May 1990.

II.   Sources Referenced  in Method 1613

     A.   Method 8290: Analytical Procedures  and                 II-A     240
         Quality Assurance for Multimedia Analysis
         of Polychlorinated Dibenzo-p-dioxins and
         Dibenzofurans by High-Resolution Gas
         Chromatography/High-Resolytion Mass
         Spectrometry, prepared for Environmental
         Monitoring Systems Laboratory-Las  Vegas,
         USEPA, by Yves Tondeur, June 1987.

     B.   Measurement of 2,3,7,8-Tetrachlorinated                II-B     372
         Dibenzo-p-dioxin (TCDD) and 2,3,7,8-
         Tetrachlorinated Dibenzofuran  (TCDF) in
         Pulp, Sludges, Process Samples and
         Wastewaters from Pulp and  Paper Mills, by
         Wright State University, Dayton, Ohio, June
         1988.

     C.   NCASI Procedures for the Preparation and               II-C     414
         Isomer Specific Analysis of Pulp and Paper
         Industry Samples for 2,3,7,8-TCDD  and
         2,3,7,8- TCDF, by National Council  of the
         Paper Industry for Air and Stream
         Improvement, New York, Technical Bulletin
         No. 551, May 1989.

-------
    INDEX TO  RECORD OF TECHNICAL INFORMATION/DOCUMENTS
  SUPPORTING THE PROPOSED RULE TO  AMEND  40 CFR PART 136
             TO INCORPORATE METHOD 1613 (CONT.)

                                                                  Page
                                                         Volume    No.

D.   Analytical Procedures and Quality Assurance           II-D     476
     Plan for the Determination of  PCDD/PCDF in
     Fish, by Environmental  Research Laboratory,
     USEPA, April 1988.

E.   Determination of Tetra-,  Hexa-, Hepta-, and           II-E     520
     Octachlorodibenzo-p-dioxin Isomers in
     Particulate Samples  at  Parts per Trillion
     Levels, by L.L. Lamparski and  T.J.
     Nestrick,  Analytical Chemistry. Volume 52:
     2045-2054, 1980.

F.   Novel Extraction Device for the                       II-F     530
     Determination of Chlorinated Dibenzo-p-
     dioxins (PCDDs) and  Dibenzofurans (PCDFs)
     in Matrices Containing  Water,by L.L.
     Lamparski  and T.J. Nestrick, Chemosphere,
     Volume 19: 27-31, 1989.

G.   Control of Interferences  in the Analysis of           II-G     535
     Human Adipose Tissue for  2,3,7,8-Tetra-
     chlorodibenzo-p-dioxin, by D.G. Patterson,
     et.  al.,  Environmental Toxicological
     Chemistry. Volume 5: 355-360,  1986.

H.   Protocol for the Analysis of 2,3,7,8-                 II-H     541
     Tetrachlorodibenzo-p-dioxin by High-
     Resolution Gas Chromatography/High-
     Resolution Mass Spectrometry,  by John S.
     Stanley and Thomas M. Sack, Environmental
     Monitoring Systems Laboratory-Las Vegas,
     USEPA, EPA 600/4-86-004,  January 1986.

I.   Method 613 -- 2,3,7,8-Tetrachlorodibenzo-p-           II-I     697
     dioxin, Section 4.1, 49 Federal Register
     43234, October 26, 1984.

J.   Interpretation of Percent Recovery Data, by           II-J     703
     L.  P. Provost and R.S.  Elder, American
     Laboratory.  Volume 15:  56-83,  1983.
                                    ii                            11/90

-------
Method 1613    Revision  A      October 1990
Tetra- through Octa-  Chlorinated Dioxins and Furans
by Isotope Dilution  HRGC/HRMS
       1   SCOPE AND APPLICATION

      1.1   This method is designed  to  meet the survey
           requirements of the USEPA ITD.  The method
           is  used to determine the  tetra- through
           octa-  chlorinated  dibenzo-p-dioxins  and
           dibenzofurans  associated  with the  Clean
           Water Act  (as  amended 1987);  the Resource
           Conservation and Recovery Act (as amended
           1986); and the Comprehensive Environmental
           Response,  Compensation  and Liability Act
           (as  amended  1986); and other dioxin and
           furan compounds  amenable  to high resolu-
           tion  capillary column gas  chromatography
           (HRGC)/high  resolution  mass  spectrometry
           (HRMS).    Specificity  is provided for de-
           termination of the 17 2,3,7,8-substituted
           polychlorinated  dibenzo-p-dioxins  (PCDD)
           and polychlorinated dibenzofurans (PCOF).

      1.2   The method is based on EPA, industry, com-
           mercial  laboratory,  and academic methods
           (References 1-6).

      1.3   The  compounds  listed in  Table   1  may  be
           determined in  waters, soils,  sludges, and
           other matrices by this method.

      1.4   The  detection  limits of   the method are
           usually  dependent  on  the   level  of inter-
           ferences  rather  than  instrumental  limita-
           tions.   The  levels in Table 2 typify the
           minimum  quantities that can be determined
           in environmental  samples using the method.

      1.5   The  GCHS portions of the  method are for
           use  only  by  analysts  experienced  with
           HRGC/HRMS  or  under the close supervision
           of  such  qualified  persons.   Each  labora-
           tory  that uses  this method  must  demon-
           strate  the ability to generate acceptable
           results  using the procedure  in Section
           8.2.

       2   SUMMARY OF METHOD

      2.1   Stable  isotopicaUy labeled analogs of  15
           of  the  PCDDs  and PCOFs  are added to each
           sample prior to  extraction.   Samples con-
           taining  coarse  solids  are  prepared  for
           extraction by  grinding  or  homogenization.
           Water   samples   are  filtered  and  then
           extracted  with  methylene  chloride  using
           separatory funnel procedures;  the particu-
           lates from the water samples, soils, and
           other finely divided  solids  are  extracted
        using a combined Soxhlet extraction/Dean-
        Stark azeotropic  distillation  (Reference
        7).    Prior  to cleanup and  analysis,  the
        extracts  of  the filtered  water  and  the
        particulates  are combined.

  2.2    After extraction,   37C14-labeled  2,3,7,8-
        TCDD is added to  each  extract  to measure
        the  efficiency  of  the  cleanup  process.
        Samples  cleanup may  include  back extrac-
        tion with  acid and/or base, and gel perme-
        ation,  alumina,  silica  gel,  and activated
        carbon chromatography.    High performance
        liquid chromatography (HPLC) can  be  used
        for  further   isolation  of  the  2,3,7,8-
        isomers   or  other  specific  isomers   or
        congeners.

  2.3    After cleanup, the  extract is concentrated
        to  near  dryness.    Immediately  prior  to
        injection,   two  internal  standards   are
        added to each extract,  and 'a 1  uL aliquot
        of  the  extract  is injected  into  the  gas
        chromatograph.  The analytes are separated
        by  the GC  and detected  by a high resolu-
        tion  (>10,000)  mass  spectrometer.    Two
        exact masses (m/z's)  are  monitored  for
        each analyte.   The  isotopicaUy labeled
        compounds   serve   to  correct    for   the
        variability of the  analytical technique.

  2.4    Dioxins  and  furans  are  identified   by
        comparing  GC  retention times and the  ion
        abundance   ratios  of  the  m/z's  with  the
        corresponding  retention   time  ranges  of
        authentic   standards  and  the theoretical
        ion  abundance ratios of  the exact m/z's;
        Isomers and  congeners are identified  when
        the  retention  times   and  m/z  abundance
        ratios  agree within  pre-defined  limits..
        By using a GC column  or columns capable of
        resolving  the 2,3,7,8-substituted isomers
        from  all   other   tetra-   isomers,   the
        2,3,7,8-substituted isomers are identified
        when the retention time and m/z abundance
        ratios agree  within pre-defined limits o*f
        the  retention times  and  exact  m/z ratios
        of authentic  standards.

  2.5    Quantitative  analysis  is  performed by GCMS
        using selected ion current profile (SICP)
        areas,  in  one of two ways.

2.5.1    For  the 15 2,3,7,8-substituted  isomers for
        which labeled analogs  are available  (see
                                             001

-------
         Table  1),  the GCMS  system  is  calibrated
         and  the  compound  concentration  is  deter-
         mined  using an isotope dilution technique.
         Although  a  labeled  analog  of  the  octa-
         chlorinated dibenzofuran (OCDF)  is  avail-
         able,  using high  resolution mass spectrom-
         etry it produces  an m/z that may interfere
         with  the  identification and quantitat ion
         of  the unlabeled  octachlorinated  dibenzo-
         p-dioxin  (OCDD).    Therefore,  this  labeled
         analog has  not been  included in the cali-
         bration standards, and  the un labeled OCDF
         is  quantitared against the  labeled OCDO.
         Because the labeled analog of 1,2,3,7,8,9-
         HxCOD   is   used  as  an  internal  standard
         (i.e.,  not  added before extraction  of the
         sample),  it cannot  be used  to  quantitate
         the  un labeled compound by  strict  isotope
         dilution  procedures.   Therefore, the unla-
         beled   1,2,3,7,8,9-HxCDD   is  quantitated
         using  the average of  the  responses  of the
         labeled analogs  of the  other two 2,3,7,8-
         substituted HxCDO's,  1,2,3,4,7,8-HxCDO and
         1,2,3,6,7,8-HxCDD.   As a  result,  the con-
         centration  of  the unlabeled 1,2,3,7,8,9-
         HxCDD   is   corrected   for   the   average
         recovery  of the other two  HxCDD's.

2.5.2    For  non-2,3,7,8-substituted  isomers  and
         the  total  concentrations  of all  isomers
         within a  level   of  chlorination  (i.e.,
         total  TCDD),  concentrations are determined
         using  response factors from  the  calibra-
         tion of  labeled analogs at the  same level
         of chlorination.

  2.6    The  quality  of  the  analysis  is  assured
         through reproducible calibration and test-
         ing  of the  extraction, cleanup, and GCMS
         systems.

    3    CONTAMINATION  AND  INTERFERENCES

  3.1    Solvents,   reagents,  glassware,  and other
         sample processing  hardware may yield arti-
         facts  and/or  elevated  baselines  causing
         misinterpretation   of  chromatograms  (Ref-
         erences   8-9).     Specific  selection  of
         reagents  and  purification of solvents by
         distillation  in all-glass  systems  may be
         required.    Where  possible,   reagents  are
         cleaned by  extraction or solvent rinse.

  3.2    Proper cleaning  of glassware is extremely
         important because  glassware may not only
         contaminate  the   samples,   but  may  also
         remove   the   analytes  of   interest   by
         adsorption  on  the  glass surface.

3.2.1    Glassware  should  be  rinsed  with  solvent
         and  washed  with  a  detergent solution as
        soon  after use as  is  practical.   Sonica-
        tion  of glassware  containing  a detergent
        solution for  approximately 30 seconds may
        aid in  cleaning.  Glassware with removable
        parts,   particularly  separatory  funnels
        with     teflon    stopcocks,    must    be
        disassembled prior  to detergent washing.

3.2.2   After detergent  washing,  glassware should
        be  immediately rinsed first with methanol,
        then  with  hot tap  water.   The tap water
        rinse   is   followed by   another  methanol
        rinse,  then  acetone,  and  then methylene
        chloride.

3.2.3   Do  not  bake reusable glassware in an oven
        as a  routine part of cleaning.  Baking may
        be  warranted  after  particularly  dirty
        samples are  encountered,  but  should  be
        minimized,  as  repeated baking  of glassware
        may   cause  active   sites  on  the  glass
        surface  that  will  irreversibly  adsorb
        PCDDs/PCDFs.

3.2.4   Immediately prior  to use, Soxhlet extrac-
        tion  glassware  should   be  pre-extracted
        with  toluene  for   approximately  3  hours.
        See Section 11.1.2.3.   Separatory funnels
        should    be    shaken    with  - methylene
        chloride/toluene .(80/20  mixture)  for   2
        minutes,  drained,   and  then  shaken  with
        pure methylene chloride for 2 minutes.

  3.3   All materials used  in  the analysis shall
        be  demonstrated  to  be free from  interfer-
        ences  by  running  reference  matrix blanks
        initially   and  with   each   sample   set
        (samples  started  through  the  extraction
        process on  a given  12-hour   shift,  to  a
        maximum of  20  samples).    The  reference
        matrix  blank  must  simulate,  as closely as
        possible,  the sample  matrix  under   test.
        Reagent water (Section 6.6.1)  is  used to
        simulate  water  samples;   playground   sand
        (Section   6.6.2)   or  white   quartz  sand
        (Section  6.3.2)  can  be  used  to simulate
        soils;  filter  paper  (Section  6.6.3)  is
        used   to   simulate  papers   and  similar
        materials;  other materials (Section 6.6.4)
        can be  used to simulate other matrices.

  3.4   Interferences   coextracted  from  samples
        will  vary  considerably   from  source  to
        source,  depending  on the diversity of  the
        site  being sampled.   Interfering  compounds
        may  be present  at  concentrations several
        orders  of  magnitude higher than the  PCDDs
        and  PCDFs.   The most  frequently encoun-
        tered    interferences   are    chlorinated-
        biphenyls,   methoxy  biphenyls,  hydroxy-

-------
        diphenyl   ethers,   benzylphenyl   ethers,
        polynuclear   aromatics,   and  pesticides.
        Because very  low  levels of PCDDs and PCOFs
        are measured  by  this method, the elimina-
        tion of  interferences \s  essential.   The
        cleanup steps given  in Section  12  can be
        used to  reduce or  eliminate these  inter-
        ferences   and  thereby   permit   reliable
        determination  of  the PCDOs  and  PCOFs at
        the levels shown  in Table 2.

  3.5   Each piece of  reusable glassware should be
        numbered   in   such   a  fashion  that  the
        laboratory  can   associate  all   reusable
        glassware with the  processing  of a parti-
        cular  sample.     This  will  assist  the
        laboratory in:   1)  tracking  down possible
        sources  of  contamination for   individual
        samples,  21   identifying  glassware assoc-
        iated  with   highly   contaminated  samples
        that may  require  extra  cleaning,  and 3)
        determining   when  glassware   should  be
        discarded.

    4   SAFETY

  4.1   The  toxicity  or  careinogenicity  of  each
        compound  or   reagent  used in  this method
        has   not   been   precisely  determined;
        however,  each chemical  compound should be
        treated  as   a  potential  health  hazard.
        Exposure  to   these   compounds   should be
        reduced to the lowest possible level.

4.1.1   The 2,3,7,8-TCDO  isomer  has  been found to
   .  .   be  acnegenic,  carcinogenic,  and  terato-
        genic in  laboratory animal studies.   It is
        soluble in water  to  approximately 200 ppt
        and  in  organic solvents  to  0.14 percent.
        On    the    basis    of    the    available
        toxicological  and physical  properties of
        2,3,7,8-TCDD,  all  of the  PCDDs  and PCDFs
        should be handled only  by highly trained
        personnel    thoroughly    familiar     with
        handling  and  cautionary  procedures,  and
        who understand the associated risks.

4.1.2   It is recommended that the laboratory  pur-
        chase  dilute  standard  solutions  of  the
        analytes  in  this  method.    However,  if
        primary solutions are prepared,  they shall
        be prepared   in a hood,   and a  NIOSH/MESA
        approved  toxic  gas  respirator  shall be
        worn when high concentrations are handled.

  4.2   The  laboratory is  responsible  for  main-
        taining a current awareness  file  of  OSHA
        regulations regarding the  safe handling of
        the chemicals specified in this method.   A
        reference  file  of  data  handling   sheets
        should  also  be  made  available   to  all
        personnel   involved   in  these  analyses.
        Additional   information   on    laboratory
        safety can  be  found  in References  10-13.
        The references and bibliography  at  the end
        of  Reference  13  are  particularly  compre-
        hensive   in  dealing   with   the   general
        subject of  laboratory safety.

  4.3   The PCDOs and  PCDFs  and samples suspected
        to  contain  these compounds  are   handled
        using  essentially   the  same   techniques
        employed   in   handling   radioactive   or
        infectious  materials.    Well-ventilated,
        controlled    access     laboratories   are
        required.   Assistance  in  evaluating  the
        health  hazards  of  particular  laboratory
        conditions  may  be  obtained  from  certain
        consulting  laboratories  and  from State
        Departments  of   Health  or Labor,  many of
        which  have  an  industrial  health  service.
        The PCDDs and PCOFs are extremely  toxic to
        laboratory  animals.   Each laboratory must
        develop   a   strict   safety   program  for
        handling the PCDDs and PCDFs.  The follow-
        ing practices  are recommended (References
        2 and  14).

4.3.1   Facility  -- When finely  divided  samples
        (dusts, soils, dry chemicals)  are  handled,
        all   operations  (including   removal   of
        samples  from  sample  containers,  weighing,
        transferring,   and   mixing),   should  be
        performed  in  a  glove  box demonstrated to
        be  leak  tight  or in  a  fume  hood  demon-
        strated  to  have adequate air  flow.  Gross
        losses   to  the  laboratory  ventilation
        system must not  be allowed.   Handling of
        the  dilute  solutions  normally  used  in
        analytical  and  animal  work  presents  no
        inhalation  hazards  except in  the  case of
        an accident.

4.3.2   Protective  equipment  -- Throwaway  plastic
        gloves,  apron  or lab coat,  safety glasses
        or  mask, and  a  glove  box  or  fume hood
        adequate  for  radioactive work  should be
        utilized.    During   analytical  operations
        which  may  give  rise  to aerosols or dusts,
        personnel should wear  respirators  equipped
        with   activated  carbon   filters.    Eye
        protection  equipment  (preferably full face
        shields) must  be worn  while working with
        exposed   samples  or    pure    analytical
        standards.  Latex gloves  are commonly used
        to  reduce  exposure  of  the  hands.   When
        handling  samples  suspected  or   known  to
        contain  high  concentrations  of  the PCDDs
        or  PCDFs,  an  additional set of  gloves can
        also be worn beneath the  latex gloves.
                                                        003

-------
   4.3.3    Training --  Workers must be trained in the
            proper  method  of  removing  contaminated
            gloves and clothing without contacting the
            exterior surfaces.

   4.3.4    Personal hygiene  -•  Thorough  washing  of
            hands  and forearms after each manipulation
            and  before   breaks  (coffee,   lunch,   and
            shift).

   4.3.5    Confinement  -- Isolated  work area, posted
            with   signs,   segregated   glassware   and
            tools,  plastic  absorbent  paper  on  bench
            tops.

   4.3.6    Effluent vapors --  The effluents  of sample
            splitters for  the  gas  chromatograph  and
            roughing pumps  on  the GC/MS should  pass
            through  either a  column of activated char-
            coal  or  be bubbled through a trap contain-
            ing oil  or high-boiling alcohols.

   4.3.7    Waste  Handling and Disposal

 4.3.7.1    Handling  •-   Good   technique    includes
            minimizing  contaminated   waste.    Plastic
            bag liners  should  be used  in waste cans.
            Janitors  and  other   personnel   must  be
            trained  in the safe handling of waste.

 4.3.7.2    Disposal

4.3.7.2.1    The PCDDs  and PCDFs  decompose  above 800
            °C.   Low-1 eve I  waste  such as  absorbent
            paper,   tissues,    animal   remains,   and
            plastic  gloves may  be  burned  in  an appro-
            priate  incinerator.     Gross   quantities
            (milligrams)  should be  packaged  securely
            and disposed through commercial  or govern-
            mental  channels  which   are  capable  of
            handling extremely toxic wastes.

4.3.7.2.2    Liquid  or  soluble  waste  should   be  dis-
            solved   in    methanol    or  ethanol   and
            irradiated with  ultraviolet light  with  a
            wavelength greater than 290 nm for several
            days.    (Use F  40  8L lamps or equivalent.)
            Analyze  liquid wastes  and  dispose  of the
            solutions when the  PCDDs  and  PCDFs can no
            longer be detected.

   4.3.8    Decontamination

 4.3.8.1    Personal Decontamination  -- Use  any  mild
            soap  with plenty of scrubbing action.

 4.3.8.2    Glassware,    tools,    and   surfaces
            Chlorothene  NU  Solvent  (Trademark  of the
            Dow Chemical  Company)  is  the  least toxic
            solvent   shown to  be  effective.    Satis-
            factory  cleaning  may  be  accomplished  by
  4.3.9
 4.3.10
 4.3.11
      5

    5.1
5.1.1.2
5.1.1.3
5.1.1.4
 rinsing  with   Chlorothene,   then  washing
 with  any detergent and  water.   If glass-
 ware  is first  rinsed with  solvent,   then
 the  dish water may be disposed of in  the
 sewer.   Given  the cost of disposal,   it  is
 prudent  to minimize solvent wastes.

 Laundry  --  Clothing  known  to be  contami-
 nated should be collected in  plastic  bags.
 Persons  who convey  the  bags  and launder
 the   clothing  should  be advised of  the
 hazard   and  trained  in   proper  handling.
 The   clothing  may be put  into  a washer
 without  contact if the  launderer  knows  of
 the  potential  problem.   The washer should
 be  run   through a  cycle  before being  used
 again for other clothing.

 Wipe  tests  -- A useful method  of  determin-
 ing  cleanliness of work  surfaces  and  tools
 is  to  wipe  the  surface with  a  piece  of
 filter  paper.   Extraction and  analysis  by
"GC can  achieve  a  limit of detection of  0.1
 ug  per   wipe.   Less   than 0.1  ug  per  wipe
 indicates acceptable  cleanliness;  anything
 higher  warrants  further cleaning.    More
 than  10  ug  on  a wipe constitutes  an  acute
 hazard  and  requires prompt cleaning before
 further  use  of   the equipment   or   work
 space,   and  indicates   that   unacceptable
 work  practices  have been employed.

 Accidents  --  Remove  contaminated  clothing
 immediately,   taking   precautions  not   to
 contaminate  skin  or  other articles.    Wash
 exposed  skin   vigorously  and repeatedly
 until medical attention  is obtained.
                                 for   discrete   or
 APPARATUS AND MATERIALS

 Sampli ng    equ i pment
 composite sampling.

 Sample  bottles and caps
 Liquid  samples (waters,  sludges and  simi-
 lar   materials  containing   five   percent
 solids  or  less)  ••  Sample  bottle,  amber
 glass,  1.1  L  minimum, with screw cap.

 Solid samples {soils, sediments,  sludges,
 paper  pulps,  filter  cake,   compost,   and
 similar materials  that  contain more  than
 five percent  solids)  --  Sample   bottle,
 wide mouth, amber glass,  500  mL minimum.

 If   amber  bottles   are  not   available,
 samples shall  be protected from  light.

 Bottle   caps   --  Threaded  to  fit  sample
 bottles.   Caps shall  be  lined with  Teflon.
                                                           004

-------
 5.1.1.5   Cleaning

5.1.1.5.1   Bottles  are detergent  water  washed,  then
           solvent rinsed before use.

5.1.1.5.2   Liners  are  detergent  water  washed,  then
           rinsed  with  reagent water (Section 6.6.1)
           and    then   solvent,    and    baked-   at
           approximately 200  °C for one hour minimum
           prior to use.

   5.1.2   Compositing   equipment   --  Automatic  or
           manual  compositing  system   incorporating
           glass   containers   cleaned   per   bottle
           cleaning procedure above.  Glass or Teflon
           tubing  only  shall be used.  If  the sampler
           uses a  peristaltic pump, a minimum length
           of compressible  silicone rubber tubing may
           be used  in the pump  only.  Before use, the
           tubing  shall  be  thoroughly   rinsed  with
           methanol,  followed  by  repeated  rinsings
           with  reagent  water to  minimize  sample
           contamination.   An  integrating flow meter
           is used to collect  proportional composite
           samples.

     5.2   Equipment  for glassware  cleaning

   5.2.1   Laboratory sink  with.overhead fume hood

     5.3   Equipment  for sample preparation

   5.3.1   Laboratory fume  hood of  sufficient size to
           contain the  sample preparation equipment
           listed  below

   5.3.2   Glove box  (optional)

   5.3.3   Tissue  homogenizer   --   VirTis Model  45
           Macro   homogenizer  (American  Scientific
           Products   H-3515,   or   equivalent)   with
           stainless   steel  Macro-shaft   and  Turbo-
           shear blade.

   5.3.4   Meat  grinder  --  Hobart,  or  equivalent,
           with 3-5 mm holes  in inner plate.

   5.3.5   Equipment  for determining  percent moisture

 5.3.5.1   Oven, capable of maintaining  a  temperature
           of 110  ±5  °C.

 5.3.5.2   Oessicator

   5.3.6   Balances

 5.3.6.1   Analytical -- Capable of weighing 0.1 mg.

 5.3.6.2   Top  loading -- Capable  of  weighing 10 mg.
    5.4   Extraction apparatus

  5.4.1   Water samples

5.4.1.1   pH    meter,    with    combination
          electrode.
              glass
5.4.1.2   pH  paper,  wide range  (Hydrion  Papers,  or
          equivalent).

5.4.1.3   Graduated cylinder, 1 L capacity

5.4.1.4   1  L filtration flasks with  side arm, for
          use in vacuum filtration of water samples.

5.4.1.5   Separatory  funnels -- 250,  500;  and 2000
          ml, with Teflon stop cocks.
  5.4.2   Soxhlet/Dean-Stark
          (Figure 1)
(SOS)
                                           extractor
          FIGURE 1   Soxhlet/Dean-Stark Extractor
                                                           005

-------
 5.4.2.1    Soxhlet  --  50  mm  i.d.,  200 mL  capacity
            with 500  mL  flask  (Cal-Glass  LG-6900,  or
            equivalent,  except substitute 500 mL round
            bottom  flask   for   300   mL  flat  bottom
            flask).

5.4.2.2     Thimble -- 43  x 123  to  fit Soxhlet (Cal-
            Glass LG-6901-122, or equivalent).

 5.4.2.3    Moisture trap -- Dean Stark or Barret with
            Teflon stopcock, to fit Soxhlet.

 5.4.2.4    Heating mantle --  Hemispherical,  to  fit
            500 ml  round bottom  flask  (Cal-Glass  LG-
            8801-112,  or  equivalent).

 5.4.2.5    Variable  transformer  --   Powerstat  (or
            equivalent),  110 volt. 10 amp.

   5.4.3    Beakers, 400-500 mL

   5.4.4    Spatulas -- Stainless steel

     5.5    Filtration apparatus

   5.5.1    Pyrex glass  wool  --  Solvent  extracted  by
            SOS  for  three  hours  minimum.     (NOTE:
            Baking glass wool  may cause  active sites
            that     will      irreversibly      adsorb
            PCDOs/PCOFs.)

   5.5.2    Glass funnel  -- 125-250 mL

   5.5.3    Glass fiber  filter paper (Whatman GF/D,  or
            equivalent)

   5.5.4    Drying column  --  15 .to  20 mm  i.d. Pyrex
            chromatographic   column   equipped   with
            coarse glass  frit or glass wool  plug.

   5.5.5    Buchner funnel, 15 cm.

   5.5.6    Glass fiber  filter paper  for above.

   5.5.7    Pressure filtration apparatus --  Millipore
            YT30 142 HW,  or equivalent.

     5.6    Centrifuge apparatus

   5.6.1    Centrifuge --  Capable of rotating  500  mL
            centrifuge bottles  or  15  mL  centrifuge
            tubes at 5,000  rpm minimum

   5.6.2    Centrifuge bottles  -- 500 mL, with screw
            caps,  to f\t  centrifuge

   5.6.3    Centrifuge tubes  --  12-15 mL, with screw
            caps,  to fit  centrifuge
    5.7   Cleanup apparatus

  5.7.1   Automated   gel   permeation  chromatograph
          (Analytical    Biochemical    Labs,    Inc,
          Columbia, HO, Model  GPC Autoprep 1002,  or
          equivalent).

5.7.1.1   Column -- 600-700  mm x 25  mm i.d., packed
          with  70  g  of  SX-3   Bio-beads  (Bio-Rad
          Laboratories,     Richmond,     CA,     or
          equivalent).

5.7.1.2   Syringe, 10 mL, with Luer fitting.

5.7.1.3   Syringe  filter  holder, stainless  steel,
          and glass  fiber  or Teflon  filters (Gelman
          4310, or equivalent).

5.7.1.4   UV  detectors  --  254-nm,  preparative  or
          semi-prep flow cell:  (Isco,  Inc., Type 6;
          Schmadzu, 5  mm path length;  Beckman-Altex
          152U,  8  uL micro-prep flow cell,  2  mm
          path; Pharmacia  UV-1,  3 mm flow cell;  LDC
          Milton-Roy   UV-3,   monitor   #1203;   or
          equivalent).

  5.7.2   Reverse  phase   high  performance  liquid
          chromatograph

5.7.2.1   Column oven and detector  -- Perkin-Elmer
          Model  LC-65T (or  equivalent) operated  at
          0.02 AUFS at 235 nm.

5.7.2.2   Injector --  Rheodyne 7120  (or  equivalent)
          with 50 uL sample loop.

5.7.2.3   Column  --  Two  6.2  x  250 "mm  Zorbax-OOS
          columns   in   series  (DuPont  Instruments
          Division, Wilmington,  DE,  or equivalent),
          operated at 50 "C with  2.0 mL/min methanol
          isocratic effluent.

5.7.2.4   Pump -- Altex 110A (or equivalent).

  5.7.3   Pipets

5.7.3.1   Disposable,  Pasteur,  150 mm x  5  mm i.d.
          (Fisher     Scientific     13-678-6A,     or
          equivalent).

5.7.3.2   Disposable,  serological,  10  mL  (6   mm
          i.d.).

  5.7.4   Chromatographic columns

5.7.4.1   150 mm ,x 8 mm i.d.,  (Kontes K-420155,  or
          equivalent)  with  coarse  glass  frit   or
          glass wool plug and 250 mL  reservoir.
                                                          OOG

-------
   5.7.5
 5.7.4.2    200  mm x  15 mm  i.d., with  coarse  glass
            frit   or   glass  wool  plug  and   250   ml
            reserve i r.

            Oven  -- For  storage  of adsorbents,  capable
            of maintaining  a  temperature of  130 ±5  °C.

     5.8    Concentration apparatus

   5.8.1    Rotary  evaporator    --    Buchi/Brinkman-
            American   Scientific   No.   E5045-10   or
            equivalent,   equipped  with   a   variable
            temperature  water bath.

 5.8.1.1    A vacuum source is required for  use of  the
            rotary evaporator.    It must  be  equipped
            with  a shutoff valve  at   the  evaporator,
            and preferably, have  a vacuum gauge.

 5.8.1.2    A recirculating water pump and  chiller  are
            recommended,  as  use  of   tap   water  for
            cooling   the   evaporator  wastes   large
            volumes   of   water    and   can   lead   to
            inconsistent    performance    as     water
            temperatures and pressures  vary.

 5.8.1.3    Round bottom flasks   -- 100 raL  and 500  mL
            or   larger,   with  ground  glass   fitting
            compatible with the  rotary evaporator.

   5.8.2    Kuderna-Danish  (K-D)

 5.8.2.1    Concentrator tube--10mL, graduated (Kontes
            K-570050-1025,    or     equivalent)   with
            calibration   verified.     Ground   glass
            stopper  (size  19/22  joint)   is  used   to
            prevent evaporation of extracts.

 5.8.2.2    Evaporation   flask--500   mL   (Kontes   K-
            570001-0500,  or  equivalent),  attached  to
            concentrator tube with springs  (Kontes  K-
            662750-0012).

 5.8.2.3    Snyder column--three  ball  macro  (Kontes K-
            503000-0232, or equivalent).

 5.8.2.4    Boiling chips

5.8.2.4.1    Glass or  silicon   carbide--approx  10/40
            mesh,  extracted  with  rnethylene  chloride
            and baked at 450 °C for one h minimum.
5.8.2.4.2  Teflon      (optional)--extracted
           methylene chloride.
                                                  with
 5.8.2.5   Water  bath—heated,  with  concentric  ring
           cover,    capable    of    maintaining    a
           temperature within +/-  2 °C,  installed  in
           a  fume hood.
5.8.3   Nitrogen  blowdown  apparatus  --   Equipped
        with water bath controlled at 35-40  °C  (N-
        Evap, Organomation Associates,  Inc.,  South
        Berlin, MA, or equivalent),  installed in a
        fume hood.

5.8.4   Sample  vials  --  Amber glass,  2-5 mL  with
        Teflon-lined screw cap.

  5.9   Gas chromatograph  --  Shall have  split less
        or on-column  injection port for  capillary
        column,     temperature    program     with
        isothermal hold, and shall meet all  of  the
        performance specifications in Section 7.

5.9.1   GC  Column  for PCDDs  and  PCDFs  and  for
        isomer  specificity for 2,3,7,8-TCDO  --  60
        ±5  m x 0.32  ±0.02  mm  i.d.;  0.25  urn 5%
        phenyl,  94%  methyl,  1%  vinyl   silicone
        bonded  phase fused silica  capillary  column
        (J & U  DB-5, or equivalent).

5.9.2   GC  Column   for   isomer  specificity   for
        2,3,7,8-TCDF  --  30 ±5  m x  0.32  ±0.02 mm
        i.d.;  0.25  urn  bonded phase  fused  silica
        capillary   column   (J  &   U  08-225,  or
        equivalent).

 5.10   Mass  spectrometer  --  28-40  eV   electron
        impact  ionization,  shall   be  capable of
        repetitively   selectively   monitoring  12
        exact   m/z's  minimum  at  high  resolution
        (>10,000) during a period  of approximately
        1  second,  and  shall  meet  all  of   the
        performance specifications in Section 7.

 5.11   GCHS  interface  --  The  mass spectrometer
        (MS)  shall  be  interfaced  to  the GC  such
        that  the  end  of  the  capillary  column
        terminates  within  1  cm  of  the ion  source
        but does not intercept the electron  or  ion
        beams.

 5.12   Data  system   --  Capable  of   collecting,
        recording and storing MS data.

    6   REAGENTS AND STANDARDS

  6.1   pH adjustment and back extraction

6.1.1   Potassium  hydroxide   --   Dissolve  20  g
        reagent grade KOH in 100 mL reagent water.

6.1.2   Sulfuric  acid  --  Reagent  grade  (specific
        gravity 1.84).

6.1.3   Sodium  chloride  --  Reagent grade, prepare
        a  five  percent (w/v)  solution  in reagent
        water.
                                                        007

-------
    6.2    Solution drying and evaporation

  6.2.1    Solution drying --  Sodium sulfate,  reagent
           grade,  granular anhydrous  (Baker 3375,  or
           equivalent),     rinsed    with    methylene
           chloride  (20 mL/g).  baked  at  400 °C  for
           one  hour  minimum, cooled  in a  dessicator,
           and  stored in  a pre-cleaned glass  bottle
           with screw  cap  that prevents moisture, from
           entering.    If   after heating  the  sodium
           sulfate develops a  noticeable grayish cast
           (due to  the presence  of   carbon  in  the
           crystal matrix), that batch of  reagent  is
           not   suitable   for   use  and  should   be
           discarded.     Extraction   with   methylene
           chloride  (as opposed to  simple  rinsing)
           and   baking  at  a  lower  temperature  may
           produce sodium  sulfate that  is  suitable
           for  use.

  6.2.2    Prepurified nitrogen

    6.3    Extraction

  6.3.1    Solvents -- Acetone, toluene, cyclohexane,
           hexane,   nonane,    methanol,    methylene
           chloride,  and nonane:  distilled-in-glass,
           pesticide  quality,   lot  certified  to  be
           free of interferences.

  6.3.2    White  quartz   sand,   60/70  mesh  --  For
           Soxhlet/Dean-Stark   extraction,   (Aldrich
           Chemical  Co,   Milwaukee    UI     Cat  No.
           27,437-9,  or  equivalent).    Bake  at 450  "C
           for  four hours  minimum.

    6.4    GPC   calibration   solution  --   Solution
           containing  300   mg/mL corn  oil,  15  mg/mL
           bis(2-ethylhexyl)   phthalate,   1.4   mg/mL
           pentachlorophenol,  0.1 rag/mL perylene,  and
           0.5  mg/mL sulfur

    6.5    Adsorbents  for  sample cleanup

  6.5.1    Silica  gel

6.5.1.1    Activated silica gel  --  Bio-Si I  A,  100-200
           mesh (Bio-Rad   131-1340,  or equivalent),
           rinsed  with methylene  chloride,  baked  at
           180  °C  for one  hour minimum,  cooled  in  a
           dessicator,  and  stored in a  pre-cleaned
           glass bottle  with screw cap that  prevents
           moisture from entering.
                   i
6.5.1.2    Acid silica  gel  (30  percent  w/w)
           Thoroughly  mix   44.0 g  of concentrated
           sulfuric  acid  with   100.0  g of  activated
           silica  gel  in a  clean container.   Break  up
           aggregates  with  a   stirring rod  until  a
               uniform  mixture is obtained.   Store in a
               screw-capped bottle with Teflon-lined cap.

    6.5.1.3    Basic silica gel -- Thoroughly nix 30 g of
               IN   sodium  hydroxide   with  100   g  of
               activated  silica gel  in a  clean  container.
               Break  up  aggregates  with  a  stirring rod
               until  a  uniform  mixture  is   obtained.
               Store   in  a   screw-capped  bottle  with
               Teflon-lined cap.

      6.5.2    Alumina  --  Either one  of  two  types  of
               alumina, acid  or basic, may be used  in the
               cleanup  of sample extracts, provided that
               the  laboratory  can  meet  the performance
               specifications  for the recovery  of  labeled
               compounds  described  in Section  8.3.   The
               same type  of  alumina  must be used for all
               samples,    including    those    used   to
               demonstrate initial precision and accuracy
               (Section 8.2)  and ongoing  precision and
               accuracy (Section  14.5).

    6.5.2.1    Acid alumina  -- Bio-Rad Laboratories 132-
               1340  Acid Alumina  AG 4  (or equivalent).
               Activate by heating to 130  °C for 12 hours
               minimum.

    6.5.2.2    Basic alumina  -- Bio-Rad Laboratories 132-
               1240 Basic Alumina AG 10  (or equivalent).
               Activate by heating to 600  °C for 24 hours
               minimum.     Alternatively,   activate  by
               heating  alumina in a  tube  furnace at 650-
               700  °C  under  an air flow  of approximately
               400  cc/min.   Do not  heat  over  700 °C, as
               this  can  lead  to  reduced capacity  for
               retaining  the  analytes.    Store  at  130 °C
               in a covered  flask.   Use within  five days
               of baking.

      6.5.3    AX-21/Celite

    6.5.3.1    Activated   carbon   --    AX-21    (Anderson
               Development   Company,   Adrian,   MI,   or
               equivalent).    Prewash  with  methanol  and
               dry  in vacuo at  110 "C.
    6.5.3.2
    6.5.3.3
Celite   545
equivalent).
                                  (Supelco   2-0199,   or
Thoroughly  mix  5.35  g  AX-21  and 62.0  g
Celite 545  to produce a  7.9% w/w mixture.
Activate  the  mixture at  130  "C  for  six
hours minimum.   Store in a dessicator.
        6.6    Reference matrices

      6.6.1    Reagent  water -- Water in which  the  PCDDs
               and  PCDFs  and  interfering  compounds  are
               not detected  by  this method.
8
003

-------
6.6.2   High solids reference matrix -- Playground
        sand  or  similar  material   in which  the
        PCDDs and  PCDFs and interfering compounds
        are not  detected by this method.   May be
        prepared   by   extraction  with  methylene
        chloride and/or  baking  at '450  °C for four
        hours minimum.

6.6.3   Filter paper  --  Gelman  type A (or equiva-
        lent) glass  fiber  filter  paper  in which
        the PCOOs  and PCDFs  and interfering con-
        pounds  are not  detected by  this  method.
        Cut the paper  to simulate the  surface area
        of the paper  sample being tested.

6.6.4   Other  matrices  --  This  method  may  be
        verified on any matrix  by  performing the
        tests given in Section  8.2.   Ideally, the
        matrix  should  be  free  of  the  PCODs  and
        PCDFs, but  in no case shall the background
        level  of   the PCDDs  and  PCDFs   in  the
        reference  matrix  exceed three  times  the
        minimum  levels given in Table 2.   If low
        background  levels  of the PCDDs and PCDFs
        are present  in  the  reference  matrix,  the
        spike   level   of   the  analytes  used  in
        Section 8.2 should be increased to provide
        a  spike-to-background ratio  in the range
        of 1/1 to 5/1  (Reference 15).

  6.7   Standard  solutions --  Purchased  as solu-
        tions  or  mixtures  with  certification to
        their  purity,  concentration,  and  authen-
        ticity,  or  prepared   from  materials  of
        known purity  and composition.   If compound
        purity   is  98 percent  or  greater,  the
        weight  may be  used  without correction to
        compute  the concentration of the standard.
        When not  being used,  standards are stored
        in the  dark at room temperature in screw-
        capped  vials   with  Teflon-lined caps.    A
        mark is  placed on  the vial  at  the  level of
        the  solution   so  that  solvent  evaporation
        loss can be detected.   If solvent  loss has
        occurred,  the solution should  be replaced.

  6.8   Stock solutions

6.8.1   Preparation -- Prepare  in  nonane  per the
        steps  below  or  purchase as  dilute solu-
        tions   (Cambridge   Isotope  Laboratories,
        Cambridge,  MA,  or equivalent).    Observe
        the  safety precautions  in  Section 4, and
        the recommendation  in Section  4.1.2.

6.8.2   Dissolve an appropriate amount of  assayed
        reference   material   in   solvent.     For
        example, weigh 1-2 mg  of  2,3,7,8-TCDD to
        three  significant  figures  in  a  10  mL
        ground  glass  stoppered  volumetric  flask
        and  fill  to the  mark  with nonane.   After
        the  TCDD  is  completely dissolved,  transfer
        the  solution to  a clean  15  ml vial  with
        Teflon-lined cap.

6.8.3   Stock standard  solutions  should be checked
        for  signs   of  degradation  prior  to  the
        preparation  of  calibration or  performance
        test  standards.  Reference standards  that
        can  be  used to determine the  accuracy  of
        calibration  standards  are available  from
        Cambridge  Isotope Laboratories.

  6.9   "Secondary  standard  -- Using  stock  solu-
        tions   (Section  6.8),  prepare  secondary
        standard   solutions   containing  the  com-
        pounds  and concentrations shown in Table 4
        in nonane.

 6.10   Labeled compound  stock  standard   --  From
        stock   standard   solutions   prepared   as
        above,  or  from  purchased mixtures, prepare
        this standard  to contain  the  labeled com-
        pounds  at  the  concentrations  shown  in
        Table   4   in nonane.    This  solution  is
        diluted with acetone  prior to use (Section
        10.3.2).

 6.11   Cleanup standard  - Prepare    Cl^-2,3,7,8-
        TCDD at the  concentration shown in Table 4
        in nonane.

 6.12   Internal  standard --  Prepare at  the con-
        centration shown in Table 4  in  nonane.

 6.13   Calibration  standards  (CS1 through CSS)  --
        Combine  the solutions   in  Sections  6.9,
        6.10,  6.11,  and  6.12  to  produce  the five
        calibration  solutions shown  in  Table 4  in
        nonane.   These solutions  permit  the rela-
        tive response  (labeled to unlabeled)  and
        response  factor  to be measured as a func-
        tion of concentration.    The CS3  standard
        is   used   for    calibration   verification
        (VER).

 6.14   Precision and  recovery standard  (PAR)  --
        Used for  determination of initial  (Section
        8.2) and  ongoing  (Section 14.5)  precision
        and  accuracy.   This  solution contains the
        analytes  and labeled  compounds  at the con-
        centrations  listed in Table 4  in nonane.
        This  solution   is  diluted  with  acetone
        prior  to  use (Section 10.3.4).

 6.15   GC   retention   time window defining  solu-
        tions  --  Used  to  define  the  beginning and
        ending  retention times for  the dioxin and
        furan isomers.
                                                         00:)

-------
 6.15.1   DB-5  column window  defining standards  --
          Cambridge  Isotope  Laboratories  ED-1732-A
          (dioxins)   and   ED-1731-A  (furans),   or
          equivalent,    containing    the    compounds
          listed  in Table 5.

   6.16   Isomer  specificity test standards -- Used
          to  demonstrate isomer specificity for  the
          2.3,7,8-tetra-   isomers   of  dioxin   and
          furan.

 6.16.1   Standards for  the  DB-5  column  •-  Cambridge
          Isotope  Laboratories ED-908, ED-908-C,  or
          ED-935,  or   equivalent,    containing   the
          compounds listed  in  Table  5.

 6.16.2   Standards   for   the  08-225   column
          Cambridge  Isotope Laboratories  EF-937  or
          EF-938,  or   equivalent,    containing   the
          compounds listed  in  Table  5.

   6.17   Stability   of    solutions  --    Standard
          solutions  used for  quantitative  purposes
          (Sections  6.9-6.14)  shall  be   analyzed
          within  48  hours  of  preparation  and on a
          monthly  basis  thereafter  for   signs   of
          degradation.      Standards  will   remain
          acceptable   if  the  peak  area  at   the
          quant i tat ion   ai/z   remains   within   ±15
          percent  of   the   area  obtained  in   the
          initial  analysis  of the   standard.     Any
          standards  failing to meet  this  criterion
          should   be   assayed   against    reference
          standards,  as  in Section 6.8.3.,  before
          further use.

      7   CALIBRATION

    7.1   Assemble   the  GCMS  and   establish   the
          operating conditions necessary  to meet  the
          relative retention time specifications  in
          Table 2.

  7.1.1   The following  GC  operating  conditions  may
          be  used  for  guidance  and adjusted   as
          needed to meet the relative retention time
          specifications in  Table 2:
          Injector temp:  270  °C
          Interface temp:  290 °C
          Initial temp and  time: 200 °C, 2  min
          Temp Program:  200-220 °C  at 5 °C/min
                         220 °C for  16 min
                         220-235 °C  at 5 "C/min
                         235 °C for  7 min
                         235-330 °C  at 5 "C/min

          NOTE:   All  portions of  the column which
          connect  the GC  to  the ion source  shall
          remain   at   the   interface   temperature
          specified   above   during   analysis,   to
          preclude  condensation  of  less  volatile
          compounds.

  7.1.2   Mass   spectrometer   (MS)   resolution
          Obtain  a  selected  ion  current  profile
          (SICP) of  each analyte in  Table  4  at the
          two exact masses  specified  in Table 3 and
          at >10,000 resolving power by injecting an
          authentic standard  of  the  PCDDs and PCOFs
          either singly  or as part of  a  mixture in
          which  there  is  no  interference  between
          closely   eluted  components,    using   the
          procedure in Section 13.

7.1.2.1   The analysis time for  PCDDs and PCDFs may
          exceed the long-term mass stability of the
          mass spectrometer.  Because the instrument
          is  operated  in  the high-resolution mode,
          mass drifts  of a few ppm (e.g.,  5  ppm- in
          mass) can have serious  adverse  effects on
          instrument  performance.     Therefore,  a
          mass-drift  correction  is   mandatory.    A
          lock-mass ion  from  the reference compound
          (PFK)   is   used  for   tuning   the   mass
          spectrometer.     The   lock-mass   ion  is
          dependent  on   the   masses  of   the   ions
          monitored within each descriptor, as shown
          in  Table  3.    The  level of  the reference
          compound  (PFK)   metered   into   the   ion
          chamber during HRGC/HRMS analyses  should
          be  adjusted  so that the amplitude  of the
          most intense selected lock-mass ion signal
          (regardless of the descriptor number) does
          not  exceed  10  percent  of  the  full-scale
          deflection  for a given  set  of  detector
          parameters.     Under   those   conditions,
          sensitivity  changes   that   might   occur
          during   the    analysis  can   be   more
          effectively monitored.   NOTE:   Excessive
          PFK (or any other reference substance) may
          cause noise  problems and contamination of
          the ion source resulting in an increase in
          time lost in cleaning the source.

7.1.2.2   By  using  a PFK  molecular  leak,  tune the
          instrument  to  meet  the minimum  required
          resolving  power  of  10,000  (10  percent
          valley) at m/z 304.9824 (PFK)  or any other
          reference  signal close to  m/z  303.9016
          (from TCDF).   By using the peak matching
          unit and  the  PFK  reference  peak,  verify
          that the exact mass of m/z 380.9760 (PFK)
          is within 5 ppm of the  required value.

    7.2   Ion  abundance  ratios,  minimum  levels,
          signal-to-noise   ratios,   and   absolute
          retention   times   --    Inject    the   CS1
10
                                                          010

-------
          calibration  solution  (Table  4)  per  the
          procedure in Section 13 and  the conditions
          in Table 2.

  7.2.1   Measure  the  SICP  areas  for  each  analyte
          and  compute   the   ion  •abundance  ratios
          specified  in   Table   3A.     Compare  the
          computed  ratio  to  the  theoretical  ratio
          given in Table 3A.

7.2.1.1   The  groups  of  m/z's  to be  monitored are
          shown   in   Table  3.     Each   group  or
          descriptor shall  be monitored  in succes-
          sion as a function of GC retention time to
          ensure  that   all   PCDDs   and  PCDFs  are
          detected.      The   theoretical   abundance
          ratios  for  the  m/z's  are given  in Table
          3A,  along with the control  limits of each
          ratio.

7.2.1.2   The mass spectrometer shall be operated in
          a  mass  drift   correction mode,  using per-
          fluorokerosene   (PFIO   to   provide  lock
          masses.   The   lock mass for  each  group of
          m/z's is shown  in Table 3.  Each lock mass
          shall be  monitored and shall  not  vary by
          more  than  ±10  percent   throughout  its
          respective retention  time  window.   Varia-
          tions  of  the  lock mass  by  more  than  10
          percent indicate the presence of coeluting
          interferences   that   may   significantly
          reduce   the    sensitivity   of   the  mass
          spectrometer.     Re-injection  of  another
          aliquot  of  the sample  extract will  not
          resolve  the  problem.   Additional  cleanup
          of the  extract may be  required to remove
          the  interferences.

  7.2.2   All  PCDDs and  PCDFs  shall  be within their
          respective  ratios;  otherwise,  the  mass
          spectrometer  shall  be  adjusted  and  this
          test  repeated  until  the  m/z  ratios  fall
          within  the  limits  specified.     If  the
          adjustment alters  the  resolution  of  the
          mass  spectrometer,  resolution  shall  be
          verified (Section  7.1)  prior to repeat of
          the  test.

  7.2.3   Verify that the HRGC/HRMS instrument meets
          the  minimum levels  in  Table  2.  The peaks
          representing  both   unlabeled  and   labeled
          analytes in the calibration  standards must
          have a signal-to-noise ratio (S/N) greater
          than or 'equal  to  10;  otherwise,  the mass
          spectrometer  shall  be  adjusted  and  this
          test repeated  until  the  minimum levels in
          Table 2 are met.
                                                            7.2.4   The  absolute  retention  time  of
                                               "12
        1,2,3,4-TCDD  (Section 6.12)  shall  exceed
        25.0  minutes  on the  DB-5  column,  and  the
        retention  time  of    C12-1,2,3,4-TCDO  shall
        exceed  15.0  minutes on the DB-225  column;
        otherwise,   the  GC  temperature   program
        shall  be adjusted  and  this test  repeated
        until  the above-stated  minimum  retention
        time  criteria are met.

  7.3   Retention  time windows   --   Analyze  the
        window  defining  mixtures  (Section  6.15)
        using the  procedure in Section 13  (Figures
        2A-2D).   Table 5  gives  the elution  order
        (first/last) of the compound pairs.

  7.4   Isomer  specificity

7.4.1   Analyze   the   isomer   specificity   test
        standards    (Section  6.16)    using   the
        procedure  in Section 13.

7.4.2   Compute the  percent valley between  the GC
        peaks  that  elute   most   closely   to  the
        2,3,7,8-  TCDD   and  TCDF  isomers,  on  their
        respective columns,  per Figure 3.

7.4.3   Verify   that   the   height   of   the  valley
        between the most   closely eluted  isomers
        and  the 2,3,7,8-   isomers  is   less  than 25
        percent (computed  as 100  x/y  in Figure 3).
        If  the valley  exceeds  25  percent,  adjust
        the  analytical conditions and repeat  the
        test  or replace the GC column and  recali-
        brate (Section  7.2  through 7.4).

  7.5   Calibration   with   isotope   dilution
        Isotope dilution   is  used   for   the 15
        2,3,7,8-substituted PCDOs  and PCDFs  with
        labeled compounds   added  to  the  samples
        prior to extraction, and  for  1,2,3,7,8,9-
        HxCDD and OCDF (see Section  16.1).   The
        reference   compound  for   each   unlabeled
        compound is  shown  in Table 6.

7.5.1   A   calibration   curve   encompassing   the
        concentration   range  is  prepared   for  each
        compound  to be determined.    The  relative
        response  (RR)   (unlabeled  to  labeled)  vs.
        concentration   in   standard  solutions  is
        plotted or computed using  a linear regres-
        sion.    Relative   response  is determined
        according   to   the   procedures   described
        below.   A  minimum  of five data points are
        employed for calibration.

7.5.2   The   relative  response  of each  unlabeled
        PCDD/PCDF   and   its  labeled  analog   is
        determined using   the  area   responses of
                                                         Oil
                                                11

-------
                       6-MAY-88    Sir: Voltage 705    Sys: D85US
           Sample 1 Injection 1    Group 2    Mass 303.9016
       100
        80
        60
        40
        20
                   1,3,6,8-TCDF
                                                                     Norm:    3044
                                                    1,2,8.9-TCDF
          25:20  26:40  28:00  29:20  30:40  32:00  33:20  34:40  36:00  37:20  38:40
                     6-MAY-88    Sir: Voltage 705    Sys: DB5US
          Sample 1 Injection 1    Group 2    Mass 319.8965
       100
        80-
        60
        40-
        20
                 1,3,6,8-TCDD
                                                                      Norm:   481
                                                    1,2,8.9-TCDD
                                                   1,2.3.7/1,2.3,8-TCDD
                                                                      2,3,7,8-TCDD
          25:20  26:40  28:00  29:20   30:40- 32:00^x33:20  34:40  36:00   37:20  38:40
                FIGURE 2A  First and Last Eluted Tetra- Oioxin and Furan Isomers
12
012

-------
              6-MAY-88    Sir: Voltage 705    Sys: DB5US
    Sample 1 Injection 1    Group 2    Mass 339.8597
  100
   80
   60
   40
  20-
                                           Norm:    652
           1,3,4.6,8-PeCDF
1,2,3,8,9-PeCDF
     29:20  30:40  32:00  33:20  34:40   36:00  37:20  38:40
                6-MAY-88    Sir: Voltage 705    Sys: DB5US
     Sample 1 Injection 1    Group 2    Mass 355.8546
 100-1
  80
  60-
  40-
  20-
                 1,2,4,7,9-PeCDD
                                           Norm:
                                              1,2,3,8,9-PeCDD

    29:20  30:40  32:00  33:20  34:40  36:00  37:20  38:40
FIGURE 2B  First and Last Eluted Penta- Dioxin and Furan Isomers
                                 013
                              13

-------
                         6-MAY-88    Sir: Voltage 705    Sys: DB5US
              Sample 1 Injection  1    Group 3    Mass 373.8208
         100
          80
         60
         40-
         20-
                                                                      Norm:    560
                   1.2.3.4,6.8-HxCDF
                                                             1.2,3,4.8,9-HxCDF
             39:30  40:00 40:30  41:00  41:30  42:00  42:30 43:00  43:30  44:00  44:30
                       6-MAY-88    Sir: Voltage 705    Sys: DB5US
            Sample 1 Injection 1    Group 3    Mass 389.8156
        100
        00
        60
        40-
        20-
                           1.2,4,6,7,9/1,2,4,6,8,9-HxCDO
                                                                      Norm:    384
                                                           1,2,3,4,6,7-HxCDO
            39:30 40:00  40:30 41:00  41:30  42:00  42:30  43:00 43:30  44:00   44:30
                 FIGURE 2C   First and Last Eluted Hexa- Dioxin and Furan Isomers
14
014

-------
             6-MAY-88    Sir: Voltage 705    Sys: D85US
  Sample 1  Injection 1    Group 4    Mass 407.7818

             1,2,3,4,6,7,8-HpCDF
 100    I   ..  '     •-	                                  Norm:    336
 80
 60^
 40
 20
                          1.2,3,4,7,8,9-HpCDF
    45:20   46:40   48:00   49:20   50:40   52:00   53:20   54:40   56:00   57:20

             6-MAY-88    Sir: Voltage 705    Sys: DB5US
  Sample 1 Injection 1    Group 4    Mass 423.7766
             1,2,3,4,6,7,9-HpCDO
         ,X~
80-
60-
40
20
                                                                Norm:    282
                     1,2,3,4.6,7.8-HpCDD
    45:20   46:40   48:00  49:20   50:40   52:00   53:20   54:40   56:00   57:20

             6-MAY-88    Sir: Voltage 705    Sys: DB5US
  Sample 1 Injection 1     Group 4    Mass 441.7428


100,                                            ,                Norm:    13
                                                   OCDF
                                                I
 80-j

 60-

 40

 20-I
                                               \
    45:20   46:40  48:00   49:20   50:40   52:00   53:20   54:40   56:00   57:20

            6-MAY-88    Sir: Voltage 705  '   Sys: DB5US
 Sample 1 Injection 1     Group 4    Mass  457.7377

100-,                                           ,   OCDD         Norm:

 80-I

 60

 40

 20-I
   45:20   46:40  48:00   49:20   50:40   52:00   53:20   54:40   56.00   57:20


         FIGURE 2D  First and Last Eluted Hepta- Dioxin and  Furan Isomers
                                         015
                                                                                     15

-------
              3A   DB225 Column

                         21-APR-88   Sir: Voltage 705    Sys: DB225
              Sample 1 Injection 1    Group 1    Mass 305.8987
              Text: COLUMN PERFORMANCE
                                                2,3,7,8-TCDF
                Norm:
3466
             100
                                                             1,2,3,9-TCDF
             20
                16:10 16:20 16:30 16:40 16:50 17:00 17:10 17:20 17:30 17:40 17:50 18:00
            3B    DBS Column
            100
  FIGURE 3  Valley between 2,3,7,8- Tetra Dioxin and Furan Isomers and Other Closely Eluted Isomers
16
016

-------
7.5.3
        both   the   primary   and  secondary  m/z's
        specified in Table 3, for each calibration
        standard, as follows:
            RR  =
                           (A
                              1
                                          l
                          (A,
where,

A   and A    are the areas of  the primary
   and  secondary  m/z's for  the  unlabeled
   compound.

A.   and A.   are the areas of  the primary
   and  secondary  m/z's   for  the  labeled
   compound.

C.   is  the  concentration of  the  labeled
   compound  in the calibration standard.

C  is  the  concentration of  the  unlabeled
   compound  in the calibration standard.

To  calibrate  the  analytical   system  by
isotope dilution,   inject a 1.0 uL aliquot
of calibration  standards  CS1  through  CSS
(Section  6.13  and  Table 4)   using  the
procedure in Section 13 and the conditions
in Table 2.  Compute the relative response
     at each concentration.
7.5.4   Linearity  --  If  the  relative response for
        any  compound  is  constant  (less  than  20
        percent coefficient of variation) over the
        5-point  calibration  range,  an  averaged
        relative  response may  be  used  for  that
        compound;  otherwise,  the complete calibra-
        tion curve for that compound shall be used
        over the 5-point calibration range.

  7.6   Calibration  by internal  standard  --  The
        internal  standard method   is   applied  to
        determination  of  non-2,3,7,8-substituted
        compounds having no labeled analog in this
        method,  and   to  measurement  of  labeled
        compounds  for  intralaboratory  statistics
        (Sections 8.4 and 14.5.4).

7.6.1   Response  factors  --   Calibrat.ion requires
        the determination of  response factors (RF)
        defined by the following equation:
                    1

RF  =    ' s


where,

A   and  A    are the areas of  the primary
   and  secondary m/z's  for  the  compound
                                                          7.6.2
                                                          7.6.3
                                                   7.7
                                                   7.8
                                                 7.8.1
                                                          7.8.2
                                                             to  be calibrated.
                                                             only  one  m/z  for
                                                             See Table 3.)
                                                                                   (NOTE:    There  is
                                                                                   Cl4-2,3,7,8-TCDD.
                                                          A.    and A.   are the areas of the primary
                                                             and   secondary  m/z'
                                                             internal standard.
                                                                                              for   the  GCMS
                                                                  C.   is  the  concentration  of  the  GCMS
                                                                      internal   standard   (Section  6.12  and
                                                                      Table 4).
                                                                  C  is the concentration of the compound in
                                                                     the  calibration  standard.
                                                          To  calibrate  the  analytical  system  by
                                                          internal standard, inject a 1.0 uL aliquot
                                                          of  calibration standards CS1  through CSS
                                                          (Section  6.13  and  Table  4)  using  the
                                                          procedure in Section 13 and the conditions
                                                          in  Table  2.    Compute  the response factor
                                                          (RF) at each concentration.

                                                          Linearity --  If   the  response  factor (RF)
                                                          for any compound  is constant (less than 35
                                                          percent coefficient of variation) over the
                                                          5-point ^calibration  range,  an  averaged
                                                          response  factor  may  be  used  for  that
                                                          compound; otherwise, the complete calibra-
                                                          tion curve for that compound shall be used
                                                          over the 5-point  range.

                                                          Combined calibration  --  By using calibra-
                                                          tion solutions (Section  6.13  and Table 4)
                                                          containing   the   unlabeled   and  labeled
                                                          compounds, and the  internal  standards,  a
                                                          single  set   of analyses  can  be used  to
                                                          produce calibration curves for the isotope
                                                          dilution  and  internal   standard  methods.
                                                          These  curves  are   verified   each  shift
                                                          (Section  14.3)  by analyzing  the calibra-
                                                          tion verification standard (VER, Table 4).
                                                          Recalibration  is  required  if  calibration
                                                          verification   criteria   (Section  14.3.4)
                                                          cannot be met.
                                                          Data   storage    --   MS   data  shall
                                                          collected, recorded, and stored.
                                                                                                  be
                                                          Data  acquisition  --  The  signal  at  each .
                                                          exact m/z shall  be collected repetitively
                                                          throughout   the   monitoring   period   and
                                                          stored on a mass storage device.

                                                          Response     factors     and     multipoint
                                                          calibrations --  The data  system  shall  be
                                                          used  to  record  and  maintain  lists  of
                                                          response  factors   (response  ratios   for
                                                       017
                                                                                                 17

-------
           isotope     dilution)    and    multipoint
           calibration   curves.     Computations  of
           relative  standard  deviation  (coefficient
           of   variation)   shall  be  used  to   test
           calibration   linearity.     Statistics  on
           initial   performance  (Section  8.2)  and
           ongoing  performance  (Section  14.5)  shall
           be computed and maintained.

      8    QUALITY ASSURANCE/QUALITY COMTROL

    8.1    Each  laboratory that uses  this  method is
           required   to   operate  a  formal   quality
           assurance  program  (Reference  16).   The
           minimum   requirements   of   this   program
           consist  of  an  initial   demonstration of
           laboratory capability, analysis of  samples
           spiked with  labeled compounds to evaluate
           and document data quality, and analysis of
           standards and blanks  as  tests of continued
           performance.    Laboratory  performance is
           compared    to    established   performance
           criteria  to determine  if the  results of
           analyses meet the performance characteris-
           tics  of  the method.   If  the method  is to
           be applied routinely  to  samples containing
           high  solids  with   very   little  moisture
           (e.g., soils,  filter cake,  compost)  or to
           an   alternate   matrix,   the  high  solids
           reference  matrix (Section  6.6.2)  or the
           alternate  matrix (Section 6.6.4)   is  sub-
           stituted  for   the   reagent  water  matrix
           (Section 6.6.1)  in all performance  tests.

  8.1.1    The  analyst  shall make  an initial  demon-
           stration   of   the   ability   to  generate
           acceptable  accuracy  and  precision   with
           this  method.   This  ability  is established
           as described in Section 8.2.

  8.1.2    The  analyst  is  permitted to  modify  this
           method to improve separations or lower the
           costs  of  measurements,  provided  that all
           performance specifications  are  met.   Each
           time  a modification  is made  to the  method,
           the analyst is  required  to repeat the pro-
           cedures  in Sections  7.2 through  7.4 and
           Section   8.2    to    demonstrate   method
           performance.

  8.1.3    Analyses of blanks  are required to demon-
           strate freedom  from  contamination (Section
           3.2).    The  procedures  and  criteria for
           analysis  of  a  blank   are  described  in
           Section 8.5.

  8.1.4    The   laboratory  shall  spike all   samples
           with  labeled  compounds  to  monitor method
           performance.   This  test  is  described in
         Section 8.3.  When results of these spikes
         indicate atypical  method  performance  for
         samples, the samples are  diluted to bring
         method   performance   within   acceptable
         limits.    Procedures  for  dilutions  are
         given in Section 16.4.

 8.1.5    The laboratory shall, on an ongoing basis,
         demonstrate through  calibration verifica-
         tion and the analysis of the precision and
         recovery standard that  the analytical sys-
         tem is  in  control.   These procedures  are
         described in Sections 14.1 through  14.5.

 8.1.6    The laboratory  shall maintain  records  to
         define the quality of  data that is gener-
         ated.    Development of  accuracy statements
         is described in Section 8.4.

  8.2    Initial  precision   and   accuracy   --   To
         establish the ability  to  generate  accept-
         able precision  and  accuracy,  the  analyst
         shall  perform the following operations.

 8.2.1    For low solids (aqueous samples), extract,
         concentrate,  and  analyze  four  1-liter
         aliquots of reagent  water  spiked with  the
         diluted  precision and  recovery  standard
         (PAR)  (Sections 6.14 and 10.3.4) according
         to the  procedures in Sections  10  through
         13.  For an alternate  sample matrix, four
         aliquots of the alternate matrix are used.
         All  sample  processing  steps,  including
         preparation   (Section    10),   extraction
         (Section  11),   and   cleanup  (Section  12)
         that are to be used for processing samples
         shall  be included in this test.

 8.2.2    Using  results of the set of four analyses,
         compute  the  average  concentration  (X)  of
         the extracts  in  ng/mL  and  the  standard
         deviation  of  the  concentration  (s)   in
         ng/mL    for   each  compound,   by  isotope
         dilution  for  PCODs  and  PCDFs  with   a
         labeled analog,  and  by  internal  standard
         for  labeled  compounds.    Calculate  the
         recovery of the labeled compounds.

 8.2.3    For each  unlabeled  and  labeled compound,
         compare  s  and  X  with  the. corresponding
         limits for  initial precision and accuracy
         in Table 7.   If  s and  X for all compounds
         meet   the   acceptance    criteria,   system
         performance is acceptable  and  analysis  of
         blanks  and   samples  may   begin.     If,
         however,  any  individual  s  exceeds  the
         precision limit or any  individual  X falls
         outside  the  range   for  accuracy,  system
18
013
                                                                o

-------
        performance   is   unacceptable   for   that
        compound.   Correct  the  problem and repeat
        the test (Section 8.2).   The concentration
        limits  in   Table  7 for  labeled compounds
        are  based  on  the requirement  that   the
        recovery of each  labeled  'compound  be  in
        the range of 25-150%.

  8.3   The laboratory shall spike alt samples  and
        OC  aliquots   with   the  diluted   labeled
        compound  spiking solution  (Sections  6.10
        and  10.3.2) to  assess  method performance
        on the sample matrix.
8.3.1
8.3.2
8.3.3
8.4.1
8.4.2
Analyze  each  sample   according   to
procedures in Sections  10 through  13.
                                                the
Compute  the  percent  recovery  (R)  of  the
labeled compounds  in  the  labeled compound
spiking standard and  the  cleanup standard
using   the   internal   standard   method
(Section 7.6).

The recovery of each labeled compound must
be within 25-150%.   If the recovery of any
compound  falls  outside  of  these   limits,
method  performance  i's  unacceptable  for
that compound in that sample.  To overcome
such   difficulties,   water   samples   are
diluted  and  smaller  amounts  of  soils,
sludges, sediments  and  other matrices are
reanalyzed per Section 17.
  8.4   Method   accuracy   for   samples  shall   be
        assessed and records shall be maintained.
After  the  analysis of  five samples  of a
given  matrix  type  (water,   soil,  sludge,
pulp, etc)  for  which  the labeled compound
spiking  standards   pass   the   tests   in
Section 8.3,  compute the  average percent
recovery (R) and the standard deviation  of
the percent recovery  (SR)  for  the labeled
compounds  only.    Express  the  accuracy
assessment as a  percent  recovery interval
         from  R
                  2SR to R
                      2S   for each matrix.
                       K
For example,  if  R  = 90% and  SR  = 10% for
five  analyses   of  pulp,   the  accuracy
interval is expressed as 70-110%.

Update  the accuracy  assessment  for  each
compound in each matrix on a  regular basis
(e.g.,   after   each  5-10   new  accuracy
measurements).
  8.5   Blanks   --  Reference  matrix   blanks  -are
        analyzed   to   demonstrate   freedom   from
        contamination (Section 3.2).
S.5.1   Extract  and  concentrate a  1-liter  reagent
        water  blank  (Section  6.6.1),  high  solids
        reference  matrix  blank  (Section  6.6.2),
        paper  matrix   blank   (Section   6.6.3)   or
        alternate  reference matrix  blank  (Section
        6.6.4)   with   each  sample  set   (samples
        started  through the extraction  process  on
        the same 12-hour  shift,  to  a maximum of  HO
        samples).   Analyze  the blank  immediately
        after   analysis   of   the   precision  and
        recovery   standard   (Section   14.5)   to
        demonstrate freedom  from contamination.

8.5.2   If any of  the PCDDs or  PCDFs (Table 1)  or
        any  potentially   interfering  compound  is
        found  in blank  at  greater  than  the minimum
        level   (Table   2),  assuming  a   response
        factor of  1 relative  to the  C.^-1,2,3,4-
        TCOD  internal   standard  for compounds not
        listed in  Table 1, analysis of  samples  is
        halted until  the  source of  contamination
        is   eliminated   and   a   blank   shows   no
        evidence  of  contamination  at  this  level.
        NOTE:    All  samples   associated  with   a
        contaminated  method   blank must   be re-
        extracted   and    reanalyzed   before  the
        results  may  be   reported  for   regulatory
        compliance purposes.

  8.6   The   specifications   contained    in  this
        method can be met  if  the apparatus used  is
        calibrated properly and then maintained  in
        a  calibrated state.    The  standards  used
        for  calibration  (Section  7),  calibration
        verification   (Section   14.3),   and   for
        initial  (Section  8.2)  and ongoing (Section
        14.5)  precision   and  recovery  should  be
        identical,   so   that    the  most   precise
        results  will  be obtained.   A GCMS  instru-
        ment  will   provide  the  most  reproducible
        results  if dedicated  to the settings and
        conditions required  for  the  analyses  of
        PCDDs  and  PCDFs by this  method.

  8.7   Depending  on   specific  program  require-
        ments,  field replicates may be  collected
        to determine  the  precision of  the sampling
        technique,   and   spiked  samples   may  be
        required  to  determine the  accuracy  of  the
        analysis when  the internal  standard method
        is used.
                                                              9   SAMPLE    COLLECTION,
                                                                  HANDLING
                                PRESERVATION,    AND
                                                    9.1    Collect samples  in amber glass  containers
                                                          following  conventional  sampling  practices
                                                          (Reference  17).     Aqueous   samples  which
                                                          flow  freely  are  collected in  refrigerated
                                                                                                          19

-------
          bottles  using  automatic   sampling  equip-
          ment.  Solid samples are collected as grab
          samples using wide mouth jars.

   9.2    Maintain  samples  at  0-4   °C  in  the  dark
          from   the   time    of   collection   until
          extraction.     If   residual   chlorine  is
          present  in  aqueous  samples,  add  80  mg
          sodium  thiosulfate  per  liter  of  water.
          EPA Methods 330.4 and 330.5 may be used to
          measure residual chlorine (Reference 18).

   9.3    Perform sample  analysis within  40 days of
          extraction.

     10    SAMPLE PREPARATION

          The  sample  preparation process  involves
          modifying the physical  form  of  the sample
          so   that   the   PCOOs  and  PCDFs   can  be
          extracted  efficiently.    In  general,  the
          samples must be in a liquid form or in the
          form of finely divided solids in order for
          efficient extraction to take place.  Table
          8   lists   the    phase(s)   and   quantity
          extracted  for  various  sample  matrices.
          Samples  containing   a   solid  phase  and
          samples  containing particle  sizes  larger
          than  1   mm require  preparation  prior  to
          extraction.     Because   PCODs/PCDFs   are
          strongly associated with particulates, the
          preparation  of  aqueous samples  is depen-
          dent on  the  solids content of the sample.
          Aqueous  samples containing less  than one
          percent   solids   are  extracted  in   a
          separator/   funnel.     A   smaller  sample
          aliquot   is   used   for   aqueous   samples
          containing  one  percent  solids   or  more.
          For  samples  expected or  known  to contain
          high levels of  the PCDDs and/or PCDFs, the
          smallest sample size representative of the
          entire  sample  should  be used,  and  the
          sample   extract   should  be   diluted,  if
          necessary, per  Section 16.4.

   10.1    Determine percent solids

 10.1.1    Weigh 5-10 g  of sample (to three signifi-
          cant figures)  into a tared beaker.  NOTE:
          This aliquot  is used only for determining
          the  solids content  of the sample, not for
          analysis of PCDDs/PCDFs.

 10.1.2    Dry overnight (12 hours minimum) at 110 ±5
          °C, and cool in a dessicator.
     10.1.3    Calculate percent  solids  as  follows:
                                             x 100
       % sol ids =
       weight of sample after drying
       weight of sample before drying

10.2   Determine particle size
     10.2.1    Spread  the  dried   sample   from  Section
              10.1.2  on  a  piece  of  filter  paper  or
              aluminum foil  in a fume hood or  glove box.

     10.2.2    Estimate the size of the particles  in the
              sample.     If   the  size  of  the  largest
              particles  is   greater  than   1   mm,  the
              particle size must  be  reduced to 1  mm or
              less prior to  extraction.

       10.3    Preparation of aqueous  samples  containing
              one  percent   solids  or   less  --   The
              extraction procedure  for  aqueous  samples
              containing  less  than   or   equal  to  one
              percent   solids   involves   filtering  the
              sample,   extracting  the particulate  phase
              and the  filtrate separately, and combining
              the extracts  for analysis.   The  aqueous-
              portion   is  extracted  by   shaking   with
              methylene chloride in a separatory funnel.
              The  particulate  material   is   extracted
              using the SOS  procedure.

     10.3.1    Mark the  original  level  of  the  sample on
              the sample  bottle  for  reference.   Weigh
              the sample in the bottle on a top loading
              balance  to ±1  g.

     10.3.2    Dilute a sufficient  volume  of the labeled
              compound stock solution by  a factor  of 50
              with  acetone   to  prepare   the  labeled
              compound spiking solution.    1.0  mL  of the
              diluted   solution  is  required  for  each
              sample,   but  no  more  solution   should  be
              prepared  than  can  be  used  in  one  day.
              Spike 1.0 mL of  the diluted solution into
              the sample bottle.  Cap the bottle and mix
              the sample by careful  shaking.   Allow the
              sample to equilibrate  for  1-2 hours, with
              occasional shaking.

     10.3.3    For  each  sample  or  sample  set  (to  a
              maximum   of  20  samples)  to  be  extracted
              during the  same  12-hour shift,  place two
              1.0  liter aliquots  of reagent  water  in
              clean 2  liter  separatory flasks.

     10.3.4    Spike  1.0  mL   of   the  diluted  labeled
              compound  spiking  standard  (Section  6.10)
              into  one  reagent   water  aliquot.    This
              aliquot  will serve  as  the  blank.   Dilute
              10  uL   of  the   precision   and  recovery
20
020

-------
          standard (Section  6.14)  to  2.0  mt_  with
          acetone.   Spike  1.0 mL  of  the  diluted
          precision and  recovery  standard  into  the
          remaining  reagent  water  aliquot.    This
          aliquot   will  serve  as  the  PAR  (Section
          14.5).

 10.3.5    Assemble a  Buchner  funnel  on  top  of  a
          clean  1 L  filtration  flask.    Apply  a
          vacuum  to the  flask,  and  pour  the  entire
          contents of  the sample bottle  through  a
          glass fiber filter (Section 5.5.4)  in  the
          Buchner   funnel,   swirling   the   sample
          remaining  in  the  bottle  to  suspend  any
          particulates.

 10.3.6    Rinse the sample bottle  twice  with 5 ml of
          reagent   water  to  transfer  any  remaining
          particulates  onto the filter.

 10.3.7    Rinse the  any  particulates  off  the sides
          of the  Buchner  funnel with small  quanti-
          ties  of  reagent water.

 10.3.8    Weigh the  empty sample bottle  on  a  top-
          loading  balance  to ±1  g.   Determine  the
          weight  of  the  sample by  difference.   Do
          not discard the bottle at  this point.

 10.3.9    Extract  the filtrates using  the  procedures
          in Section  11.1.1.

10.3.10    Extract  the particulates using  the proce-
          dures in Section 11.1.2.

   10.4    Preparation of   samples  containing greater
          than  one percent solids

 10.4.1    Weigh a  well-mixed aliquot of each sample
          (of  the same matrix  type)  sufficient  to
          provide  10 g of dry  solids (based  on  the
          solids  determination in  10.1.3)   into  a
          clean beaker  or glass jar.

 10.4.2    Spike   1.0  ml   of   the   diluted  labeled
          compound spiking solution  (Section 10.3.2)
          into  the sample aliquot(s).

 10.4.3    For  each  sample  or  sample   set  (to  a
          maximum  of  20  samples)  to  be  extracted
          during  the same 12-hour shift,   weigh  two
          10 g  aliquots of the appropriate reference
          matrix (Section 6.6)  into  clean  beakers or
          glass jars.

 10.4.4    Spikre   1.0  mL   of   the   diluted  labeled
          compound   spiking   solution    into   one
          reference matrix  aliquot.    This  aliquot
          will  serve as the  blank.  Spike 1.0 mL of
          the   diluted   precision    and    recovery
         standard    (Section    10.3.4)    into    the
         remaining  reference  matrix aliquot.   This
         aliquot  will  serve  as  the  PAR  (Section
         14.5).

10.4.5   Stir   or   tumble   and   equilibrate   the
         aliquots for  1-2  hours.

10.4.6   Extract  the aliquots using  the  procedures
         in Section  11.

  10.5   Multiphase  samples

10.5.1   Pressure filter  the  sample,  blank,  and PAR
         aliquots through  Whatman GF/0 glass  fiber
         filter  paper.    If  necessary,  centrifuge
         these  aliquots for  30 minutes  at  greater
         than 5000  rpm prior  to filtration.

10.5.2   Discard  any  aqueous  phase  (if  present).
         Remove any  non-aqueous liquid (if present)
         and  reserve   for  recombination  with  the
         extract   of   the  solid  phase  (Section
         11.1.2.5).   Prepare  the filter  papers  of
         the  sample  and  QC  aliquots  for  particle
         size   reduction   and    blending  (Section
         10.6).

  10.6   Sample grinding,  homogenization, or blend-
         ing -- Samples with  particle sizes greater
         than  '1   mm   (as  determined  by  Section
         10.2.2)  are  subjected  to grinding,  homo-
         genization,  or  blending.    The   method  of
         reducing  particle size  to less   than  1  mm
         is  matrix  dependent.     In  general,  hard
         particles  can be reduced  by grinding with
         a mortar and pestle.  Softer particles can
         be  reduced by grinding  in a  Wiley mill  or
         meat  grinder,  by  homogenization,  or  by
         blending.

10.6.1   Each  size  reducing  preparation procedure
         on  each  matrix  shall  be verified  by run-
         ning  the tests  in  Section 8.2  before the
         procedure  is employed  routinely.

10.6.2   The  grinding, homogenization, or blending
         procedures  shall  be  carried  out  in a glove
         box or  fume hood to  prevent  particles from
         contaminating the work environment.

10.6.3   Grinding  --  Tissue samples,  certain papers
         and  pulps,  slurries,  and  amorphous solids
         can  be  ground in  a  Wiley  mill  or  heavy
         duty  meat  grinder.   In  some cases,  reduc-
         ing   the   temperature   of  the  sample  to
         freezing  or to dry  ice  or liquid nitrogen
         temperatures  can  aid   in   the  grinding
         process.    Grind  the  sample  aliquots from
         Section   10.4.5  or   10.5.2   in  a  clean
                                                        021
                                                 21

-------
          grinder.  Do not allow  the  sample  tempera-
          ture to exceed 50 °C.   Grind  the blank and
          reference  matrix aliquots  using  a clean
          grinder.

 10.6.4   Homogenization  or  blending  --  Particles
          that   are  not   ground  effectively,   or
          particles greater  than  1 mm  in size after
          grinding, can  often be  reduced in  size by
          high  speed  homogenization  or  blending.
          Homogenize and/or blend the sample, blank,
          and  PAR  aliquots   from Section   10.4.5,
          10.5.2, or 10.6.3.

 10.6.5   Extract  the  aliquots using the procedures
          in Section 11.

     11   EXTRACTION AND CONCENTRATION

   11.1   Extraction  of  filtrates  --   extract  the
          aqueous  samples,  blanks, and PAR  aliquots
          according to the following  procedures.

 11.1.1   Pour the  filtered aqueous  sample  from the
          filtration  flask   into  a  2-1 separatory
          funnel.   Rinse the  flask twice  with  5 ml
          of  reagent  water and add these  rinses to
          the  separatory funnel.    Add  60  mL methy-
          lene   chloride   to  the  'Sample  bottle
          (Section   10.3.8)°,    seal,and  shake   60
          seconds to rinse the  inner  surface.

 11.1.2   Transfer  the  solvent  to  the separatory
          funnel  and  extract  the sample by  shaking
          the  funnel   for  2  minutes  with   periodic
          venting.    Allow   the   organic   layer  to
          separate  from  the  water  phase  for   a
          minimum  of  10 minutes.   If  the  emulsion
          interface between layers  is more  than one-
          third  the  volume  of   the  solvent layer,
          employ  mechanical  techniques to  complete
          the  phase  separation (e.g., a glass stir-
          ring rod).   Drain  the  methylene  chloride
          extract  into a sol vent-rinsed glass funnel
          approximately   one-ha If   full   of  clean
          sodium  sulfate.   Set up the glass funnel
          so  that  it will   drain  directly  into   a
          solvent-rinsed   500-mL    K-0   concentrator
          fitted  with  a  10  ml  concentrator tube.
          NOTE:    Experience  with  aqueous  samples
          high in dissolved organic materials (e.g.,
          paper mill effluents) has shown that acid-
          ification   of    the    sample  prior   to
          extraction  may  reduce   the  formation  of
          emulsions.  Paper  industry  methods suggest
          that  the  addition  of   up  to  400 mL  of
          ethanol  to  a  1 L effluent  sample  may also
          reduce   emulsion   formation.      However,
          studies by  the Agency to date suggest that
          the  effect may be a result  of  the  dilution
            of  the  sample,  and  that  the addition of
            reagent  water  may  serve  the  same  function.
            Mechanical  techniques may still be  neces-
            sary  to  complete the  phase separation.   If
            either  of  these  techniques  is  utilized,
            the   laboratory  must  perform  the  startup
            tests described in Section  8.2 using  the
            same  techniques.

  11.1.3    Extract  the water sample two  more  times
            using 60  mL   of  fresh  methylene  chloride
            each  time.  Drain  each extract  through  the
            funnel  containing  the sodium sulfate  into
            the   K-0  concentrator.    After  the  third
            extraction,  rinse the  separatory  funnel
            with  at  least  20 mL of  fresh  methylene
            chloride,  and  drain this  rinse  through  the
            sodium   sulfate   into   the   concentrator.
            Repeat  this rinse  at  least twice.

  11.1.4    The   extract   of   the  filtrate   must   be
            concentrated   before  it   is  combined  with
            the   extract   of   the  particulates   for
            further  cleanup.   Add  one  or  two  clean
            boiling  chips  to the  receiver and attach  a
            three-ball  macro  Snyder  column.    Pre-wet
            the column by  adding  approximately 1  mL of
            hexane   through  the  top.    Place  the  K-D
            apparatus  in  a  hot water bath so  that  the
            entire  lower   rounded surface of  the  flask
            is bathed  with steam.

  11.1.5    Adjust   the   vertical   position  of   the
            apparatus   and  the  water  temperature  as
            required to complete the  concentration in
            15-20 minutes.    At   the  proper  rate  of
            distillation,  the  balls  of  the  column will
            actively chatter but  the chambers will  not
            flood.

  11.1.6    When  the  liquid  has  reached an  apparent
            volume  of 1 mL,  remove  the  K-D  apparatus
            from  the -bath  and allow the  solvent  to
            drain and  cool for  at  least  10  minutes.
            Remove   the  Snyder  column  and  rinse  the
            flask and  its  lower joint  into  the concen-
            trator  tube with  1-2  ml  of  hexane.  A 5 mu
            syringe is recommended  for  this operation.

  11.1.7    The  concentrated  extracts of  the filtrate
            and   the  particulates are  combined  using
            the  procedures in  Section  11.2.13.

     11.2    Soxhlet/Dean-Stark extraction of  solids --
            Extract  the  solid samples,  particulates,
            blanks,  and PAR aliquots using  the follow-
            ing  procedure.

  11.2.1    Charge  a clean extraction thimble with 5.0
            g of  100/200  mesh  silica (Section 6.5.1.1)
22
022

-------
          and 100 g  of  quartz sand (Section 6.3.2).
          NOTE:    Do  not disturb  the  silica  layer
          throughout the extraction process.

 11.2.2   Place  the  thimble  in  a  clean  extractor.
          Place 30-40 ml  of  toluene' in the receiver
          and 200-250 mL of toluene in the flask.

 11.2.3   Pre-extract the glassware by heating  the
          flask until the toluene  is  boiling.   When
          properly  adjusted,  1-2  drops  of  toluene
          per second will  fall  from  the condenser
          tip  into   the   receiver.     Extract  the
          apparatus for three hours minimum.

 11.2.4   After pre-extraction,  cool and disassemble
          the  apparatus.    Rinse  the  thimble  with
          toluene and allow to air dry.

 11.2.5   Load the wet  sample from Sections 10.4.6,
          10.5.2,  10.6.3, or 10.6.4,  and  any non-
          aqueous  liquid from  Section 10.5.2  into
          the thimble and manually mix into the sand
          layer with a clean metal spatula carefully
          breaking up any large lumps of sample.   If
          the  material   to   be   extracted  is  the
          particulate matter  from  the filtration of
          an aqueous sample, add the filter paper to
          the thimble also.

 11.2.6   Reassemble the  pre-extracted SOS apparatus
          and add  a  fresh charge  of  toluene  to  the
          receiver and reflux flask.

 11.2.7   Apply power to  the heating mantle to begin
          refluxing.    Adjust  the reflux  rate  to
          match the  rate  of  percolation through  the
          sand and  silica beds until  water removal
          lessens  the  restriction  to  toluene flow.
          Check the apparatus for  foaming frequently
          during  the first  2 hours  of extraction.
          If foaming  occurs,  reduce the reflux rate
          until foaming subsides.

 11.2.8   Drain  the  water from the receiver  at  1-2
          hours  and  8-9 hours,  or  sooner   if  the
          receiver  fills with  water.    Reflux  the
          sample  for  a  total of  16-24  hours.   Cool
          and disassemble the apparatus.  Record the
          total volume of water collected.

 11.2.9   Remove  the distilling  flask.    Drain  the
          water from the  Oean Stark receiver and add
          any toluene in  the  receiver to the extract
          in the flask.

11.2.10   For  solid  samples,  the  extract must  be
          concentrated  to approximately  10 mL prior
          to back extraction.   For the particulates
          filtered   from  an aqueous  sample,  the
          extract  must  be  concentrated  prior  to
          combining   with   the   extract    of   the
          filtrate.  Therefore, add one or two clean
          boiling  chips  to  the round  bottom flask
          and  attach  a   three-ball   macro  Snyder
          column.    Pre-wet  the   column  by adding
          approximately 1 mL  of toluene through the
          top.   Place the  round  bottom  flask  in a
          heating mantle  and  apply heat as  required
          to  complete  the  concentration   in 15-20
          minutes.   At the  proper rate of distilla-
          tion,   the  balls  of   the  column  will
          actively chatter but  the chambers  will not
          flood.

11.2.11    When  the  liquid  has reached  an  apparent
          volume  of  10 mL,  remove  the round bottom
          flask  from the  heating mantle  and allow
          the solvent to drain  and cool for  at least
          10 minutes.  Remove the Snyder column.

11.2.12    If the extract  is from a solid sample, not
          the  particulates  from  an  aqueous  sample,
          transfer the concentrated extract  to a 250
          mL  separatory   funnel.    Rinse  the flask
          with  toluene  and add  the  rinse  to  the
          separatory  funnel.    Proceed  with  back
          extraction per Section 11.3.

11.2.13    If  the extract  is  from  the partjculates
          from  an aqueous  sample,  it  must  be com-
          bined with the  concentrated  extract of the
          filtrate  (Section  11.1.7)  prior  to  back
          extraction.    Assemble  the  glass funnel
          filled  approximately  one-half   full  with
          sodium  sulfate  from  Section  11.1.2  such
          that  the  funnel  will drain  into  the K-D
          concentrator from Section  11.1.7  contain-
          ing  the  concentrated  methylene   chloride
          extract of the  filtrate.  Pour the concen-
          trated  toluene  extract of the particulates
          through  the  sodium  sulfate  into  the K-0
          concentrator.     Rinse   the  round-bottom
          flask  with  three   15-20   mL  volumes  of
          hexane,  and  pour  each   rinse  through   the
          sodium  sulfate  into  the K-0  concentrator.
          Add one  or two fresh boiling chips to the
          receiver  and attach  the three-ball macro
          Snyder  column  to  the   K-0  concentrator.
          Pre-wet  the column by adding approximately
          1 mL  of hexane to  the  top  of the column.
          Concentrate   the   combined   extract   to
          approximately  10  mL  (the  volume  of  the
          toluene).   Remove  the  K-0  apparatus  from
          the  bath  and  allow  the  solvent   to drain
          and cool  for at least 10 minutes.  Remove
          the Snyder column.    Transfer the  contents
          of  the K-0  concentrator to a pre-rinsed
          250 mL  separatory  funnel.   Rinse  the flask
                                                         023
                                                  23

-------
          and lower joint with  three 5 ml volumes of
          hexane, and  add  each rinse to the separa-
          tory funnel.  Proceed with back extraction
          per Section  11.3.

   11.3    Back extraction with  base and acid

 11.3.1    Spike  1.0   mL  of   the   cleanup  standard
          (Section 6.11) into  the  separatory funnels
          containing   the  sample   and  QC  extracts
          (Section 11.2.12 or  11.2.13).

 11.3.2    Partition  the extract   against  50  mL  of
          potassium   hydroxide   solution   (Section
          6.1.1).  Shake for 2  minutes with periodic
          venting  into a hood.   Remove  and discard
          the aqueous  layer.    Repeat the base wash-
          ing  until  no  color  is  visible   in   the
          aqueous  layer,  to  a  maximum of four wash-
          ings.   Minimize contact  time  between  the
          extract  and  the base to prevent degrada-
          tion  of  the PCDDs  and  PCDFs.   Stronger
          potassium   hydroxide   solutions   may  be
          employed  for  back   extraction,  provided
          that  the laboratory  meets  the specifica-
          tions  for   labeled  compound  recovery  and
          demonstrates  acceptable performance using
          the procedures in Section 8.2.

 11.3.3    Partition  the extract   against  50  mL  of
          sodium  chloride  solution (Section  6.1.3)
          in the same  way as with  base.  Discard  the
          aqueous layer.

 11.3.4    Partition  the extract   against  50  mL  of
          sulfuric acid  (Section  6.1.2)  in the same
          way as with  base.  Repeat  the acid washing
          until  no  color is  visible  in  the aqueous
          layer, to a  maximum  of  four washings.

 11.3.5    Repeat  the  partitioning against   sodium
          chloride solution  and discard the aqueous
          layer.

 11.3.6    Pour each  extract  through a drying  column
          containing 7 to  10 cm of anhydrous  sodium
          sulfate.  Rinse the  separatory funnel with
          30-50  mL  of  toluene  and pour  through  the
          drying column.   Collect each extract in  a
          500  mL round  bottom flask.   Concentrate
          and  clean  up the  samples and  QC aliquots
          per Sections 11.4 and 12.

   11.4    Macro-concentration   --  Concentrate   the
          extracts  in  separate 100-mL  round  bottom
          flasks on a  rotary evaporator.

 11.4.1    Assemble  the rotary evaporator  according
          to  manufacturer's   instructions,  and warm
          the  water  bath  to  45  °C.    On  a daily
          basis,  preclean  the  rotary  evaporator by
          concentrating  100 mL  of clean  extraction
          solvent  through the system.   Archive  both
          the  concentrated solvent  and the  solvent
          in the catch  flask  for contamination check
          if necessary.   Between  samples, three 2-3
          mL  aliquots  of toluene  should  be rinsed
          down  the  feed  tube  into a waste  beaker.

11.4.2    Attach  the round  bottom flask  containing
          the sample extract  to  the rotary  evapora-
          tor.   Slowly  apply  vacuum  to the  system,
          and begin rotating  the sample flask.

11.4.3    Lower  the  flask  into  the water bath and
          adjust   the  speed   of  rotation  and  the
          temperature as  required to  complete the
          concentration   in 15-20 minutes.   At the
          proper  rate of concentration, the  flow of
          solvent  into  the receiving  flask  will be
          steady,  but no bumping  or visible  boiling
          of the  extract will occur.   NOTE:   If the
          rate  of  concentration  is too  fast,  analyte
          loss  may  occur.

11.4.4    When  the  liquid in  the concentration  flask
          has  reached  an apparent volume of  2 mL,
          remove  the flask from the  water bath and
          stop  the rotation.   Slowly and  carefully,
          admit air into the  system.  Be  sure not to
          open  the valve so quickly that  the sample
          is blown  out  of the flask.  Rinse  the  feed
          tube  with approximately 2 mL  of  hexane.

11.4.5    Transfer  the  extract  to a vial  using  three
          2-3  mL  rinses of  hexane.    Proceed  with
          micro-concentration and solvent  exchange.

  11.5    Micro-concentration and solvent  exchange

11.5.1    Toluene  extracts  to be subjected to GPC or
          HPLC  cleanup are exchanged into methylene
          chloride.   Extracts that are  to  be cleaned
          up  using  silica  gel,   alumina,   and/or AX-
          21/Celite are  exchanged into  hexane.

11.5.2    Transfer   the  vial   containing   the  sample
          extract  to a  nitrogen evaporation  device.
          Adjust  the  flow  of nitrogen  so  that the
          surface   of  the  solvent is  just   visibly
          disturbed.  NOTE:   A   large vortex, in the
          solvent  may cause analyte  loss.

11.5.3    Lower the vial into a  45 3C water  bath and
          continue  concentrating.

11.5.4    When  the volume of  the liquid  is  aoproxi-
          mately  100 uL, add 2-3 mL of  the  desired
          solvent  (methylene  chloride or  hexane) and
24
02-1

-------
11.5.5
11.5.6
11.5.7
11.5.8
    12
  12.1
12.1.2
12.1.3
continue  concentration  to  approximately
100 uL.   Repeat  the addition  of  solvent
and concentrate once more.

If the extract  is  to  be  cleaned up by GPC
or HPLC, adjust  the  volume  of  the extract
to  5.0   ml   with   methylene   chloride.
Proceed with GPC cleanup (Section 12.2).

If  the extract  is  to  be  cleaned  up  by
column  chromatography   (alumina,   silica
gel, AX-21/Celite), bring the final volume
to  1.0  ml   with  hexane.    Proceed  with
column cleanups (Sections 12.3-12.5).

For   extracts   to  be   concentrated  for
injection into  the GCHS --  add  10  uL  of
nonane to the vial.  Evaporate the solvent
to the level of the nonane.   Evaporate the
hexane  in  the  vial  to  the  level  of the
nonane.

Seal  the  vial   and  label with  the sample
number.  Store  in the dark  at room temper-
ature until  ready for GCMS  analysis.

EXTRACT CLEANUP

Cleanup   may   not   be   necessary   for
relatively  clean  samples  (e.g.,  treated
effluents,   groundwater,  drinking  water).
If  particular   circumstances  require the
use of a cleanup procedure,  the analyst
may use any or  all of the procedures  below
or   any  other   appropriate   procedure.
Before  using   a  cleanup  procedure,  the
analyst must demonstrate that the  require-
ments  of Section  8.2  can be  met using the
cleanup procedure.

Gel   permeation  chromatography   (Section
12.2)  removes  many high molecular weight
interferences   that   cause   GC   column
performance  to  degrade.    It may  be used
for all soil and sediment extracts and may
be  used  for   water   extracts   that  are
expected to  contain  high molecular weight
organic   compounds    (e.g.,    polymeric
materials,  humic acids).

Acid,   neutral,  and basic silica  gel, and
alumina (Sections  12.3 and  12.4)  are used
to     remove    nonpolar     and      polar
interferences.
         AX-21/Celite  (Section  12.5)   is
         remove nonpolar interferences.
                                   used   to
  12.1.4   HPLC  (Section  12.6)   is.  used  to  provide
           specificity .  for   the   2,3,7,8-substituted
           and other PCDO and PCOF isomers.

    12.2   Gel permeation chromatography (GPC)

  12.2.1   Column packing

12.2.1.1   Place 70-75 g of  SX-3 Bio-beads  in a 400-
           500 ml beaker.

12.2.1.2   Cover  the  beads  with  methylene  chloride
           and  allow  to  swell  overnight  (12  hours
           minimum).

12.2.1.3   Transfer  the  swelled  beads  to  the  column
           and pump  solvent  through the column,  from
           bottom  to  top,  at 4.5-5.5 mL/min  prior  to
           connecting  the  column  to  the detector.

12.2.1.4   After purging the column with  solvent  for
           1-2 hours,  adjust  the  column head pressure
           to  7-10 psig  and purge  for  4-5  hours  to
           remove  air.   Maintain  a  head  pressure  of
           7-10  psig.    Connect   the  column  to  the
           detector.

  12.2.2   Column calibration

12.2.2.1   Load  5  mL  of  the  calibration  solution
           (Section 6.4)  into  the  sample  loop.

12.2.2.2    Inject  the  calibration solution and record
           the signal  from the detector.   The elution
           pattern  will   be  corn  oil,   bis(2-ethyl
           hexyl)     phthalate,     pentachlorophenol,
           perylene, and sulfur.

12.2.2.3   Set  the "dump  time"  to allow  >85 percent
           removal  of the  corn  oil  and   '85  percent
           collection  of  the phthalate.

12.2.2.4   Set the "collect  time"  to the  peak minimum
           between perylene  and  sulfur.

12.2.2.5   Verify  the calibration with   the  calibra-
           tion  solution  after  every  20  extracts.
           Calibration is  verified if  the  recovery of
           the  pentachlorophenol  is  greater  than  85
           percent.   If  calibration is not verified,
           the system  shall  be recalibrated using  the
           calibration solution,  and the  previous  20
           samples  shall be re-extracted  and cleaned
           up using the  calibrated GPC system.

  12.2.3   Extract  cleanup  --  GPC requires  that  the
           column  not  be  overloaded.    The  column
           specified  in  this  method  is   designed  to
           handle   a   maximum   of   0.5    g   of   high
           molecular   weight  material   in   a   5   mL
                                                        0
                                                  or,-
                                                                                                  25

-------
           extract.    If  the  extract  is  known  or
           expected  to contain more  than  0.5 g,  the
           extract is  split  into  a liquors  for GPC  and
           the  aliquots  are combined  after elution
           from  the  column.   The residue content of
           the     extract     may     be     obtained
           gravimetricaUy  by  evaporating  the solvent
           from a 50 uL aliquot.

12.2.3.1   Filter  the  extract or  load  through  the
           filter  holder  to  remove   particulates.
           Load the 5.0 mL  extract onto  the column.

12.2.3.2   Elute  the  extract  using  the  calibration
           data   determined   in  Section   12.2.2.
           Collect the eluate  in a  clean 400-500 ml
           beaker.

12.2.3.3   Rinse  the  sample loading  tube thoroughly
           with  methylene  chloride  between  extracts
           to prepare  for the next sample.

12.2.3.4   If   a  particularly  dirty   extract  is
           encountered,  a 5.0  ml methylene  chloride
           blank  shall be  run  through  the system to
           check for carry-over.

12.2.3.5   Concentrate the  eluate per Section 11.2.1,
           11.2.2, and 11.3.1  or 11.3.2  for further
           cleanup or  for injection into the GCHS.

    12.3   Silica gel  cleanup

  12.3.1   Place  a  glass wool  plug  in  a  15 mm  i.d.
           chromatography column.  Pack  the  column in
           the following  order (bottom to top):   1 g
           silica gel  (Section  6.5.1.1), four g  basic
           silica gel  (Section 6.5.1.3),  1  g  silica
           gel,   8   g  acid   silica  gel   (Section
           6.5.1.2), 2 g  silica gel.   Tap the  column
           to settle the  adsorbents.

  12.3.2   Pre-rinse  the column   with  50-100  mL  of
           hexane.     Close  the   stopcock  when  the
           hexane  is  within   1   mm  of   the   sodium
           sulfate.   Discard the eluate.   Check  the
           column for  channeling.   If channeling is
           present,  discard  the  column  and prepare
           another.

  12.3.3   Apply  the  concentrated   extract  to  the
           column.     Open   the  stopcock  until   the
           extract  is  within  1   mm  of  the  sodium
           sulfate.

  12.3.4   Rinse  the   receiver   twice   with  1  mL
           portions of hexane and apply  separately to
           the  column.   Elute the  PCDDs/PCDFs with
           100 mL hexane and collect  the eluate.
12.3.5   Concentrate the eluate per Section 11.4 or
         11.5 for  further  cleanup or for  injection
         into the HPLC or GCMS.

12.3.6   For  extracts  of samples  known to contain
         large   quantities    of    other   organic
         compounds  (such as  paper mill effluents)
         it  may   be   advisable  to  increase  the
         capacity  of  the sflica  gel  column.    This
         may  be   accomplished  by  increasing  the
         strengths  of  the  acid  and basic   silica
         gels.     The   acid  silica  gel  (Section
         6.5.1.2)  may  be increased  in  strength to
         as  much  as 44% w/w (7.9  g su If uric  acid
         added  to  10  g  silica  gel).    The  basic
         silica   gel   (Section   6.5.1.3)   may  be
         increased  in  strength to as  much  as 33%
         w/w  (50  mL IN NaOH  added to 100 g  silica
         gel).   NOTE:   The  use   of  stronger  acid
         silica gel  (44% w/w) may lead to charring
         of  organic  compounds  in  some   extracts.
         The  charred material  may  retain  some of
         the analytes  and  lead to lower recoveries
         of  PCOOs/PCDFs.   Increasing the  strengths
         of  the acid  and basic silica gel may  also
         require  different  volumes  of  hexane  than
         those   specified   above,   to   elute  the
         analytes  off  the column.   Therefore, the
         performance   of  the  method  after   such
         modifications  must  be   verified  by  the
         procedures in Section 8.2.

  12.4   Alumina cleanup

12.4.1   Place  a  glass  wool  plug in a  15 mm  i.d.
         chromatography column.

12.4.2   If  using  acid alumina,  pack the column by
         adding 6  g acid alumina  (Section  6.5.2.1).
         If  using  basic alumina,  substitute  6  g
         basic  alumina  (Section 6.5.2.2).   Tap the
         column to settle the adsorbents.

12.4.3   Pre-rinse  the  column  with  50-100  mL of
         hexane.    Close  the  stopcock   when  the
         hexane is within 1 mm of  the alumina.

12.4.4   Discard the eluate.   Check the column for
         channeling.    If   channeling  is  present,
         discard the column and prepare another.

12.4.5   Apply  the  concentrated extract  to  the
         column.    Open  the  stopcock  until  the
         extract is within 1 mm of the alumina.

12.4.6   Rinse  the  receiver  twice   with   1  mL
         portions  of hexane and apply separately to
         the   column.      Elute   the   interfering
         compounds  with  100 mL hexane  and discard
         the eluate.
 26
                                                           02

-------
  12.A.7   The choice of eluting solvents will.depend
           on the  choice  of alumina  (acid  or  basic)
           made in Section 12.4.2.

12.4.7.1   If using acid alumina,  elute the PCDDs and
           PCDFs from the column with 20 ml methylene
           chloride:hexane (20:80 v/v).   Collect the
           eluate.

12.4.7.2   If using  basic  alumina,  elute  the PCDOs
           and  PCDFs  from  the column  with  20  mL
           methylene  chloriderhexane   (50:50   v/v).
           Collect the eluate.

  12.4.8   Concentrate the eluate per Section 11.4 or
           11.5 for  further  cleanup  or for injection
           into the HPLC or GCHS.

    12.5   AX-21/Celite

  12.5.1   Cut  both   ends  from  a  10 ml  disposable
           serological  pipet  to  produce  a  10  cm
           column.   Fire  polish both  ends  and flare
           both ends if desired.  Insert a glass wool
           plug at one end, then pack the column with
           1 g of  the activated AX-21/Celite to form
           a 2 cm  long adsorbent bed.  Insert a glass
           wool plug on top of  the bed to  hold the
           adsorbent in place.

  12.5.2   Pre-rinse  the  column  with  five  mL  of
           toluene   followed  by   2  ml   methylene
           chloride:methanol:toluene  (15:4:1 v/v),  1
           ml  methylene   chloride:cyclohexane  (1:1
           v/v),  and five  mL  hexane.   If  the flow
           rate of  eluate  exceeds   0.5  mL  per min,
           discard the column.

  12.5.3   When the  solvent is within  1  mm  of the
           column  packing,  apply the sample extract
           to the column.  Rinse the sample container
           twice  with 1  mL portions of  hexane and
           apply separately  to the  column.  Apply 2
           mL of hexane to complete the transfer.

  12.5.4   Elute the interfering compounds  with 2 mL
           of    hexane,     2   mL     of    methylene
           chloride:cyclohexane (1:1  v/v),  and  2 mL
           of   methylene   chloride:methanol:toluene
           (15:4:1 v/v).  Discard the eluate.

  12.5.5   Invert  the column and  elute the PCDDs and
           PCDFs with  20 mL of toluene.   If  carbon
           particles  are   present   in  the  eluate,
           filter through glass fiber filter paper.—

  12.5.6   Concentrate the eluate per Section 11.4 or
           11.5 for  further  cleanup or for injection
           into the  HPLC or GCMS.
    12.6    HPLC (Reference 6)

  12.6.1    Column calibration

12.6.1.1    Prepare a calibration  standard  containing
            the 2,3,7,8- isomers and/or  other  isomers
            of interest  at  a concentration of.approxi-
            mately 500 pg/uL in methylene chloride.

12.6.1.2    Inject 30 uL  of the calibration  solution
            into the  HPLC  and  record the signal  from
            the detector.    Collect  the  eluant  for re-
            use.  The elution order will be  the tetra-
            through octa-isomers.

12.6.1.3    Establish the collect  time  for  the tetra-
            i seniors  and  for  the  other  isomers  of
            interest.   Following  calibration,  flush
            the    injection    system    with    copious
            quantities of  methylene  chloride,  includ-
            ing  a minimum  of  five  50-uL  injections
            while the detector is monitored,  to ensure
            that residual PCDDs and  PCDFs are removed
            from the system.

12.6.1.4    Verify the  calibration with  the  calibra-
            tion  solution   after   every  20  extracts.
            Calibration is verified if the recovery of
            the  PCDDs  and  PCDFs from the calibration
            standard  (Section  12.6.1.1)   is  75-125
            percent   compared   to   the   calibration
            (Section 12.6.1.2).  If calibration is not
            verified, the system shall be recalibrated
            using  the calibration  solution,   and the
            previous 20 samples  shall be re-extracted
            and   cleaned   up  using   the  calibrated
            system.

  12.6.2    Extract cleanup  --  HPLC  requires  that the
            column  not  be  overloaded.    The  column
            specified  in  this  method is designed to
            handle a maximum of 30 uL of extract.  If
            the extract cannot be concentrated to  less
            than 30 uL, it  is split into fractions and
            the  fractions  are combined  after elution
            from the column.

12.6.2.1    Rinse  the sides  of  the vial twice with 30
            uL  of  methylene chloride  and reduce to 30
            uL  with the blowdown apparatus.

12.6.2.2    Inject the 30 uL extract  into the HPLC.

12.6.2.3    Elute  the extract  using  the calibration
            data  determined in  12.6.1.   Collect the
            fraction(s) in  a clean 20 mL concentrator
            tube  containing  5 mL of  hexaneracetone
            (1:1 v/v).
                                                        027
    n
                                                                                                             27

-------
12.6.2.4    If  an extract containing  greater  than  100
            ng/mL  of  total  PCOD  or  PCDF is  encoun-
            tered,  a  30  uL methylene chloride  blank
            shall  be run through  the system  to  check
            for carry-over.

12.6.2.5    Concentrate the  eluate  per   Section  11.5
            for injection  into the GCHS.

      13    HRGC/HRMS  ANALYSIS

    13.1    Establish  the  operating  conditions  given
            in Section 7.1.

    13.2    Add  10 uL  of  the  internal standard  solu-
            tion  (Section 6.12) to  the sample extract
            immediately prior  to injection to minimize
            the  possibility  of  loss by  evaporation,
            adsorption, or  reaction.  If  an extract is
            to  be  reanalyzed  and  evaporation  has
            occurred,   do  not  add  more  instrument
            internal standard  solution.   Rather,  bring
            the  extract  back   to  its previous  volume
            (e.g.,  19  uL) with pure nonane only.

    13.3    Inject  1.0 uL of  the  concentrated extract
            containing the  internal standard solution,
            using  on-column   or  split less  injection.
            Start  the  GC  column   initial  isothermal
            hold   upon  injection.    Start   MS   data
            collection after  the  solvent  peak elutes.
            Stop  data  col lection.after the octachloro-
            dioxin  and furan  have  eluted.  Return  the
            column   to  the  initial  temperature  for
            analysis of the  next extract  or standard.

      14    SYSTEM AND LABORATORY  PERFORMANCE

    14.1    At  the  beginning  of  each  12-hour  shift
            during  which analyses  are performed,  GCMS
            system   performance  and  calibration  are
            verified   for  all unlabeled  and  labeled
            compounds.   For  these  tests,  analysis of
            the   CS3   calibration  verification  (VER)
            standard  (Section 6.13  and  Table 4)  and
            the   isomer  specificity  test  standards
            (Sections  6.16 and Table 5)  shall be used
            to   verify  all   performance   criteria.
            Adjustment   and/or   recalibration   (per
            Section  7)  shall   be  performed  until  all
            performance criteria are  met.  Only after
            all   performance   criteria  are   met  may
            samples,    blanks,   and   precision   and
            recovery standards be analyzed.

    14.2    MS  resolution --  A static resolving  power
            of  at  least  10,000   (10 percent  valley
            definition) must be demonstrated at appro-
            priate   masses  before   any   analysis   is
            performed.   Static  resolving  power checks
         must be performed  at  the beginning and at
         the end of each 12-hour shift according to
         procedures  in  Section 7.1.2.   Corrective
         actions must  be  implemented  whenever the
         resolving   power   does    not   meet   the
         requirement.

  14.3   Calibration verification

14.3.1   Inject   the   VER '  standard    using   the
         procedure in Section 13.

14.3.2   The m/z abundance ratios for all PCDDs and
         PCDFs shall be within the limits  in  Table
         3A; otherwise, the mass spectrometer  shall
         be adjusted until the m/z  abundance ratios
         fall within the  limits  specified,  and the
         verification    test   (Section   14.3.1)
         repeated.   If  the adjustment  alters the
         resolution of  the mass spectrometer,  reso-
         lution  shall  be  verified  (Section 7.1.2)
         prior to repeat of the verification test.

14.3.3   The peaks  representing  each unlabeled and
         labeled compound  in  the  VER standard must
         be  present with  a  S/N  of  at  least 10;
         otherwise,  the mass  spectrometer shall be
         adjusted   and    the   verification   test
         (Section 14.3.1) repeated.

14.3.4   Compute    the    concentration   of   each
         unlabeled  compound  by  isotope   dilution
         (Section  7.5)  for  those   compounds  that
         have  labeled  analogs (Table  1).   Compute
         the concentration of the labeled compounds
         by  the internal  standard method.    These
         concentrations  are computed  based on the
         averaged  relative  response  and  averaged
         response factor  from the  calibration data
         in Section 7.

14.3.5   For each  compound,  compare the  concentra-
         tion  with  the   calibration   verification
         limit  in  Table 7.   If  all compounds meet
         the  acceptance  criteria,  calibration has
         been verified.   If,  however,  any  compound
         fails,  the  measurement  system  is  not
         performing properly for  that  compound.  In
         this  event,  prepare  a  fresh  calibration
         standard  or  correct  the  problem causing
         the  failure   and  repeat   the  resolution
         (Section  14.2)  and  verification  (Section
         14.3.1) tests, or recalibrate  (Section 7).

  14.4   Retention  times and GC resolution

14.4.1   Retention  times
 28

-------
14.4.1.1   Absolute  --The  absolute retention  times
           of   the    3C12-1,2,3,4-TCDD   and    C^-
           1.2,3,7.8.9-HxCDD  GCMS internal  standards
           shall   be  within   ±15   seconds  of   the
           retention times  obtained during  calibra-
           tion  (Section  7.2.4).

14.4.1.2   Relative  -- The  relative retention  times
           of  unlabeled and  labeled PCDDs and  PCDFs
           shall  be  within the limits given  in  Table
           2.

  14.4.2   GC  resolution

14.4.2.1   Inject  the  isomer  specificity  standards
           (Section    6.16)    on   their    respective
           columns.

14.4.2.2   The valley  height  between 2,3,7,8-TCDO  and
           the other  tetra-  dioxin isomers  at  m/z
           319.8965, and  between  2.3,7,8-TCDF  and  the
           other  tetra- furan isomers at m/z  303.9016
           shall   not   exceed  25 percent   on  their
           respective  columns (Figure 3).

  14.4.3   If  the  absolute  retention  time  of  any
           compound    is   not   within   the   limits
           specified or the 2,3,7,8- isomers  are  not
           resolved,   the   GC   is   not   performing
           properly.    In  this event,  adjust the GC
           and repeat  the verification test  (Section
           14.3.1) or  recalibrate (Section  7).

    14.5   Ongoing precision  and  accuracy

  14.5.1   Analyze  the   extract   of   the   diluted
           precision  and  recovery  standard   (PAR)
           (Section   10.3.4   or   10.4.4)   prior  to
           analysis  of  samples  from  the same set.

  14.5.2   Compute the  concentration of each  PCDD  and
           PCDF   by   isotope  dilution   for   those
           compounds   that   have   labeled   analogs
           (Section  7.5).   Compute  the concentration
           of  each  labeled  compound by the  internal
           standard method.

  14.5.3   For each  unlabeled  and   labeled  compound,
           compare the concentration with  the  limits
           for ongoing accuracy  in  Table  7.   If  all
           compounds  meet  the  acceptance  criteria,
           system  performance   is   acceptable   and
           analysis   of   blanks   and   samples   may
           proceed.     If,   however,  any   individual
           concentration  falls outside  of  the  range
           given,     the     extraction/concentration
           processes are not  being performed  properly
           for that  compound.   In this event,  correct
           the problem,  re-extract   the  sample  set
           (Section   10)   and   repeat   the   ongoing
         precision   and   recovery  test   (Section
         14.5).   The concentration limits  in  Table
         7  for  labeled compounds  are  based on  the
         requirement   that   the   recovery   of  each
         labeled  compound be  in  the  range of  25-
         150%.

14.5.4   Add  results which pass  the specifications
         in Section  14.5.3 to initial and  previous
         ongoing  data  for  each   compound  in each
         matrix.     Update   QC  charts   to   form a
         graphic    representation   of   continued
         laboratory  performance.   Develop  a  state-
         ment of  laboratory accuracy for each PCDD
         and  PCDF  in each matrix  type by calculat-
         ing  the  average  percent  recovery (R)  and
         the  standard  deviation  of percent  recovery
         (SR).  Express the accuracy as a  recovery
         interval  from R  -  2S   to R  +  2SR.    For
         example,  if  R  =  95%  and SR  =  5%,  the
         accuracy  is 85-105%.
    15   QUALITATIVE DETERMINATION

         For  a  gas   chromatographic   peak  to   be
         identified  as  a PCDD  or  PCDF  (either  a
         unlabeled or  a  labeled compound),  it must
         meet all of the criteria  in Sections 15.1-
         15.4.

  15.1   The signals  for the two exact m/z's being
         monitored  (Table 3) must  be present,  and
         must  maximize  within  + 2  seconds  of  one
         another.

  15.2   The signal-to-noise  ratio  (S/N)  of each of
         the two exact  m/z's must  be greater than
         or equal  to  2.5 for a  sample  extract,  and
         greater than  or equal  to  10 for  a calibra-
         tion  standard  (see  Sections   7.2.3  and
         14.3.3).

  15.3   The ratio of  the  integrated  ion  currents
         of both  the  exact m/z's monitored must be
         within the limits  in Table  3A.

  15.4   The relative   retention  time  of  the peaks
         representing     a    unlabeled    2,3,7,8-
         substituted  PCDD  or  PCDF  must  be  within
         the   limits   given  in  Table  2.     The
         retention  time of peaks  representing non-
         2,3,7,8-substituted  PCDDs  or  PCDFs must be
         within    the     retention   time   windows
         established in Section 7.3.

  15.5   Confirmatory     analysis     --     Isomer
         specificity for all  of the 2,3,7,8-substi-
         tuted   analytes  cannot  be   attained   by
         analysis  on  the DB-5  (or  equivalent)  GC
         column alone.   The  lack of specificity is
                                                                                                             29

-------
          of  greatest  concern   for   the   un labeled
          2,3,7,8-TCDF.    Therefore,  any  sample  in
          which   2.3,7,8-TCDF   is   identified   by
          analysis  on  a  OB-5  (or  equivalent)  GC
          column  must  have  a confirmatory  analysis
          performed on a DB-225,  SP-2330,  or equiva-
          lent GC  column.   The operating  conditions
          in  Section  7.1.1   may  be  adjusted  for
          analyses on  the  second GC  column, but  the
          GCHS  must  meet  the mass  resolution  and
          calibration specifications  in Section 7.

   15.6   If any  gas  chromatographic peak meets  the
          identification criteria in Sections  15.1,
          15.2, and 15.4,  but does not meet the ion
          abundance ratio  criterion  (Section  15.3),
          and  is  not  a  labeled  analog, that  sample
          must be analyzed on a second  GC  column,  as
          in Section  15.5  above.   Interferences  co-
          eluting  in  either   of  the two  m/z's  may
          cause the ion abundance ratio to fall out-
          side of  the limits in Table 3A.   If  the
          ion  abundance  ratio of  the peak fails  to
          meet the criteria on the second  GC column,
          then the peak does  not  represent a PCDD or
          PCDF.   If  the peak does meet  all  of  the
          criteria  in  Sections   15.1-15.4  on  the
          second   GC   column,  then  calculate  the
          concentration  of    that   peak   from  the
          analysis on  the  second GC  column, accord-
          ing to the procedures in Section 16.

   15.7   If   any  gas  chromatograpoic   peak   that
          represents a  labeled analog does not meet
          all  of  the   identification  criteria  in
          Sections  15.1-15.4  on  the   second   GC
          column,  then  the  results  may  not   be
          reported    for    regulatory    compliance
          purposes and  a  new aliquot of  the  sample
          must be extracted and analyzed.

     16   QUANTITATIVE DETERMINATION

   16.1   Isotope  dilution   --  By  adding  a   known
          amount  of  a  labeled  compound  to  every
          sample prior to  extraction, correction for
          recovery of  the  un labeled  compound can be
          made  because  the   unlabeled  compound  and
          its  labeled analog  exhibit  similar effects
          upon  extraction,  concentration,  and  gas
          chromatography.    Relative  response  (RR)
          values are  used  in  conjunction  with  cali-
          bration  data  described  in  Section 7.5  to
          determine concentrations directly,  so long
          as  labeled  compound  spiking   levels  are
          constant, using  the following equation:
                                                          where,  C£x  is  the concentration  at  the
                                                          unlabeled compound in the extract  and the
                                                          other  terms   are   as  defined  in  Section
                                                          7.5.2.

                                                 16.1.1    Because of a  potential  interference,  the
                                                          labeled analog of  OCDF is not added to the
                                                          sample.  Therefore, this unlabeled analyte
                                                          is quant itated  against  the  labeled  OCDD.
                                                          As   a    result,    the   concentration   of
                                                          unlabeled  OCDF   is   corrected   for   the
                                                          recovery  of   the  labeled   OCDD.     In
                                                          instances  where   OCDD  and   OCDF   behave
                                                          differently   during   sample   extraction,
                                                          concentration,  and   cleanup  procedures,
                                                          this may decrease  the  accuracy of the OCDF
                                                          results.   However, given  the low toxicity
                                                          of  this  compound  relative  to  the  other
                                                          dioxins and furans, the potential decrease
                                                          in accuracy is not considered significant.

                                                 16.1.2    Because the labeled analog of 1,2.3,7,8,9-
                                                          HxCDD  is  used  as an  internal  standard
                                                          (i.e.,  not added  before  extraction of the
                                                          sample),  it cannot be used  to  quantitate
                                                          the unlabeled  compound by  strict  isotope
                                                          dilution  procedures.     Therefore,   the
                                                          unlabeled 1,2,3,7,8,9-HxCDD is quantitated
                                                          using the average  of  the  responses of the
                                                          labeled analogs of the  other two 2,3,7,8-
                                                          substituted HxCDD's,  1,2,3,4,7,8-HxCDD and
                                                          1,2,3,6,7,8-HxCDD.    As  a   result,   the
                                                          concentration     of      the     unlabeled
                                                          1.2,3,7,8,9-HxCDD   is   corrected  for  the
                                                          average recovery of the other two HxCDD's.

                                                 16.1.3    Any peaks representing non-2,3,7,8-substi-
                                                          tuted  dioxins or  furans are quantitated
                                                          using an  average  of   the  response factors
                                                          from all  of  the  labeled  2,3,7,8-  isomers
                                                          in the same level  of  chlorination.

                                                   16.2    Internal standard --  Compute  the concen-
                                                          trations of  the   C-labeled analogs and
                                                          the    C-labeled cleanup  standard  in the
                                                          extract using the response  factors deter-
                                                          mined from calibration  data  (Section 7.6)
                                                          and the following  equation:
                                                              (ng/mL)  =
                                                                                           Cis
                  (Ajs
                                                                                      Ais)  RF
C   (ng/mi.)  =
 ex
                             (A   + A  ) C.
                               n     n    is
                                .      2
                             (A'  -•• A  ) RR
where,  C    is  the  concentration  of  the
compound  in  the   extract  and  the other
terms  are  as  defined  in Section  7.6.1.
(NOTE:   There  is  only  one  m/z  for  the
37Cl-labeled standard.)
30
                                                           u '

-------
16.3   The   concentration   of   the   unlabeled
       compound in the solid  phase  of  the sample
       is computed using the concentration of the
       compound in the extract  and  the weight of
       the solids (Section 10),  as follows:

                            (C    x  V   )
       Concentration    =     ex    ex
       in solid (ng/Kg)           w
                                  s
       where,
       V   is the extract volume in ml.
       W  is the sample weight in Kg.

16.4   The   concentration   of   the   unlabeled
       compound  in  the  aqueous  phase  of  the
       sample is computed using the concentration
       of  the compound  in the  extract  and the
       volume of water  extracted  (Section 10.3),
       as follows:
(C
                                   Vex>
                                  Vs
         Concentration
         in aqueous phase
         (P9/L)

         where,
         V   is the extract volume in ml.
         V  is the sample volume in liters.

  16.5   If  the  SICP  areas  at  the  quantitation
         m/z's for any compound exceed the calibra-
         tion range of the system, a smaller sample
         aliquot is extracted.

16.5.1   For aqueous samples containing one percent
         solids  or  less,  dilute  100 ml,  10  ml,
         etc.,  of  sample to  1  liter  with  reagent
         water and extract per Section 11.

16.5.2   For  samples containing  greater  than  one
         percent  solids,  extract   an  amount   of
         sample equal to  1/10, '1/100,  etc.,  of  the
         amount  determined   in   Section  10.1.3.
         Extract per Section 10.4.

16.5.3   If  a  smaller   sample  size  will  not  be
         representative   of   the   entire  sample,
         dilute the  sample  extract by a  factor of
         10,   adjust   the  concentration   of   the
         instrument  internal  standard  to  100 pg/uL
         in the extract,  and  analyze an  aliquot of
         this  diluted  extract   by  the  internal
         standard method.

  16.6   Results are  reported to  three significant
         figures  for  the  unlabeled  and  labeled
         isomers  found  in  all  standards,  blanks,
         and  samples.    For  aqueous  samples,  the
         units  are  pg/L; for  samples  containing
         greater  than one  percent  solids  (soils,
         sediments,  filter   cake,   compost),   the
         units are ng/Kg based on the dry weight of
         the sample.

16.6.1   Results   for   samples  which   have  been
         diluted  are  reported at the  least dilute
         level at  which  the  areas  at the quantita-
         tion  m/z's  are  within  the  calibration
         range (Section 16.5).

16.6.2   For  unlabeled  compounds having  a  labeled
         analog,  results are  reported  at the  least
         dilute  level   at  which  the  area  at the
         quantitation m/z  is within  the calibration
         range   (Section   16.5)   and  the   labeled
         compound  recovery  is  within  the normal
         range for the method (Section 17.4).

16.6.3   Additionally,  the total  concentrations of
         all  isomers  in  an  individual   level  of
         chlorination   (i.e.,.  total  TCDD,   total
         PeCDD,  etc.) are  reported  to three signi-
         ficant  figures  in units of pg/L,  for both
         dioxins  and  furans.   The  total  or  ng/Kg
         concentration  in  each  level  of  chlorina-
         tion  is the sum  of  the concentrations of
         all  isomers   identified  in  that  level,
         including    any   non-2,3,7,8-substituted
         isomers.

    17   ANALYSIS OF COMPLEX SAMPLES

  17.1   Some samples may  contain  high levels (>10
         ng/L;  >1000 ng/Kg)  of  the  compounds  of
         interest,  interfering  compounds,  and/or
         polymeric  materials.    Some extracts will
         not  concentrate  to  10  uL  (Section  11);
         others  may overload the GC column and/or
         mass spectrometer.

  17.2   Analyze  a smaller  aliquot .of  the sample
         (Section  16.4)  when the extract will not
         concentrate  to 20  uL  after  all  cleanup
         procedures have been exhausted.

  17.3   Recovery   of    labeled   compound   spiking
         standards  --  In  most  samples,  recoveries
         of  the  labeled compound spiking  standards
         will  be  similar  to  those  from  reagent
         water   or  from   the   alternate  matrix
         (Section  6.6).   If  recovery  is  outside of
         the   25-150%   range,   a   diluted  sample
         (Section  16.4)  shall  be analyzed.  If the
         recoveries of  the labeled compound spiking
         standards   in   the   diluted   sample  are
         outside  of the  limits (per  the  criteria
         above),   then   the   verification  standard
         (Section   14.3)   shall   be   analyzed  and
         calibration verified  (Section 14.3.4).  If
         the  calibration cannot  be  verified,  a new
                                                      031
                                                                                                        31

-------
          calibration  must  be   performed  and   the
          original  sample extract  reanalyzed.    If
          the   calibration  is   verified  and   the
          diluted  sample does  not meet  the  limits
          for  labeled compound   recovery,  then  the
          method does  not apply  to the sample being
          analyzed   and   the  result   may   not   be
          reported    for    regulatory   compliance
          purposes.

     18   METHOD PERFORMANCE

          The  performance   specifications  in   this
          method are based on  the  analyses of  more
          than  400  samples,  representing  matrices
          from at  least  five industrial  categories.
          These   specifications  will   be  updated
          periodically  as more   data  are  received,
          and each time  the  procedures  in the method
          are revised.

       REFERENCES

      1   Tondeur,  Yves,  "Method  8290:  Analytical
          Procedures   and   Quality  Assurance   for
          Multimedia   Analysis    of   Polychlorinated
          Oibenzo-p-dioxins   and  Dibenzofurans   by
          High-Resolution  Gas   Chromatography/High-
          Resolution Mass  Spectrometry",  USEPA EMSL,
          Las Vegas, Nevada, June 1987.

      2   "Measurement   of   2,3,7,8-Tetrachlorinated
          Dibenzo-p-dioxin (TCOO) and 2,3,7,8-Tetra-
          chlorinated  Dibenzofuran (TCDF)  in Pulp,
          Sludges, Process Samples and Waste-waters
          from  Pulp  and  Paper  Mills", Wright State
          University, Dayton, OH  45435, June 1988.

      3   "HCASI Procedures  for  the  Preparation and
          Isomer Specific Analysis  of Pulp  and Paper
          Industry   Samples   for   2,3,7,8-TCDD   and
          2,3,7,8-  TCDF". National  Council  of  the
          Paper Industry for Air  and  Stream Improve-
          ment,  260  Madison Avenue,  New  York,  NY
          10016,  Technical  Bulletin  No.  551,   Pre-
          release Copy,  July 1988.

      4   "Analytical    Procedures    and   Quality
          Assurance  Plan  for  the  Determination  of
          PCDD/PCDF  in  Fish",  USEPA.  Environmental
          Research    Laboratory,     6201   Congdon
          Boulevard, Duluth, MN 55804, April 1988.

      5   Tondeur, Yves,  "Proposed  GC/MS  Methodology
          for  the Analysis  of  PCDDs and  PCDFs  in
          Special   Analytical   Services   Samples",
          Triangle    Laboratories,     Inc.,   801-10
          Capitola  Or,  Research  Triangle  Park,  NC
          27713,  January  1988;  updated  by personal
          communication  September 1988.
     6   Lamparski,   L.L.,   and   Nestrick,   T.J.,
         "Determination of  Tetra-,  Hexa-,  Hepta-,
         and Octachlorodibenzo-p-dioxin  Isomers  in
         Particulate Samples at  Parts  per Trillion
         Levels",  Analytical  Chemistry.   52:  2045-
         2054,  1980.

     7   Lamparski,   L.L.,   and   Nestrick,   T.J.,
         "Novel    Extraction    Device    for    the
         Determination  of  Chlorinated  Dibenzo-p-
         dioxins (PCDDs)  and  Dibenzofurans (PCDFs)
         in     Matrices     Containing     Water",
         Chemosphere. 19:27-31, 1989.

     8   Patterson,  D.G.,  et.   al.   "Control   of
         Interferences  in  the  Analysis  of  Human
         Adipose    Tissue    for    2,3,7,8-Tetra-
         chlorodibenzo-p-dioxin".     Environmental
         Toxicological Chemistry. 5: 355-360,  1986.

     9   Stanley,  John S.,  and  Sack,  Thomas  M.,
         "Protocol  for  the Analysis  of  2,3,7,8-
         Tetrachlorodibenzo-p-dioxin    by    High-
         Resolution     Gas     Chromatography/High-
         Resolution Mass Spectrometry", USEPA EMSL,
         Las Vegas, Nevada 89114, EPA 600/4-86-004,
         January 1986.

    10   "Working  with  Carcinogens",  DHEW,   PHS,
         CDC,  NIOSH.  Publication  77-206,  August
         1977.

    11   "OSHA Safety and Health Standards, General
         Industry" OSHA 2206,  29 CFR 1910, January
         1976.

    12   "Safety      in      Academic     Chemistry
         Laboratories", ACS Committee on Chemical
         Safety, 1979.

    13   "Standard  Methods  for  the Examination of
         Water  and Wastewater",  16th edition  and
         later  revisions,  American  Public  Health
         Association,    1015     15th   St,    N.W.,
         Washington,  DC  20005,  46:  Section  108
         (Safety), 1985.

    14   "Method 613  -- 2,3,7,8-Tetrachlorodibenzo-
         p-dioxin".  40  CFR  136   (49  FR  43234),
         October 26,  1984, Section 4.1.

    15   Provost,    L.P.,    and    Elder,    R.S.,
         "Interpretation of Percent Recovery Data",
         American Laboratory. 15: 56-83,  1983.

    16   "Handbook of Analytical Quality Control in
         Water  and Wastewater Laboratories",  USEPA
         EMSL,   Cincinnati,  OH 45268, EPA-600/4-79-
         019, March 1979.
32
03


-------
17   "Standard  Practice  for  Sampling  Water",
     ASTH Annual Book of Standards, ASTH, 1916
     Race Street, Philadelphia, PA  19103-1187,
     1980.

18   "Methods   330.4   and  330.5   for  Total
     Residual Chlorine",  USEPA,  EMSL,  Cincin-
     nati, OH  45268,  EPA  600/4-70-020,  March
     1979.

19   Barnes,  Donald G.,  Kutz,  Frederick U., and
     Bottimore,  David  P.,  "Update of  Toxicity
     Equivalency Factors (TEFs) for Estimating
     Risks   Associated   with  Exposures   to
     Mixtures of  Chlorinated  Dibenzo-p-Dioxins
     and   Dibenzofurans   (CDDs/CDFs)",   Risk
     Assessment   Forum,  USEPA,  Washington,  DC
     20460, February 1989.
                                                  033
33

-------
                                                   Table 1
         POLYCHLORINATED DIBENZODIOXINS AND  FURANS DETERMINED  BY  ISOTOPE DILUTION AND  INTERNAL STANDARD
               HIGH  RESOLUTION GAS CHROMATOGRAPHY (HRGO/HIGH  RESOLUTION MASS SPECTROMETRY  (HRHS)
PCDDs/PCDFs (1)
Isomer/Congener CAS Registry
2,3,7,8-TCDD 1746-01-6
Total -TCDD 41903-57-5
2,3,7.8-TCDF 51207-31-9
Total -TCDF 55722-27-5
1,2,3,7,8-PeCDD 40321-76-4
Total -PeCDD 36088-22-9
1,2,3,7,8-PeCDF 57117-41-6
2,3,4,7,8-PeCDF 57117-31-4
Total-PeCDF 30402-15-4
1,2,3.4,7.8-HxCDD 39227-28-6
1,2,3,6,7,8-HxCDD 57653-85-7
1,2.3,7.8,9-HxCOD 19408-74-3
Total-HxCDD 34465-4608
1,2.3.4.7,8-HxCDF 70648-26-9
1,2,3,6,7.8-HxCDF 57117-44-9
1,2,3,7,8,9-HxCDF 72918-21-9
2,3,4,6,7,8-HxCDF 60851-34-5
Total-HxCDF 55684-94-1
1,2,3,4,6,7-,8-HpCDD 35822-46-9
Total-HpCDD 37871-00-4
1,2.3,4,6,7,8-HpCDF ' 67562-39-4
1,2,3,4,7,8,9-HpCOF 55673-89-7
Total-HpCDF 38998-75-3
OCDD 3268-87-9
OCOF 39001-02-0
(1) Polychlorinated dioxins and furans
TCDD = Tetrachlorodibenzo-p-dioxin
PeCDD = Pentachlorodibenzo-p-dioxin
HxCDD = Hexachlorodibenzo-p-dioxin
HpCDO = Heptachlorodibenzo-p-dioxin
OCDD = Octachlorodibenzo-p-dioxin
Labeled Analog
13C12-2,3,7,8-TCDD
37Cl4-2,3,7,8-TCDD
13C12-2,3,7,8-TCDF
13C12-1,2.3,7,8-PeCDD
13C12-1,2,3.7,8-PeCDF
13C12-2.3,4,7,8-PeCOF
13C,,-1,2,3,4,7,8-HxCDD
Cl2-1,2,3,6,7.8-HxCDD
13C12-1.2,3,7,8,9-HxCDD<2>
13C12-1,2,3,4,7,8-HxCDF
13C12-1,2,3.6,7.8-HxCDF
13C,--1. 2,3, 7,8,9- HxCDF
C,,-2,3,4,6,7,8-HxCDF
13C12-1,2,3,4,6,7,8-HpCDD
13C12-1, 2,3,4. 6,7,8-HpCOF
13C12-1,2,3,4,7.8,9-HpCDF
13C12-OCDD
none
TCDF = Tetrachlorodibenzofuran
PeCDF = Pentachlorodibenzofuran
HxCDF = Hexachlorodibenzofuran
HpCDF = Heptachlorodibenzofuran
OCDF = Octachlorodibenzofuran
CAS Registry
76523-40-5
85508-50-5
89059-46-1
109719-79-1
109719-77-9
116843-02-8
109719-80-4
109719-81-5
109719-82-6
114423-98-2
116843-03-9
116843-04-0
116843-05-1
109719-83-7
109719-84-8
109719-94-0
114423-97-1

(2)   Labeled  analog  is used as an internal  standard  and  therefore  is  not  used for quant i tat ion  of  the  native
     compound.
 34
                                                           031

-------
                                                    Table  2
                             RETENTION TIMES AND MINIMUM LEVELS FOR PCDDs AND PCDFs
                                                                                          Minimum Level (1)
Compound
Compounds using C..--1,2,3,4-TCDD
Native Compounds
2,3,7,8-TCDF
2,3,7,8-TCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,7,8-PeCDD
Labeled Compounds
;3C12-2,3.7,8-TCDF
1JC12-1.2.3,4-TCDD
^3C12-2,3,7,8-TCDD
37Cl4-2,3,7,8-TCOD
13C12-1,2,3,7,8-PeCDF
13C12-2,3,4,7,8-PeCDF
13C12-1,2,3,7,8-PeCDD
Compounds using C12-1,2,3,7,8,9-
Native Compounds
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,6,7,8-HpCDD
1,2,3,4,7,8,9-HpCDF
OCDD
OCDF
Labeled Compounds
13C12-1,2,3,4,7,8-HxCDF
13C,_-1,2,3.6,7,8-HxCDF
C.2-1,2,3,7,8,9-HxCDF
13C12-2,3,4,6,7,8-HxCDF
13C12-1,2,3,4,7,8-HxCDD
13C12- 1,2,3,6,7,8-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,4,6,7,a-HpCDF
13C.-- 1,2,3,4,6,7,8-HpCDD
C,--1,2,3,4,7,8,9-HpCDF
C12-OCDD
Retention
Time
Reference
as internal standard
J3C12-2,3.7.8-TCDF
X2-2,3,7,8-TCOD
C12-1,2,3,7,8-PeCDF
13C12-2,3,4,7,8-PeCDF
C12-1,2,3,7,8-PeCDD
13C12-1,2,3,4-TCOD
13C.,-1,2.3.4-TCDD
13 12
I3C12-1.2,3.4-TCOO
^•I^S^TCDD
"C12-1,2,3,4-TCOO
13C12-1,2,3,4-TCDD
HxCDD as internal standard
13C.,-1,2,3,4,7,8-HxCDF
C12-1.2,3,6,7.8-HxCDF
13C12-1,2,3,7,8,9'HxCDF
13C,_-2,3,4,6,7,8-HxCDF
C12-1,2.3,4,7.8-HxCDD
13C12-1,2,3,6,7,8-HxCDD
13C12-1,2,3,6.7,8-HxCDD
13C12-1,2,3,4,6.7,8-HpCDF
13C.,- 1,2,3,4,6,7,8-HpCDD
C12-1,2.3,4,7,8,9-HpCDF
13C..--OCOD
13 12
C12-OCOD
13C.2-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
Relative Water Solid
Retention pg/L ng/kg
Time ppq ppt
0.993 -
0.993 -
0.918 -
0.999 -
0.987 -
0.931 -
1.000 •
0.993 -
1.002 -
1.091 -
1.123 -
1.134 -
0.986 -
0.973 -
0.937 -
0.999 -
0.999 -
0.992 -
0.986 -
0.930 -
0.986 -
0.896 -
0.996 -
0.995 -
0.947 -
0.940 -
0.993 -
0.971 -
0.974 -
0.975 -
1.000 -
0.953 -
1.023 -
1.024 -
1.050 -
1.009 10 1
1.009 10 1
1.076 50 5
1.001 50 5
1.016 50 5
0.994
1.000
1.036
1.013
1.371
1.408
1.428
1.015 50 5
1.025 50 5
1.068 50 5
1.001 50 5
1.001 50 5
1.009 50 5
1.016 50 5
1.022 50 5
1.016 50 5
1.079 50 5
1.005 100 10
1.013 100 10
0.992
1.006
1.017
1.000
1.002
1.006
1.000
1.172
1.125
1.148
1.275
Extract
pg/uL
ppb
0.5
0.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
5.0
5.0
(1)   Level  at  which  the analytical system will give acceptable SICP and calibration.
                                                           0
O r-
                                                                                                            35

-------
DESCRIPTORS,  MASSES,  M/Z TYPES,
                                              Table 3
                                        AND ELEMENTAL COMPOSITIONS OF THE CDDs AND CDFs (1)
Descriptor
Number
1












2










3











Accurate
m/z (2)
292.9825
303.9016
305.8987
315.9419
317.9389
319.8965
321 .8936

327.8847
330.9792
331 .9368
333.9339
375.8364
339.8597
341 .8567
351.9000
353.8970
354.9792

355.8546
357.8516
367.8949
369.8919
409.7974
373.8208
375.8178
383.8639
385.8610
389.8157
391.8127
392.9760

401.8559
403.8529
430.9729
445.7555
m/z
Type
Lock
M
M+2
M
M+2
M
M+2

M
QC
M
M+2
M+2
M+2
M+4
M+2
M+4
Lock

M+2
M+4
M+2
M+4
M+2
M+2
M+4
M
M+2
M+2
M+4
Lock

M+2
M+4
QC
M+4
Elemental Composition
C7F11
C12 H4 35C14 0
C12 H4 35cl337(:l °
13C12 H4 35C14 0
13C12 H, 35C13 37Cl 0
C12 H4 35cl4 °2
C.- H. 35Cl, 37Cl 0,
12 4 3 2
C12 H4 37cl4 °2
C7F13
13C12 H, 35C14 O,
13 35 37
C12 H4 C13 Ct °2
C12 H4 35C15 37Cl 0
C,2 H3 35C14 37Cl 0
C12 H, 35C13 37C12 0
13 35 . 37
C12 H3 C14 Cl °
13 35 37
C12 H3 C13 C12 °
C0 F.,
9 13
C H 35Cl 37Cl 0
C12 H3 35C13 37C12 °2
13 35 37
C12 H3 C14 Cl °2
13 35 37
C12 H3 CL3 C12 °2
C12 H3 35cl6 37cl °
C12 H2 35cl5 3/Ct °
C12 H2 35cl4 37cl2 °
13C H 35
C12 H2 C16 °
13C12 H, 35C15 37Cl 0
C,, H2 35C15 37Cl 02
C12 H2 35cl4 37cl2 °2
C0 F -
9 15
13 35 37
C12 H2 C15 Cl °2
13 35 37
C12 H2 C14 C12 °2
C9 F13

Compound
(3)
PFK
TCDF
TCDF
TCDF(4)
TCDF(4)
TCDD
TCDD

TCDD(S)
PFK
TCDD(4)
TCDO(4)
HxCDPE
PeCDF
PeCDF
PeCDF(4)
PeCDF(4)
PFK

PeCDD
PeCDD
PeCDD (4)
PeCDD(4)
HpCDPE
HxCDF
HxCDF
HxCDF(4)
HxCDF(4)
HxCDD
HxCDD
PFK

HxCDD(4)
HxCDD(4)
PFK
OCDPE
Primary
m/z?

Yes

Yes

Yes


-

Yes


Yes

Yes



Yes

Yes


Yes

Yes

Yes



Yes



36
                                                   036

-------
                                             Table 3 (continued)
              DESCRIPTORS, MASSES, M/Z TYPES,  AND ELEMENTAL COMPOSITIONS OF  THE  CDDs AND CDFs (1)
Descriptor Accurate
Number m/z (2)





















(1)
(2)


(3)

4 407.7818
409.7789
417.8253
419.8220
423.7766
425.7737
430.9729
435.8169

437.8140
479.7165

5 441.7428
442.9728
443.7399
457.7377
459.7348

469.7779
471.7750
513.6775
From Reference 5
Nuclidic masses used:
H = 1.007825
0 = 15.994915
m/z
Type
M+2
M+4
M
M+2
M+2
M+4
Lock
M+2

M+4
M+4

M+2
Lock
M+4
M+2
M+4

M+2
M+4
M+4


C = 12.00000
35Cl = 34.968853

Elemental Composition
C-2 H 35C16 37Cl 0
C-2 H 35C15 37C12 0
13C12 H 35C17 0
13C12 H 35C16 37Cl 0
C-2 H 35Cl, 37Cl 02
C.2 H 35CU 37CU °2
C9F17
13C12 H 35C16 37Cl 02
17 75 77
C12 H C15 C12 °2
C12 H 35C17 37C12 0
75 77
c12 "ci7 a'ct o
C10 F17
_ 35_. 37_. _.
C12 C16 C12 °
C 35Cl 37Cl 0
C12 C17 Cl °2
C., 35Cl, 37Cl- 0,
12 6 22
13C12 35C17 37Cl 0,
13_ 35.. 37.. _
C12 C16 C12 °2
C 35Cl 37Cl 0
C12 C18 C12 °


13C = 13.003355
37Cl = 36.965903
Compound
(3)
HpCDF
HppDF
HpCDF(4)
HpCDF(4)
HpCDD
HpCDD
PFK
HpCDD (4)

HpCDD(4)
NCDPE

OCDF
PFK
OCDF
OCDD
OCDD

OCDD(4)
OCDD(4)
DCDPE


F = 18.9984

Primary
m/z?
Yes

Yes

Yes


Yes




Yes


Yes


Yes






Compound abbreviations:
Chlorinated dibenzo-p-
dioxins
TCDD = Tetrachlorodibenzo-p-dioxin
PeCDD = Pentachlorodibenzo-p-dioxin
HxCDD = Hexachlorodibenzo-p-dioxin
HpCDD = Heptachlorodibenzo-p-dioxin
OCDD = Octachlorodibenzo-p-dioxin
Chlorinated diphenyl ethers
HxCDPE =
HpCDPE =
OCDPE
NCDPE =
DCDPE
Hexachlorodiphenyl ether
Heptachlorodiphenyl ether
Octachlorodiphenyl ether
Nonachlorodiphenyl ether
Decachlorodiphenyl ether





     Chlorinated dibenzofurans
        TCDF    =    Tetrachlorodibenzofuran
        PeCDF   =    Pentachlorodibenzofuran
        HxCDF   =    Hexachlorodibenzofuran
        HpCDF   =    Heptachlorodibenzofuran
Lock mass and OC  compound
   PFK     =    Perfluorokerosene
(4)   Labeled compound

(5)   There  is only one m/z for   Cl,-2,3,7,8-TCDD (cleanup standard).
                                                        037
                                                                                                          37

-------
                       Table 3A
   THEORETICAL ION ABUNDANCE RATIOS AND CONTROL LIMITS
NO. Of
Chlorine
Atoms
4 <2>
5
6
6 (3)
7
7 (4)
a
m/z's
Forming
Ratio
M/M+2
M+2/M+4
M+2/M+4
M/M+2
M+2/M+4
M/M+2
M+2/M+4
Theoretical
Ratio
0.77
1.55
1.24
0.51
1.05
0.44
0.89
Control
Lower
0.65
1.32
1.05
0.43
0.88
0.37
0.76
Limitsd)
Upper
0.89
1.78
1.43
0.59
1.20
0.51
1.02
(1)   Represent + 15% windows around the theoretical ion
     abundance ratios.
(2)   Does not apply to    Cl.-2,3,7,8-TCDD (cleanup
     standard).
(3)   Used for 13C-HxCDF only.
(4)   Used for 13C-HpCDF only.
 38
                                                      038

-------
                                                    Table 4
                 CONCENTRATIONS OF SOLUTIONS CONTAINING LABELED AND UNLABELED PCDDS AND PCDFS -•
                                          STOCK AND SPIKING SOLUTIONS
           Compound
                                  Labeled
                                  Compound
                                   Stock
                                Solution (1)
                                  (ng/mL)
              Labeled
              Compound
              Spiking
            Solution (2)
              (ng/mL)
    PAR
   Stock
Solution (3)
   (ng/mL)
   Cleanup
  Standard
   Spiking
Solution (4)
   (ng/mL)
  Internal
  Standard
  Spiking
Solution (5)
  (ng/mL)
Native CDDs and CDFs
             2,3,7,8-TCDD
             2,3,7,8-TCDF
          1,2,3,7,8-PeCDD
          1,2,3,7,8-PeCDF
          2,3,4,7,8-PeCDF
        1,2,3,4,7,8-HxCDD
        1,2,3,6,7,3-HxCDD
        1,2,3,7,8.9-HxCDD
        1,2,3.4,7,8-HxCDF
        1.2,3,6,7,8-HxCDF
        1,2,3.7,8,9-HxCDF
        2,3,4,6,7.8-HxCDF
      1,2.3,4,6,7,8-HpCDD
      1,2,3,4,6,7,8-HpCDF
      1,2,3,4,7,8,9-HpCDF
                     OCDD
                     OCDF
Labeled CDDs and CDFs
       13C12-2.3,7.8-TCDD
    13
    13
    13,
       13C12-2,3,7,8-TCDF
      C12-1,2.3,7,8-PeCDD
      C12-1.2,3,7,8-PeCDF
      'c12-2.3,4.7,8-PeCDF
    Cl2-1,2.3,4.7,8-HxCDD
    C12-1,2,3,6,7.8-HxCDD
    C12-1,2,3,4,7,8-HxCDF
    C12-1,2,3,6,7,8-HxCDF
  13C12-1,2,3,7,8,9-HxCDF
  13C12-2,3,4,6,7,8-HxCDF
13C12-1,2,3,4,6,7,8-HpCDD
13C12-1,2,3,4,6f7,8-HpCOF
13ci:)-i,2,3,4,7,8f9-HpcDF
  13,
  13
  13
  "12
               13,
                 C-2-OCDD
Cleanup Standard
       37Cl4-2,3,7,8-TCDD
Internal Standards
       13C12-1.2,3,4-TCDD
  13
    C12-1,2,3,7,8,9-HxCDD
100
100
100
100
100
100
100
100
100
100
100
100
100
100
200
                                                                      40
                                                                      40
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     200
                                                                     400
                                                                     400
                                                                                      0.8
                                                                                                     200
                                                                                                     200
(1)  Section 6.10 -  prepared in nonane and diluted  to prepare spiking solution.
(2)  Section 10.3.2  -  prepared from stock solution  daily.
(3)  Precision and Recovery (PAR)  standard. Section 6.14  • prepared in nonane and diluted to prepare spiking
     solution in Section 10.3.4.
(4)  Section 6.11  -  prepared in nonane.
(5)  Section 6.12  -  prepared in nonane.
                                                       039
                                                                                                             39

-------
                                           Table 4 (continued)
                CONCENTRATIONS OF SOLUTIONS CONTAINING LABELED AND UNLABELED PCDDS AND PCDFS
                                   CALIBRATION AND  VERIFICATION SOLUTIONS
Compound
Native CDDs and CDFs
2,3,7.8-TCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDD
1,2,3,7,3-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDD
1,2,3,4.6,7.8-HpCDF
1,2,3,4,7.8,9-HpCDF
OCDD
OCDF
Labeled CDDs and CDFs
13C12-2,3,7,8-TCDD
13C12-2.3,7.8-TCOF
'3C12-1,2,3,7,8-PeCDD
13Cl2-1,2,3,7,8-PeCOF
13C12-2.3,4,7.8-PeCOF
C12-1,2,3,4,7,8-HxCOD
13C12-1,2,3,6,7,8-HxCDD
13C12-1,2,3,4,7,8-HxCDF
13C,,-1,2,3,6,7,8-HxCDF

^C^-I^.S./.S^-HxCDF
13C12-2,3,4,6,7,8-HxCDF
13C,,-1,2,3,4,6,7,8-HpCOD

IOC.,-1,2,3,4,6,7,8-HpCDF
-17 \f.
C,-- 1,2.3, 4. 7,8,9- HpCDF
12 13

Cleanup Standard
37Cl4-2,3,7,8-TCDD
Internal Standards
13C12-1,2,3,4-TCDD
13C12-1,2,3,7,8,9-HxCDD
CS1
(ng/mL)

0.5
0.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
5.0
5.0

100
100
100
100
100
100
100
100
100

100
100
100

100
100

200

0.5

100
100
CS2
(ng/mL)

2
2
10
10
10
10
10
10
10
10
10
10
10
10
10
20
20

100
100
100
100
100
100
100
100
100

100
100
100

100
100

200

2

100
100
VER(6)
CS3
(ng/mL)

10
10
50
50
50
50
50
50
50
50
50
50
50
50
50
100
100

100.
100
100
100
100
100
100
100
100

100
100
100

100
100

200

10

100
100
CS4
(ng/mL)

40
40
200
200
200
200
200
200
200
200
200
200
200
200
200
400
400

100
100
100
100
100
100
100
100
100

100
100
100

100
100

200

40

100
100
CSS
(ng/mL)

200
200
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
2000
2000

100
100
100
100
100
100
100
100
100

100
100
100

100
100

200

200

100
100
(6)   Section 14.3 -  calibration verification (VER)  solution.
 40
040

-------
                        Table 5
 GC RETENTION TIME WINDOW DEFINING  STANDARD MIXTURES AND
        ISOMER SPECIFICITY TEST STANDARD MIXTURES

DB-5 Column GC  Retention Time Window Defining Standard
(Section 6.15)

Congener        First Eluted            Last  Eluted
TCDF
TCDD
PeCDF
PeCDD
HxCDF
HxCDD
HpCDF
HpCDO
1,3,6,8-
1,3,6,8-
1,3,4,6.8-
1,2,4,7,9-
1,2,3,4,6,8-
1,2,4,6,7,9-
1,2,3,4,6,7,8-
1,2,3,4,6,7,9-
1.2,8,9-
1,2,8,9-
1,2,3,8,9-
1,2,3,8,9-
1,2,3,4,8,9-
1,2,3,4,6,7-
1,2,3,4,7,8,9-
1,2,3,4,6,7,8-
DB-5 TCDD Isomer Specificity Test Standard
(Section 6.16.1)
                  1,2,3,4-TCDD          1,2,3,7-TCDD
                  1,2,7,8-TCDD          1,2,3,8-TCDD
                  1,4,7,8-TCOD          2,3,7,8-TCDD

DB-225 Column  TCDF  Isomer Specificity Test Standard
(Section 6.16.2)
                  2,3,4,7-TCDF
                  2,3,7,8-TCDF
                  1,2,3,9-TCDF
                                                      04i
                                                                                                           41

-------
                                                 Table 6
                  REFERENCE COMPOUNDS FOR QUANT ITAT IOH OF NATIVE AND  LABELED PCODS AND PCDFS
 Native PCDDs and PCDFs
Reference Compound










1
1
1




1
1
1
1

1
1
2
p

,2
.2


1,2
1.2
2,3
.2.3
.2.3
? 3
? 3

? 3

.2.3
,3.4
1 4

,3.4
.3.4


2,3,
2,3,
,3,7
,3,7
,4,
,4,
,6,
7
4

f>

,7,
,6,
6

.6,
.7.


7
7
7
8
7

7

3
7
7

7
8


7,3-TCDD
7,3-TCDF
,8-PeCDD
,8-PeCDF
,S-PeCDF
,8-HxCDD
,8-HxCDD
9-HxCDD
8-HxCDF

8-HxCDF
,9-HxCDF
,8-HxCDF
8-HpCDD
,8-HpCDF
,9-HpCDF
OCDD
OCDF
13c
13c
1-*
1
13c
13c

13c
1-1
13c
13c
c
13C
-C12
13C12
C
C12
JC

12
12


12
12
12
12
-1
-1
-1
13C .2
12 '
13
13C12-2,
12-1.2.3
12-1,2,3
12-2
-1 ?
-1 ?


-1 ?

-1 ?

-1 ?

-2.3
2 3

.2.3
.2.3
.3,4
T 4
1 6


3 4

T 0

3 7

.4.6
4 6

,4.6
,4,7
3, 7,8-TCDD
3,7,8-TCDF
,7,8-PeCDD
,7,8-PeCDF
,7,
7
7


7

7

a

,7,
7

.7,
.3,
n
8-PeCDF
8-HxCDD
8-HxCDD
(1)
8-HxCDF

8-HxCOF
9-HxCDF
8-HxCDF
8-HpCDD
8-HpCDF
9-HpCOF
"C1?-OCDD




c
12-OCDD
(1)   1,2,3,7,8,9-HxCOD  is quantified using the average
     1,2,3,6,7.8-HxCDD.
Labeled PCDDs and PCDFs
13C,,-2. 3, 7,8-TCDD
13
IJC12-2,3,7,8-TCDF
13C12-1,2,3,7,8-PeCDD
13C12-1.2,3,7,8-PeCDF
13C12-2,3,4,7,8-PeCDF
13C12-1,2,3,4,7,8-HxCDD
13C12-1,2,3,6,7,8-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1.2,3,4,7,8-HxCDF
13C,,-1,2,3,6.7,8-HxCDF

13C12"2,3,4,6.7,8-HxCDF
13C12-1,2,3,4,6.7,8-HpCDD
13C12-1,2,3,4,6,7,8-HpCDF
12 13^ .QCDD
37Cl4-2, 3, 7,8-TCDD
ponses for the 13C,--1,2,3,4,
Reference Compound
13C.,-1,2,3,4-TCDD
13
C12-1,2,3,4-TCDD
13C12-1,2,3,4-TCDD
13C12-1,2,3,4-TCDD
13C12-1,2,3,4-TCDD
13C.,-1,2,3,7,8,9-HxCOD
.-12
C.2-1,2,3,7,8,9-HxCDD
13Cl2-1,2,3,7,8,9-HxCDD
13C.,-1,2,3,7,8,9-HXCDD
12
C12-1,2,3,7,8,9-HxCOD
13C12-1.2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,7,8,9-HxCDD
13C.,-1,2,3,7,8,9-HxCDD

13 12 ' ' ' ' ' D
°C12-1,2,3,7,8,9-HxCDD
13C12-1,2,3,4-TCDD
,7,8-HxCDD and 13C.--
 42
                                                      042

-------
                                                    Table 7
                                    ACCEPTANCE CRITERIA FOR PERFORMANCE TESTS
Compound
2,3,7,8-TCDO
2,3,7.8-TCDF
1.2,3,7,8-PeCDD
1,2,3,7,8-PeCDF
2.3,4,7,8-PeCDF
1,2.3,4,7,8-HxCDD
1,2.3,6,7,8-HxCDD
1,2,3,7.8,9-HxCDD
1.2.3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,a,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3.4.6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCOF
OCDD
OCOF
13C12-2,3.7,8-TCDD
13C12-2.3,7,8-TCDF
1X
l:>C12-1,2.3,7,8-PeCDD
13C12-1.2.3,7,8-PeCDF
13C12-2.3,4.7.8-PeCDF
•]•»
C,,-1, 2,3,4.7, 8-HxCDD

l:>C.,-i.2.3,6,7,8-HxCDD
.,12
C,,-1,2,3,4,7,8-HxCDF
•!•» '^
'JC..,-1,2,3,6.7,8-HxCDF
.,12
°C,, -1,2.3. 7,8, 9-HxCDF
.,12
C.--2,3,4,6,7,8-HxCDF
13 12
'^.,-1,2,3,4,6,7,8-8^00

C.,-1,2,3,4,6.7,8-HpCDF
12 ...... K-
C., -1,2,3,4, 7.8, 9-HpCDF
13
c.2-ocoo
37Cl4-2,3.7.8-TCDD
Test

Cone. (1) s
(ng/tnL) (ng/mL)
10
10
50
50
50
50
50
50
50
50
50
50
50
50
50
100
100
100
100
100
100
100
100
100

100
100
100
100

100
100
100

200
10
1.5
2.0
4.2
4.6
4.2
5.5
5.5
9.5
6.3
4.0
4.0
5.0
6.4
3.6
4.2
13.0
45.0
-
-
-
-
-
-
.

-
-
-
.

-
-
-

-
-
IPR (2)

X
( ng/mL )
3.9 -
3.2 -
47.5 -
44.2 -
45.3 -
30.9 -
33.2 -
22.7 -
25.2 -
39.1 -
37.9 -
27.4 -
27.4 -
39.5 -
36.6 -
69.4 -
46.1 -
25.0 -
25.0 -
25.0 -
25.0 -
25.0 -
25.0 -
25.0 -

25.0 -
25.0 -
25.0 -
25.0 -

25.0 -
25.0 -
25.0 -

50.0 -
2.5 -
20.6
26.8
50.5
54.0
50.3
70.2
65.9
90.9
92.0
54.4
62.9
85.5
76.5
62.1
64.9
154.6
139.8
150.0
150.0
150.0
150.0
150.0
150.0
150.0

150.0
150.0
150.0
150.0

150.0
150.0
150.0

300.0
15.0
OPR(2)
(ng/mL)
5.9 -
6.6 -
35.6 -
36.7 -
37.8 -
35.1 -
33.3 -
31.8 -
36.9 -
34.8 -
37.1 -
35.7 -
37.5 -
37.4 -
36.9 -
75.6 -
69.5 -
25.0 -
25.0 -
25.0 -
25.0 -
25.0 -
25.0 -
25.0 -

25.0 -
25.0 -
25.0 -
25.0 -

25.0 -
25.0 -
25.0 -

50.0 -
2.5 -
14.2
12.7
58.1
57.3
56.9
60.4
64.4
61.2
58.8
58.8
55.7
60.0
56.8
60.5
60.6
118.7
127.0
150.0
150.0
150.0
150.0
150.0
150.0
150.0

150.0
150.0
150.0
150.0

150.0
150.0
150.0

300.0
15.0
VER
(ng/mL)
8.6 -
8.8 -
44.2 -
46.7 -
47.2 -
37.6 -
39.7 -
42.6 -
41.5 -
40.5 -
45.7 -
44.1 -
41.6 -
43.1. -
43.6 -
87.5 -
83.9 -
90.0 •
87.7 -
80.6 -
81.8 -
83.0 -
76.1 -
84.0 -

85.2 -
85.0 -
89.5 -
85.7 -

82.2 -
88.5 -
89.0 -

164.2 -
6.1 -
11.6
11.3
56.6
53.5
53.0
66.5
63.0
58.7
60.2
61.7
54.5
56.7
60.2
58.0
57.3
114.4
119.2
111.2
114.0
124.0
122.3
120.5
131.3
119.1

117.4
117.7
111.7.
116.7

121.6
113.1
112.4

243.6
11.6
(1)  All  specifications are given as concentrations in the final extract or standard solution.
(2)  s  =  standard deviation of the concentration; X = average concentration.  Concentration limits  for  labeled
     compounds in IPR and OPR aliquots are based on requirements for labeled compound recovery  of 25-150% (Sections
     8.2.3 and 14.5.3).
                                                       043
                                                                                                         43

-------
                                                    Table  S
                            SAMPLE PHASE AND QUANTITY EXTRACTED FOR VARIOUS MATRICES
Sample Matrix (1)
SINGLE PHASE
Aqueous
Solid
Organic
MULTIPHASE
Liquid/Solid
Aqueous/solid
Example

Drinking water
Groundwater
Treated wastewater
Dry soi I
Compost
Ash
Waste solvent
Waste oil
Organic polymer


Wet soil
Percent Quant i ty
Solids Phase Extracted

<1 (2) 1000 mL
'1
>20 Solid 10 9
<1 Organic 10 g


1-30 Solid 10 g
        Organic/solid


       Liquid/Liquid

        Aqueous/organ ic
        Aqueous/organic/
        solid
Untreated effluent
Digested municipal  sludge
Filter cake
Paper pulp
Tissue

Industrial sludge
Oily waste
tn-process effluent
Untreated effluent
Drum waste

Untreated effluent
Drum waste
                                                               1-100
                                                                  >1
                                                                                  Both
Organic
Organic
& solid
                     10 g
10 g
                                                                                                      10 g
(1)  The exact matrix may be vague for some samples.  In general, when the CDDs and CDFs  are  in  contact  with  a
     multiphase system in which one of the phases  is water, they will be preferentially dispersed in or adsorbed
     on the alternate phase,  because of their  low  solubility in water.

(2)  Aqueous samples are filtered after spiking with labeled analogs.   The filtrate and the material  trapped on
     the filter are extracted separately,  and  then the extracts are combined for cleanup and analysis.
 44
                                                        04

-------
           PERFORMANCE EVALUATION OF METHOD 1613

    .  High Resolution Gas Chromatography/High Resolution Mass Spectrometry
Determination of Tetra- through Octa-Chlorinated Dibenzo-p-Dioxins and Dibenzofurans
                             by Isotope Dilution
                                 March 1990
                           USEPA OFFICE OF WATER
                 OFFICE OF WATER REGULATIONS AND STANDARDS
                  INDUSTRIAL TECHNOLOGY DIVISION (WH-552)
                          SAMPLE CONTROL CENTER
                            Washington, DC 20460
                                              'i' J

-------
                                   TABLE OF CONTENTS




Section




Introduction




Determination of Method Requirements




Method Selection and Initial Development




Single Laboratory Testing




Confirmatory Testing




Interim Method Testing




Field Validation Study




Conclusions




Outliers




Estimation of Variance Components




Derivation of Quality Control Limits for Accuracy




Derivation of Quality Control Limits for Precision




Calibration Linearity




Relative Retention Time




Method 1613,12 September 1988 Draft




Method 1613, July 1989
    Page Number




         1




         1




         2




         4




        11




        12




        14




        19




Appendix A




Appendix B




Appendix C




Appendix D




Appendix E




Appendix F




Appendix G




Appendix H
                                                       04G

-------
                                    UST OF TABLES

Table        Description

  1           Comparison of Candidate Methods

  2           Soxhlet/Dean Stark versus Soxhlet t Test Results

  3           Method Detection Limit

  4           1613 Field Validation Study - Samples by Industry and Lab

  5           Initial Precision and Recovery - All Labs

  6           Ongoing Precision and Recovery - All Labs

  7           Precision and Recovery of Labeled Compounds in EPA Samples

  8           Calibration Linearity QC Limits

  9           Calibration Verification QC Limits

 10           Relative Retention Time QC Limits

11A          Initial Precision and Recovery QC Limits for Accuracy

11B          Initial Precision and Recovery QC Limits for Precision

12            Ongoing Quality Assurance Limits

13            Precision and Recovery of Labeled Compounds - EPA Samples, All Labs

14            Precision and Recovery of Labeled Compounds - EPA Samples, Labs 2
              and 3

15            QC Limits for Precision and Recovery of Labeled Compounds - EPA
              Samples
                                                       047

-------
                                    LIST OF FIGURES

Figure                   Description

  1                       Initial Precision and Recovery of Labeled and Native
                          Compounds - Frequency Distribution Plot of All Labs Combined

  2                       Ongoing Precision and Recovery of Labeled and Native
                          Compounds - Frequency Distribution Plot of All Labs Combined

  3                       Precision and Recovery of Labeled Compounds
                          Frequency Distribution Plot by Lab

  4                       Precision and Recovery of Labeled Compounds
                          Frequency Distribution Plot of All Labs Combined
                                             111
                                                         048

-------
1.    INTRODUCTION

      In 1976 the U.S. Supreme Court issued a consent decree requiring the Environmental Protection
Agency (EPA) to measure and limit 65 compounds and classes of compounds in effluents discharged to
receiving waters in the United States. The list of 65 was subsequently refined by EPA to a list of 129 specific
analytes termed the "Priority Pollutants" and codified as the Section 307(a) list of "toxic pollutants" in the
1977 Clean Water Act (CWA) amendments. Priority Pollutant Number 129 is 2,3,7,8 -tetrachlorodibenzo-
p-dioxin (23,7,8-TCDD), one of the most toxic substances known to man.

      Currently, EPA Method 613 is  the only method for the analysis of  2,3,7,8-TCDD that  has been
promulgated  under CWA Section 304(h).  This method, developed in  the  late 1970's, utilizes  gas
chromatography and low resolution mass spectrometry, with minimal extract  cleanup steps. The  detection
limit for 2,3,7,8-TCDD in water is 2000 ppq for Method 613. As a  result, it does not achieve the  detection
limit for 23,7,8-TCDD currently required by the Agency for analysis of treated effluents (10 - 25 ppq).

      Method 1613 was developed by the Agency in response to the need for analyses of treated effluents at
low levels of 23,7,8-TCDD  (10 - 25  ppq).  It was designed for regulatory development purposes and
compliance monitoring under the National Pollutant Discharge Elimination System (NPDES, CWA Section
402).

      Method 1613 is a survey method designed to quantitate 23,7,8-TCDD as well as  the other sixteen
23,7,8 substituted dioxins and furans.   It employs gas chromatography coupled with a state-of-the-art high
resolution mass spectrometric method for the analysis.  Method 1613 incorporates the  Agency's 500/600
series QA/QC  program,  including rigorous start-up tests and on-going demonstrations of laboratory
performance. It employs isotope dilution as a means of quantifying the analytes of interest. Isotope dilution
is a technique in which stable isotopically labeled analogs of the target compounds are added to each sample
and used to quantify the native analytes present, in order to reduce the variability of the analysis and correct
for recovery bias.

      This report presents results of various single-laboratory development studies, sample and QA/QC
analyses performed  to date using Method 1613.  The objective is to demonstrate that the Method achieves
acceptable precision and bias across a range of industrial effluents and contract laboratories, and is therefore
a valid method for  promulgation as a  Section 304(h) alternate test procedure for the analysis of 2,3,7,8-
TCDD. A secondary objective is the promulgation of Method 1613 under Section 304(h) for the analysis of
the 23,7,8 substituted dioxin and furan isomers other than 2378-TCDD.

2.    DETERMINATION OF METHOD REQUIREMENTS

      The  CWA  (as amended  1987) requires  the Agency  to  consider  water  quality-based effluent
limitations in conjunction with traditional Best Available Technology (BAT) efforts. The Agency's water
quality criteria for 23,7,8-TCDD of 0.013 ppq (at a 10"6 estimated human cancer risk) is below the limit of
detection of current analytical techniques.

      Section 304(m) of the CWA requires the Agency to periodically review existing effluent guidelines,
and to develop guidleines for new industries and new compounds.  In response to this requirement, the
Office of Water (OW) will address the issue of regulation of 23,7,8-TCDD in the treated effluents  of several
industries. In order to collect data to support guideline development and support current regional NPDES
efforts, OW must measure 23,7,8-TCDD in treated effluents at the lowest levels practical using available
analytical technology.  Further,  methods  utilized by  OW must  contain a rigorous QA/QC program
consistent with the 600/1600 series methods that have been promulgated under 304(h).
                                                       040

-------
      Therefore, the method implemented by the Office of Water for the measurement of 2,3,7,8-TCDD
must meet or exceed the following data quality objectives (DQOs):

     o   The method must be appropriate for nationwide use by EPA and contract laboratories as well as
         by the regulated  community.  This requires a method that is sufficiently rugged that minor
         variations in use do not significantly affect the usability (precision and bias) of the data obtained.

     o   The method must utilize high resolution gas chromatography combined with high resolution mass
         spectrometry (HRGC/HRMS). The HRMS instrument must have a resolving power in excess of
         10,000 atomic mass units.

     o   Sample preparation  procedures must minimize sample  handling  and therefore reduce  the
         potential for contamination.

     o   Extraction and cleanup procedures must be sufficient to allow removal of interferences so that
         target detection limits can be reached.

     o   Isotope dilution quantitation must be utilized to allow precise estimates of analytical  variability,
         as this technique has proven useful  in the development of BAT effluent limits for other
         compounds.

     o   A Limit of Quantitation  (LOQ) for 2^,7,8-TCDD of 1 ppt for solid samples and 10 ppq for
         effluents is required.

     o   In order to collect data on other potentially toxic dioxins and furans, the method must  measure
         the 16 other 2,3,7,8 substituted dioxins and furans.

      Due to  the  recent discovery of 2,3,7,8-TCDD in pulp and paper export vectors,  EPA Regional
personnel are currently working on NPDES permits for 2,3,7,8-TCDD, and require a standard method for
nationwide use. Industry representatives argue that the Agency can not require utilization of a method that
is not promulgated under 304(h), and therefore, the  only method that can be utilized is Method 613.
However, given that the MDL published in Method 613 is 2000 ppq, Method 613 would be inappropriate for
monitoring a pollutant for which the water quality criteria is five orders of magnitude lower (0.013 ppq).

3.     METHOD SELECTION AND INITIAL DEVELOPMENT

      After reviewing OWs method  requirements, several meetings were conducted with EPA Regional
personnel and  various dioxin experts from commercial and industrial laboratories.  Concurrent with these
meetings, ITD Sample Control Center staff reviewed seven candidate methods and reviewed data for 2,3,7,8-
TCDD analysis from contract laboratories using various existing methods. The information shown in Table
1 is a result of these efforts.
                                                     050

-------
                    TABLE 1 - COMPARISON OF CANDIDATE METHODS

METHOD                    POSITIVE ATTRIBUTES              NEGATIVE ATTRIBUTES

613                          304(h) Promulgation                    Low Resolution MS
                             500/600 Series QA/QC                  MDL2000ppq
                                                                  Limited to 2,3,7,8-TCDD

ERL-Duluth                  Extensive use                          Used only by ORD lab
                             MDL adequate                         Primarily Fish Tissue

Wright State                  Extensive use on                       Used primarily by WSU
University                    pulp and paper matrix                   Labor intensive, High cost
                                                                  Hybrid GC column
                                                                  Limited to 2,3,7,8-TCDD/F

NCASI                      Appropriate matrices                   Used only by NCASI
                                                                  High Cost
                                                                  Limited to 2,3,7,8-TCDD/F

Dow Chemical                Extraction & Cleanup                   Low Resolution MS
                                                                  Used only by Dow
                                                                  Limited to 2,3,7,8-TCDD/F

8290                         Extensive use                          Published version is "draft",
                                                                  Extensively modified,
                                                                  Not standardized across labs

Triangle                      Highly refined 8290                    "Proprietary"
Laboratories
"8290x"
      After review of the candidate methods, OW decided to develop a method specifically designed to
meet the data quality objectives listed in Section 2 above. The initial version of Method 1613 was prepared
by combining the "best" features of the candidate methods and adding other features required to meet the
stated data quality objectives. Those features included the following:

    o   The EMSL-Cincinnati format and the 600 series QA/QC Program were taken from Method 613
    o   Sample preparation features were taken from the ERL-Duluth, Wright State University, NCASI,
         Dow, and 8290 methods

    o   Chromatography and mass spectrometry specifications were taken from 8290

    o   Isotope dilution quantitation was specified, based on ITD's lengthy experience with Method 1625

      The initial  draft  of  Method  1613 is dated 12 September  1988, and employs  isotope dilution
techniques and a 600 series QA/QC program (Attached as Appendix G).
                                                         051

-------
4.    SINGLE LABORATORY TESTING

      Two concurrent single laboratory tests were initiated by ITD during the early development of Method
1613. The first test involved the use of a new extraction technique for solid matrices.  The second test
involved the use of a new extraction technique for aqueous matrices. Both techniques were taken from work
performed at Dow, and were modified for use in Method 1613. A third single laboratory test was initiated by
ITD to determine the Method Detection Limit (MDL) of the July 1989 version of the method.

4.1   Soxhlet-Dean  Stark (SDS) Extraction Evaluation

      A soxhlet  extraction procedure  is specified in many analytical methods  for the extraction of
polychlorinated dibenzo-/?-dioxins and polychlorinated dibenzofurans (PCDDs/PCDFs) from solid  and
semi-solid matrices  such as soil and sludge.  Basically, the procedure involves the repeated refluxing of an
organic solvent such that the solvent percolates through the sample matrix and extracts the compounds of
interest by dissolution.

      Typically, when using soxhlet extraction, steps must be taken to remove the water from the sample
matrix prior to extraction because the organic solvents used for extraction are not water miscible.  This is
particularly true when extracting samples with high moisture contents such as sludge. Techniques  such as
filtration and centrifugation have been employed to remove the water in other analytical procedures.  The
addition of sodium sulfate to adsorb the water from the sample has been extensively employed in the
analysis of organic compounds from environmental matrices. Each of these techniques involves additional
handling of the sample, and therefore increases the potential for introduction of contaminants, loss of
analytes,  or loss  of the entire sample. Since each sample handling step has the potential  to increase the
variability of the data produced by the overall analytical method, it is critical to minimize sample handling
steps when dealing with 2,3,7,8-TCDD at the levels addressed in the Method 1613 (10-25 ppq).

      Another technique for removing water from a sludge sample and other types of wet samples involves
the use of a Dean Stark water separator in conjunction with a soxhlet extractor. This piece of glassware fits
between the soxhlet extractor and the condenser, and offsets the condenser to one side. This offset provides
space for a dog leg that drops straight down from the bottom of the condenser.  Ground glass joints connect
the separator to the soxhlet extractor and the condenser, and the dog leg ends in either a graduated receiving
tube or a stopcock.  The resulting combined apparatus is referred to as the Soxhlet-Dean Stark apparatus,
abbreviated as SDS.

      Water is removed from the sample  during extraction by the process of azeotropic distillation.
Originally designed for processes involving xylene, the  separator works  equally well with toluene, the
extraction solvent specified in Method 1613, because  toluene and water  form an azeotrope that boils at a
temperature of 85 °C. When the azeotropic vapor condenses, the liquid drops into the dog leg, where the
toluene floats on top of the more dense water, and eventually flows back into the distilling flask. The use of
a separator with a stopcock allows the laboratory to draw off the collected water without interrupting the
extraction process.

      The SDS extraction technique offers the potential for enhancing analytical precision  and decreasing
bias by reducing the number of sample handling steps and more effectively extracting the sample. It has a
noteworthy advantage over simple soxhlet extraction  in that the percent moisture  in the sample may be
determined directly  from the sample being extracted, rather than from  another aliquot of the sample which
may not be truly representative of the aliquot that is extracted.

      The SDS combination offers a significant advantage over the use of sodium sulfate to adsorb the
water from the sample matrix.  Sodium sulfate drys the sample by hydrating itself with the water in the
                                                             0-!

-------
sample. During the process of extraction, some of this water of hydration may be lost back to the sample or
to the solvent and, as a result, the dehydrated sodium sulfate may seal off pores in the surface of the solid
matrix.  This process effectively traps the analytes  of interest within  the matrix, thus preventing their
extraction.

      Another potential problem with the use of sodium sulfate in the analysis of very low levels of
PCDDs/PCDFs is the loss of analytes by absorption on contaminants in the reagent itself.  The reagent is
typically purified of any organic contaminants by heating it in a muffle furnace at high temperatures.  Any
organic material present is  charred,  often giving the reagent a light gray cast.  While the heat treatment
effectively prevents this organic material from being extracted from the sodium sulfate, PCDDs/PCDFs are
strongly adsorbed by activated carbon, and the charred material represents a source of activated carbon.
Because of the potential loss of analytes through the use of sodium sulfate, ITD has chosen to avoid the use
of this reagent during the extraction of samples.

      Although Dow had published data on the use of the SDS in other solid matrices, the first phase in the
validation of Method 1613 consisted of an intralaboratory study to ascertain the comparability of the SDS
procedure with the more commonly used soxhlet procedure when applied to municipal sewage sludge.

4.1.1  Experimental Design

      To  demonstrate the comparability of the SDS procedure proposed for EPA Method 1613 with the
soxhlet procedure currently  employed in other analytical methods for PCDDs/PCDFs, a five gallon sample
of sewage sludge was sent to Triangle Laboratories for analyses.  ITD SAS 131, Episode 1519 consisted of
nine analyses of the sludge sample, as follows:

      - 3 analyses of the unspiked sludge, extracted by soxhlet

      - 3 analyses of sludge spiked with PCDDs/PCDFs, extracted by soxhlet

      - 3 analyses of sludge spiked with PCDDs/PCDFs, extracted by SDS

      A preliminary examination of the unspiked sludge data indicated that very few of the PCDDs/PCDFs
were detected in these samples. Hence, it would have been impossible to collect statistically  meaningful
results from an experiment in which the unspiked sludge was extracted by both procedures.  Therefore, the
unspiked data were not considered further.

      The sludge was spiked by the laboratory with all 17 2,3,7,8-substituted PCDD/PCDF isomers prior to
extraction. Replicate aliquots were extracted by soxhlet alone and by the SDS procedure. Because this work
was done during the earliest stages of development of Method 1613, Triangle Laboratories utilized Method
"8290x" for the analyses.  (For  the purpose of testing the SDS procedure, this instrumental aspects of "8290x"
were deemed to be sufficiently similar to Method 1613). The resulting data from Triangle laboratories were
evaluated  to determine if the SDS procedure provided comparable or better  results than  the soxhlet
procedure alone.

      Data for all the samples were reviewed to determine if they met the requirements of Method "8290x"
for identification of analytes. All data were consistent with those requirements.

4.1.2  Statistical Results

      Because the extraction procedures were compared in a controlled experiment rather than during the
routine analyses of field samples for other purposes,  a more rigorous statistical analysis of the results was
possible. The results of the triplicate analyses of the spiked sludge by both the soxhlet and SDS  procedures
were compared using standard statistical techniques.  The mean concentrations of the three measurements
of each isomer were determined for each extraction procedure.  For each isomer, the mean concentration for
                                                         053

-------
the SDS procedure was compared to the mean concentration for the Soxhlet procedure using a two-tailed r-
test. The r-test was repeated for each of the 17 isomers.

      The results of those tests are given in Table 2. The table lists the mean concentration of each isomer
for the  samples extracted by both methods and the variance (s2) of each isomer by each method.  The t
statistic was calculated for each pair of means, and is given in Table 2. The value of t for each pair of means
was used to determine the probability,  P(r,4), that the mean concentrations of each isomer determined by
the different extraction procedures were not the same.  The null hypothesis was that the means were not
different  The null hypothesis would be rejected, i.e. the means were different, if the probability value that
was calculated was less than 0.05.

      All 17 isomers were detected in all six samples, with one exception.  1,2,3,4,7,8,9-HpCDF was not
detected in one of the soxhlet-extracted replicates. In the case of this isomer, the r-test was performed two
ways:  once using the estimated detection limit  (EDL) calculated by the laboratory in place of the
undetected concentration; and once using a value of 0.0 for the undetected concentration. The results of
both r-tests appear near the bottom of Table 2.

      Only two analytes yielded values for P(r,4) that were less than 0.05. Those analytes were 1,2,3,4,6,7,8-
HpCDD, with P equal to 0.038, and 23,7,8-TCDF, with P equal to 0.032. OCDD had a P value of 0.060, but
none of the results for the other analytes were close to the 0.05 level. The r-test results indicate that, for
1,2,3,4,6,7,8-HpCDD  and 2,3,7,8-TCDF,  there  is a  statistically  significant difference in  the  mean
concentrations of these analytes extracted by the two procedures.  In both cases, the SDS procedure yielded
a higher mean concentration than the soxhlet alone.  Therefore, it would appear that for at least two
analytes, the SDS  extraction procedure gave higher results than extraction by soxhlet alone, and for the
remaining 15 analytes, the SDS results comparable to those from the soxhlet alone.

4.13 Limitations

      The r-test used for these data assumes that the variances of the two means being compared are equal,
and can be estimated from the sample data. All  the r-tests use a value of 4 for the number of degrees of
freedom for the test, i.e. 3+3-2.  A review of the data in Table 2 suggests that the variances may not be equal
and that, in general, the variance of the SDS data is less than that of the Soxhlet data (12 of 17 compounds
had lower variances by SDS).  The notable exceptions to this trend are the variances for the two isomers
which were statistically different between  the two extraction methods.  The variances for 1,2,3,4,6,7,8-
HpCDD and 2,3,7,8-TCDF were higher for the SDS data.

      Therefore, r-tests were performed  for several of the isomers assuming that the variances are not
equal.  This version of the r-test calculates the same r statistic as is reported on Table 2, but uses  the sample
variances and the numbers of observations for each isomer (n) to determine a different number of degrees of
freedom (n') than  is  used when the variances are assumed to be equal. This  new value for the degrees of
freedom was calculated for four of the isomers, and resulted in n'<0. Values of n' less than zero have no
validity, indicating that there are not enough data to perform this variation of the r-test.

      The method used by Triangle Laboratories also includes isotopically labeled "surrogates".  Unlike the
labeled  analogs used in an isotope dilution method such as Method 1613, the surrogates are not used for
quantifying the target analytes, but may be used  to judge the effectiveness of the  sample extraction
procedures. Triangle reported data for the recovery of these surrogates and, in general, the recoveries were
good for both extraction methods. Two-tailed r-tests were performed on some of the surrogate recovery data
as well.  No  statistically significant differences between the mean  recoveries of the surrogates could be
found, presumably also due to the small size of the data set.

      In an attempt to overcome the small number of observations in the data set, the data for most of the
native isomers were grouped together and tested.  Eleven of the 17 spiked isomers had  mean concentrations
near 200 ppt.  Data for these isomers were retained in the second data set tested.
                                                           054

-------
                 TABLE 2 - SOXHLET/DEAN STARK VERSUS SOXHLET t TEST RESULTS
                                    SDS
Soxhlet
Isomer
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,23,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
23,4,7,8-PeCDF
1,23,4,7,8-HxCDF
1,23,6,7,8-HxCDF
23,4,6,7,8-HxCDF
1,23,7,8,9-HxCDF
1,23,4,6,7,8-HpCDF
1,23,4,7,8,9-HpCDF
mean cone.
34.449
200.503
232.960
212.913
220.462
496.677
3717.909
58.707
204.043
170.667
248.912
223.459
217.313
181.578
262.347
174.869
s2
6.121
15.124
55.071
43.294
8.738
5342.758
9.17 105
16.595
4.826
22.333
5.510
60.549
90.934
91.152
335.027
452.480
mean cone.
37.410
209.472
236.235
218.159
218.760
322.147
1711.635
48.494
206.290
183.667
252.408
222.163
228.919
178.551
242.249
160.569
s2
3.486
90.973
47.987
356.928
59.923
1039.967
2.37 105
3.228
73.55
37.333
188.022
451.416
358.418
222.782
1320.180
1283.639
t
-1.351
-1.231
-0.456
-0.371
0.290
3.089
2.64
3.244
-0.359
-2.380
-0.355
0.081
-0.774
0.242
0.698
0.485
Pfr.41
0.125
0.288
0.673
0.730
0.780
0.038 *
0.060
0.032 *
0.756
0.078
0.756
0.940
0.482
0.822
0.524
0.653
       (using EDL of 189.0 for missing soxhlet value)
1,2,3,4,7,8,9-HpCDF         174.869      452.480         97.330      7751.0
       (using 0.0 for missing soxhlet value)
OCDF                    478.340      814.661        511.207     49760
* indicates analytes whose mean concentrations are significantly different at the 0.05 level.
All concentration data are given in parts-per-trillion, as reported by the laboratory
                   1.211
                   0.207
0.292
0.850
                                                        055

-------
      Two of the other six isomers had mean concentrations that were much lower or much higher than 200
ppt. OCDD was found at relatively high concentrations (900-1900 ppt) in the unspiked sample aliquots, and
1,2,3,4,6,7,8-HpCDD was consistently found in the unspiked aliquots at levels around 100 ppt.  Of the
remaining four  isomers, 2,3,7,8-TCDD, 2,3,7,8-TCDF,  and OCDF were not consistently found in the
unspiked aliquots, and 1,2,3,4,7,8,9-HpCDF was the isomer which was not detected in one of the spiked
aliquots. Therefore, data for these six isomers were not included in the second data set.

      The data for the three replicate extractions by the two extraction procedures for the eleven remaining
isomers yielded 33 data  points for each extraction method.  Despite this number of data points, the mean
concentrations of all eleven isomers combined were not significantly different for the two procedures at the
0.05 probability level. The variation in the concentrations of each isomer due to the instrumental analysis
itself may be sufficiently large to obscure any difference between the extraction procedures.

4.1.4 Conclusions

      Based on this study, the data  for 1,23,4,6,7,8-HpCDD and 2,3,7,8-TCDF do indicate a significant
difference between the extraction procedures, with the SDS extraction yielding higher mean concentrations
(recoveries) of  these two isomers.   However,  one cannot differentiate between  the means of the
concentrations of 15 of  the spiked analytes determined using the two extraction procedures.  The lack of
differentiation between those mean concentrations may reflect the small size of the data sets tested.

      Although the strength of this conclusion is limited by the size of the data set, the data indicate that
the SDS procedure is as good an extraction method as the soxhlet procedure alone, and better for at least
two isomers. Given this, and the advantages that fewer sample handling steps are involved in the SDS
procedure and  the percent solids content of the sample may  be determined directly from the aliquot
extracted for analysis, the SDS procedure was incorporated into Method 1613.

4.2  Erlenmeyer Flask Extraction Evaluation

      The September 1988 draft of  Method 1613 contained procedures for extracting aqueous samples
through the use of a continuous liquid-liquid extractor.  This procedure has been used in numerous other
Agency methods for extraction of organic compounds from aqueous samples.

      The continuous liquid-liquid extraction procedure was criticized by various reviewers of the draft
method for several reasons. The commercial laboratories argued that it was expensive and time-consuming
to perform.  LTD was less concerned  about the time and cost elements if the procedure would result in a
significant  improvement in data  quality,  as measured through the  precision  and accuracy of labeled
compound recoveries.

      Representatives from Dow Chemical argued that they had seen only minimal improvements in
precision through the use of the continuous liquid-liquid extraction procedure, and that it posed a large risk
for contaminating the samples.  The continuous liquid-liquid extractors are difficult to adequately clean
between samples, and thus at the very low levels of interest for PCDDs/PCDFs, could pose a significant risk
of cross-contamination between samples. Dow suggested that ITD adopt a procedure that Dow had used for
many years  for extraction of aqueous samples.  Their  procedure involves extracting  the sample in the
original bottle by adding a lighter than water solvent and a magnetic stirring bar to the bottle, and drawing a
vortex of solvent into the sample for  several hours. The advantages appeared to be simplicity and lack of
sample transfer steps, thus lowering the risk of contamination. As practiced by Dow on their own samples,
the extraction is sufficiently efficient  that the water and the used sample bottle are essentially "clean", i.e.
free of measurable PCDDs/PCDFs, and thus the used bottle containing the extracted water is simply capped
and discarded.
                                                          0
5G

-------
      The use of a lighter than water solvent, typically toluene or benzene, had additional advantages.
After extracting the aqueous phase of the sample (the filtrate), the actual volume of solvent containing the
material extracted from the filtrate would be added to the SDS apparatus and used to extract the particulates
collected on the filter.  Thus,  the two fractions are extracted with one portion of solvent, and there is no
need  to combine two separate extracts. Using the same solvent also eliminates the need for a solvent
exchange step that would be required when  a heavier than water solvent and a lighter than water solvent are
combined.

      As a result of these discussions, the use of an erlenmeyer flask and a magnetic stirring bar to extract
water samples was incorporated in the second draft of Method 1613. Given the time constraints of the ITD
method development activities  at the  time, no  separate evaluation study was designed.  Rather,  two
laboratories performing analyses for ITD during the 304(m)  Pulp and Paper guidelines review utilized this
procedure for aqueous samples. However, several problems became apparent immediately. They included:

     o    Need to filter the samples before extraction

     o    Difficulties in having samplers leave sufficient room in the bottle for the solvent

     o    Difficulty with the quantitative transfer of the extract out of the bottle

Dow has much more control over the entire sampling and analytical process, and thus were not concerned
about filtering samples beforehand. They know that their aqueous samples are very low in solids, and can
control how much water is added to each bottle during sampling. The quantitative transfer of the extract is
of less concern to Dow because they do not use a true isotope  dilution method for quantifying the analytes.

      In order to accommodate the need  for filtration of the samples before extraction, ITD devised a
scheme to filter the sample through a glass fiber filter directly into an erlenmeyer flask of approximately 2
liter volume. About 100 mL of toluene was added to this flask along with a magnetic stirring bar. The flask
was then placed on a magnetic stirrer, and extracted for about an hour. While this worked on many samples,
there were a significant number of samples where  emulsions prevented  the effective extraction of the
samples. In these instances, multiple extractions were performed and the extracts added together.

      ITD anticipated difficulty in having a variety of sampling crews leave 100-150 mL of empty headspace
in a nominal 1 liter sample bottle.  Both ITD and Dow ruled out  the  possibility of withdrawing  and
discarding a representative aliquot of the sample from a completely filled bottle in order to make room for
the extraction solvent.   The use of erlenmeyer flask also solved the  potential problem of the samplers
overfilling the sample bottles, thereby not leaving space for the addition of the solvent.

      Despite the numerous changes made to the extraction  procedure, neither commercial laboratory was
able to reliably achieve recoveries of the  isotopically labelled compound added  to each sample before
extraction that met the  QC criteria (25-150% recovery) specified in the draft method  . One laboratory had
very poor  results, and was not interested in its further use. The second laboratory made numerous efforts to
make the  procedure work.  They attempted to use multiple  extractions, combining three 100 mL solvent
extracts before cleanup  and analysis.  They encountered difficulties with several darkly-colored samples,
where they could not see the vortex through the sample itself,  which led to poor recoveries if the sample was
"understirred11, and led to severe emulsions if the sample was "overstirred".

      The quantitative transfer  of the extract also caused some  problems. The Dow procedure involves
adding reagent water to the bottle (or flask,  in this case), until the solvent is driven up into the narrow neck.
From the  neck, it is simply drawn off in a clean disposable pipette.  Additional reagent water is added to
ensure all the solvent is recovered. Unfortunately, in practice, complete transfer is not possible, and the loss
of the labeled compounds contained in whatever volume of solvent is not collected caused the recoveries of
the labeled compounds  to fall below  the method specifications.  Because  Dow does not practice isotope
dilution quantitation, their method  does  not contain such rigid specifications  for  labeled compound
recovery.
                                                             057

-------
      In August and September 1989, ITD met with both laboratories, and discussed the problem with
several other laboratories familiar with PCDD/PCDF analyses at the Dioxin '89 meeting in Toronto.  As a
result, the erlenmeyer vortex extraction procedure was eliminated, in favor of the more traditional
separately funnel extraction procedure. However, the separately funnel extraction requires a heavier than
water solvent. While easily accommodated in the extraction of water samples, the use of methylene chloride
for extraction precluded the possibility of using the same actual volume of solvent in the SDS extraction of
the filtered paniculate fraction. Therefore, Method 1613 was rewritten, resulting in a solvent exchange of
the methylene chloride to toluene, and combination of the two extracts prior to further cleanup.

43   Method Detection Limit (MDL) Study

      In preparation for promulgation of Method 1613 under Section 304(h), a single laboratory method
detection limit study was initiated by ITD in December 1989. The basic design of the study was dictated by
the procedure for  determining MDLs specified in Appendix B of 40 CFR Part 136, as published in the
October 26,1984 Federal Register. Briefly, this procedure specifies the following:

     o    Spiking at least 7 aliquots of reagent water with the analytes of interest

     o    Spiking levels in the range of one to five times the laboratory's estimate of the detection limit of
          each analyte

     o    Analysis of all replicates and calculation of a mean and standard deviation of the concentration of
          each analyte

     o    Calculating the MDL as the standard deviation times the Student's r value for (n-1) degrees of
          freedom, where n is the number of replicates

      The specifications in 40 CFR 136 list four ways in which to determine an estimate of the  detection
limit of the  method.  One way  is to estimate the concentration  in a sample that corresponds  to  an
instrumental  signal-to-noise ratio of 2.5 to 5.0. Based on calculations from the single laboratory performing
the study, the lowest concentration of 2,3,7,8-TCDD in a sample which produces a signal that is 2.5 times the
background instrumental noise is 3.3 ppq. However, the concentration is determined by isotope dilution,
and the calculation assumes that the recovery of the labeled compound (the isotope) is 100%.  Because the
draft of Method 1613 allows  data  acceptance when labeled compound recovery is as low as  25%, the
concentration of 3.3 ppq was adjusted upward by the inverse of the worst-case recovery, i.e.  3.3/0.25, which
equals 13.2 ppq. Following the specifications in 40 CFR 136, the spiking level was set at 25 ppq for TCDD,
within the range of 1  to 5 times the 13.2 ppq estimated detection limit. Based on similarities in instrumental
response to the other analytes, this spiking level was used for the tetra, penta, and hexa chlorinated dioxins
and  furans,  and 50  ppq was  used for  the  hepta  and octa compounds, which  typically yield a  lower
instrumental  response.

      The mean concentrations of each analyte from the replicate analyses are given in Table 3, along with
the calculated MDL for each.  One of the QC criteria for the qualitative identification of PCDDs/PCDFs by
selected ion monitoring gas chromatography and mass spectrometry is the ratio of the abundance of the two
ions monitored for each analyte. Interferences (non-target analytes) that elute in the retention time window
of the dioxins and furans will cause the ion abundance ratio to vary from the theoretical abundances of the
two  ions.  When  the ion  abundance for a chromatographic peak  falls outside established acceptance
windows (+  15% of theoretical), the analyte cannot be positively identified.  The risk of an interference
causing the ion abundance ratio to fail the QC criteria increases as the concentration  of the analyte
decreases. As can be seen by the numbers of observations for each analyte listed in Table 3, most of the
analytes failed to meet the identification criteria at least once among the seven replicates. The results were
worst for several of the hepta chlorinated dioxins and furans, to which the instrument is less sensitive than to
the analytes with fewer chlorine atoms attached. Although less than ideal, the MDL values in Table 3 were
calculated using only those values that met the qualitative identification criteria.
                                                10

                                                        058

-------
      Except for OCDD, all the MDL values fall within a general range of 2-30 ppq. The large variability of
the OCDD measurements  is believed to be due to the  ubiquitous presence of this  compound in  the
environment. The OCDD results were not corrected for the levels of OCDD found in the blanks associated
with the samples,  although the procedure in 40 CFR 136 may permit such corrections.   Given the  low
toxicity of OCDD and a  correspondingly  lower  degree of concern with the presence of OCDD in
environmental samples, the high MDL for OCDD was not deemed to be a problem at this time.

      Except for the OCDD, the MDL values in Table  3 are all below the "minimum  levels" listed in
Method 1613.  The minimum levels correspond  to the concentration in a sample  equivalent to  the
concentration of each analyte in  the lowest of the calibration standards, assuming 100% recovery of the
labeled compounds added to the  sample and used for quantification by isotope dilution.  Thus, the MDL
data presented here indicate that the method is capable of measuring PCDDs and PCDFs at levels at least as
low as the minimum levels specified in the method description.

      Ultimately,  the MDLs need to be determined in  a variety of additional matrices, by additional
laboratories, and at additional spiking levels.

5.   CONFIRMATORY TESTING

5.1  Laboratory Testing

      Beginning with the  September 12, 1988  draft of the method, and continuing  with the revisions
through July 1989, a total of three commercial laboratories have utilized Method 1613 for contract analyses
for ITD. Those laboratories are:

     o   Triangle Laboratories, Research Triangle Park, NC
     o   Twin City Testing, St. Paul, MN
     o   California Analytical Laboratories, W. Sacramento, CA

Each of these laboratories submitted comments and suggestions to ITD based on their use of the method.
The results of the analysis of field samples by these laboratories are discussed in the section of this report on
Field Validation.

5.2   Peer Review

      In  addition  to the comments received from the laboratories listed above, the draft method  was
circulated to a variety of other Agency and industry reviewers. Comments were received from the following:

     o   Terry Nestrick and Les Lamparski, Dow Chemical, Midland, MI

     o   Dr. Tang and Dr. Alwan, Region V CRL, Chicago, IL
     o    Dr. Jill Henes, ChemWest Laboratories, W. Sacramento, CA
     o    Gary Robertson, Lockheed Environmental Services Co., Las Vegas, NV
     o    Dale Rushneck, Chemex International, Fort Collins, CO

     o    Joan Fisk, Analytical Methods Implementation Section, OERR, USEPA, Washington, DC
     o    Dr. Larry LaFleur,  National Council of the Paper Industry for Air and Stream Improvement
         (NCASI), Eugene, OR.

     o    John Stanley, Midwest Research Institute, Kansas City, MO

    o    Heather  Cavalier, Cambridge Isotope Laboratories, Cambridge, MA
                                              11
                                                       053

-------
      In addition to  a  large number of editorial comments regarding typographical errors, erroneous
paragraph citations, etc., numerous technical comments were received.  The comments from Dow Chemical
on  the  extraction procedures  are  discussed  at  length in the section on the single laboratory testing
performed.

      Three of the commercial laboratories and a commercial provider of standards commented on the
concentrations of the calibration standards specified in the original draft of the method, as well as the
concentrations of the precision and recovery standard (PAR). As a result, the concentrations of some of the
standards were adjusted.

      The solvent used for the standards drew several comments, resulting in a change from iso-octane to
nonane for the calibration standards, and the addition of a procedure for diluting the nonane-based spiking
solutions with acetone, to make them more easily incorporated into water samples.

      Dow Chemical also recommended against baking the reusable glassware at high temperatures, as such
baking promotes the formation of active sites on the glass surface which  may cause the loss of analytes
during sampling processing. As a result, kiln baking of glassware was removed from the method, and the use
of disposable glassware was maximized.

      Whereas most other dioxin methods specify the use of benzene for extracting solid matrices, OERR
recommended a change to toluene for such extractions.  Given the health concerns involved in the use and
disposal of benzene, this change was well received by the commercial laboratories. Given that toluene forms
an azeotrope with water that boils at a higher temperature than the benzene-water azeotrope, the use of
toluene in the SDS extraction process is even more effective than benzene.

      In February of  1990, Yves Tondeur of Triangle Laboratories provided data that indicated that the
multi-layer alumina column specified in the July 1989 revision of 1613 may lead to selective losses of the
tetra-chlorinated isomers.  Inconsistent recoveries of these isomers had been noted at other laboratories.
ITD speculated that the multi-layer column may not be sufficiently rugged for widespread use.

      NCASI had numerous comments on the specifics of sample extraction and cleanup. Many of these
were based on their experience with pulp and paper matrices. In particular, they  were concerned that
Method 1613 did not give sufficient guidance on dealing with the emulsions that may form during the
extraction of water samples, and that the multi-layered alumina column used for extract cleanup may not be
adequate for many sample extracts.

      As the result of a meeting between ITD and NCASI, held in February 1990,  Method 1613 will be
modified to incorporate additional guidance on treating emulsions.  Also, after review of Triangle's data on
the multi-layer alumina  column and consideration of NCASI's comments on the column, a single-layer
alumina column will be used instead of the current multi-layer cleanup column.

6.   Interim Method Description

      A copy of the July 1989 revision of Method 1613 is attached as Appendix H.  This version of the
method  contains  the changes made in response to all the comments cited  above except those made in
February 1990 from Triangle and NCASI.

      The method is written in a format consistent with that of EMSL-Cincinnati and other ITD isotope
dilution methods, including specific section for:

    o    Scope and Application

    o    Summary of the method

    o    Contamination and Interferences


                                               12
                                                         OGO

-------
     o   Safety
     o   Apparatus and materials

     o   Reagents and standards

     o   Calibration
     o   Quality Assurance and Quality Control

     o   Sample collection, preservation, and handling

     o   Sample preparation

     o   Extraction and concentration

     o   Extract cleanup

     o   Analysis

     o   System and laboratory performance

     o   Qualitative determination

     o   Analysis of complex samples

     o   Method performance

     o   References

      The  Quality Assurance  and Quality  Control  specifications of Method  1613 include a variety of
operations consistent with the 600 and 1600  series methods. The laboratory must perform a series of start
up tests to demonstrate their ability to perform the method adequately in a reference matrix  The start up
tests include spiking at least four aliquots of a reference matrix, typically reagent water or a clean solid
matrix,  with  all 17 native analytes, and carrying these  spiked  samples through the entire  analytical
procedure.  The recovery of the native and labeled analytes from the reference matrix are compared to the
specifications in the Method to determine if the laboratory is performing adequately.

      All samples and blanks are spiked with 15 isotopically labeled PCDDs/PCDFs prior to extraction.
Each extract is  spiked with an additional labeled compound prior to cleanup. The IS compounds spiked
prior to extraction are used for the quantification of the native analytes by isotope dilution, and result in
recovery-corrected data for the native analytes. The recoveries of these labeled compounds and the one
compound added prior to cleanup are compared to the Method specifications  for recovery. In this fashion,
the performance of the method is evaluated for each sample analysis.  The accuracy of the method, as
measured by the recovery of the labeled compounds, is re-evaluated by the  laboratory after 5-10 sample
analyses.

      The  method requires the analysis of blanks that are  extracted with each set of samples, up to  a
maximum of 20 samples per blank. Data from the analysis of the blank is provided along with the sample
data, and the levels of PCDDs/PCDFs in the sample are not adjusted for  the levels in blanks.  Limits are
placed on the amount of any analyte detected in the blank.  Given the ubiquitous nature and low toxicity of
OCDD, background levels of this analyte are  not a significant cause for concern.

      Guidance is given in the method on the use of field replicates and spiked samples in evaluating the
performance of the method and the laboratory.

      The performance of the laboratory and the entire analytical system is evaluated on a daily basis. Each
day on which samples  are extracted, the laboratory must prepare a reference matrix aliquot spiked with the
Precision and Recovery Standard.  This standard contains  the 17 native analytes as well as the 15 labeled
analogs, and is  carried through the entire analytical procedure, in a fashion similar to the start up test
aliquots. The recovery of the native and labeled compounds in this spiked sample are judged against the



                                               13        081

-------
requirements of the method, and may require re-extraction of all the associated samples if the spiked sample
does not meet the specifications.

      The resolution of the mass spectrometer is evaluated each 12 hours, to determine that it meets the
method specification for static resolving power of at least 10,000. The resolution of the gas chromatograph
is evaluated through the analysis of an isomer specificity standard which contains the PCDDs/PCDFs which
are most difficult to separate on the two GC columns specified in the method. The analysis of the GC
window defining mixture serves to verify that the laboratory is searching for the analytes in the appropriate
retention time windows.

      The calibration of the GCMS system is evaluated each 12 hours by the analysis of a calibration
verification standard.  If the results of this analysis fail to meet the specifications of the method, the
laboratory must analyze a  new set of  five initial calibration standards that pass the criteria for initial
calibration.

7.    FIELD VALIDATION STUDY

7.1   Scope

      Method 1613 was developed to support 304(m) studies in the Pulp  and Paper and Petroleum
Refining industries. However, prior to use in these industries 1613 was used in the National Sewage Sludge
Survey and  Pesticides industry  studies. Method 1613 was also used for analysis  of samples from the
Superfund Discharges to POTWs Study conducted under 304(m).  Three laboratories participated in the
field validation study. Table 4 provides a breakdown of the number of samples by Industrial Category and
Laboratory.  When discussing performance data from a specific laboratory the laboratory will be referred to
as Lab 1, Lab 2,... etc, rather than by name.

                                            TABLE 4

                       1613 Field Validation Study Samples by Industry and Lab

   Industry                   Lab 1          Lab 2            Lab 3         Total

   NSSS                       110            123              0             233

   Pesticides                    090               9

   Petroleum         .0              7             12              19

   Pulp & Paper                 0             32              0              32

   SF/POTW                   0             17              0              17

   Total                      108            188             12             310


7.1.1    National Sewage Sludge Survey

      In December of 1988, OW required a method for the analysis of dioxins and furans in municipal
sewage sludge in support of the National Sewage Sludge Survey (NSSS). This survey was being conducted to
provide data for the 40 CFR Part 503 regulation (CWA Section 405(d)).  The decision was made to use
Method 1613 (September 12 1988 revision) for the analysis of samples from the NSSS.  Special Analytical
Services (SAS) subcontract numbers  140 and 141 were awarded  to Twin  Cities  Testing  (TCT) and
ENSECO-California Analytical  (CAL) respectively.  As part of the SAS solicitation,  each solicited


                                               14
                                                           082

-------
laboratory submitted comments on the clarity and completeness of the method write-up.  These comments
were taken into account during subsequent method revisions.  TCT supplied extensive written comments
many of which were incorporated into future revisions. The comments from these laboratories are among
those cited earlier in this report

      Approximately eighty samples which had been stored frozen since sampling were split into two
batches and sent to TCT and CAL.  An additional 160 samples were taken and shipped to TCT and CAL.

7.1.2   Pesticide Formulators and Packagers Study

      Nine samples from a study of four pesticide formulator and packager facilities were sent to TCT for
analysis, in support of CWA Section 402 effluent guideline development.

7.1.3   304(m) Industry Studies

      Sixty-eight samples from three CWA Section 304(m) industries were analyzed using Method 1613
during FY89 studies.

      As the result of reports from Canada of the discovery of PCDDs and PCDFs in the effluents from
petroleum refinery operations,  a  study of US refineries  was undertaken by OW in  August of 1989.
Approximately 35 samples of effluents and sludges were shipped to TCT and Triangle Laboratories for
analysis.

      Thirty-two samples were taken in support of the ITD Pulp and Paper Study.  These samples were
analyzed by TCT over the course of six months, starting in the spring of 1989.

      Seventeen analyses for PCDDs and PCDFs were  performed by TCT in support of studies of
discharges from Superfund sites to POTWs.

7.2  Method and Laboratory Performance

      Prior to sample analysis using Method 1613, a laboratory must complete the start-up tests (Initial
Precision and Recovery,Method Section 8.2).  The laboratory must also analyze an ongoing precision and
accuracy aliquot  (Method Section  14.5) with each sample set.  The  results of these analyses must display
acceptable precision and recovery, as defined in the method.

      Table 5 presents the results of the start-up tests (IPR) performed by each laboratory. Table 6 details
the results of the Ongoing Precision and Accuracy (OPR) analyses by each laboratory.  The data presented
in these two tables are not differentiated by laboratory. Both tables divide the compounds into two  groups.
The first group contains the 15 native 2,3,7,8-substituted PCDDs and PCDFs that are quantified by  isotope
dilution.  The second group contains the 16 labeled compounds and the two native compounds that are
quantified by internal standard.

      As can be seen in Table 5, the initial precision of and recovery data indicate that all the laboratories
are able to achieve acceptable recoveries of the native and labeled compounds spiked into the reference
matrices. For 2,3,7,8-TCDD, the mean recovery across all labs was 88.5%. The mean recoveries of all other
native compounds were even higher.  The fact that the mean recovery of OCDD was above 100% is due to
the great difficulty in  eliminating background levels of this  compound.  The data reported in these tables
have not been adjusted  for the levels of OCDD found in the blanks. The standard deviatons of the native
compound recoveries are also relatively low.  No native compound had a standard deviation across all labs of
greater than 20%.
                                               15         063

-------
      The recoveries of the labeled compounds in Table 5 are generally lower than those of the native
compounds. The recoveries of 2,3,7,8-TCDD-C13 and 2^,7,8-TCDF-C13 are on average around 50%.  The
Cl3  labeled TCDD standard that is added after extraction but prior to cleanup has a similar average
recovery. This result confirms the belief of the laboratories that the multi-layered alumina column used for
extract cleanup may have caused the preferential loss of the tetrachlorinated compounds. As discussed
earlier, this cleanup column is being replaced in the latest revision of the method.

      The results in Table 6 for the ongoing precision and recovery data across all laboratories indicate
similiar trends.  The native compound recoveries were all above 90%, and the labeled compound recoveries
were all above 50%, though lowest for the tetrachlorinated compounds.

      In general,  the EPR and OPR  recoveries  from the  participating  laboratories  appear  to be
homogeneous.

      By comparing the recoveries of the native compounds with their labeled counterparts, one can see the
value of isotope dilution for quantification.  Despite losses of some labeled compounds as high as 50%
during the extract cleanup, the precision of the results for the native compounds, measured as the standard
deviation of the recoveries, is considerably better than for the two native compounds quantified by internal
standard techniques.

7.2.5   Field Validation Study Results

      The data for the precision and recovery of the labeled compounds spiked into the field samples are
presented by Industry and Matrix in Table 7. As these were actual field samples from a wide variety of
sources, no "true" values existed for the concentrations of the native PCDDs and PCDFs. Therefore, Table 7
contains no data for native compound recoveries.

      As can be seen in Table 7, there were significant differences in the labeled analog recoveries in the
samples from the various industries. However, the recoveries generally fell within the method specifications
despite these differences. The recoveries are generally higher in the solid matrices.

      The results from  the analyses  of  these field samples  were  used to develop quality  control
specifications for the revised Method, as discussed below.

7.3   Development of Quality Control Specifications

      A major objective of this method evaluation was to produce performance specifications from the data
collected to date and apply these specifications to the next method  revision in. order to better gauge
laboratory performance.

      Quality control  specifications were  calculated  by  constructing  statistical prediction intervals for
future observations of a quantity of interest using estimates  determined in this study. Recovery limits are
calculated using the results of a variance component analysis of the percent recoveries on three subsets of
data;  IPR samples, OPR samples and  EPA (field) samples.   The inter- and  intralaboratory variance
components of the logarithms  of the recoveries were estimated (Sg2 and SA2), along with the log mean
response (m) by the Type 1 method, using PROC VARCOMP (SAS). Details of the variance components
analysis are given in Appendix B.

      The percent recoveries of these compounds have been assumed to follow a log normal distribution
throughout the analysis described in this section.  The log normal distribution has been  frequently and
effectively applied  to model  pollutant concentrations,  and agrees with the physical interpretation of non-
negative concentration values.  Limits derived from this assumption are always non-negative.  Descriptive
                                               16            064

-------
and summary statistics  calculated for these  data  support the assumption of  log  normality for most
compounds. Although some compounds appear to be normally distributed, log normal distributions were
assumed in order to attain non-negative limits for all specifications.

      Compound-specific performance specifications have been determined at a 95% confidence level.
Using these specifications, each tested compound would have a 5% chance of falling outside QC limits.

      The specifications resulting from the three-laboratory evaluation are detailed below.  Statistical
methods are discussed in Appendices A through F.

7.3.1    Data Screening

      The data utilized  in this study were screened for outliers both on a laboratory  and individual point
basis.  The labeled analog recoveries in field samples analyzed at Lab 1 were on the order of 40-50% lower,
in similar samples, than those for Labs 2 and 3. This tended to increase the spread of the data and create the
appearance of a separate population of recovery data. Figure 3 illustrates this effect. The inclusion of data
from Lab  1 resulted in percent recovery specifications that  were extremely wide and not representative of
current laboratory performance. The lower recoveries in these samples from Lab 1 were due to problems
associated with the earliest version of Method 1613, particularly regarding the extraction procedures.

      Therefore, data from Lab 1 were not utilized for the development of recovery specifications for
labeled analogs in field samples. The remaining recovery values were then screened individually for outliers.
A robust quantile method based on the median and the interquartile range was applied, as described in
Appendix A.

7.3.2    Instrument Performance Specifications

7.3.2.1  Calibration Linearity

      In order to calculate the concentration of each compound  in a sample, the measured instrument
response is compared to the response obtained from a series of calibration samples analyzed at known
concentrations (Method 1613, Section  7). This allows an evaluation of instrument response across a range
of concentrations of the  analytes. In this study, the coefficient of variation (CV) of the instrument response
achieved by the participating laboratories was examined. A linearity specification was calculated for each
compound and the existing data were tested against this specification.  Table 8 details this analysis and
Appendix E describes the statistical approach.  In general  the specifications obtained compare favorably
with the existing CV specification of 20%  for compounds quantitated by isotope dilution  (Method 1613,
Section 7.5.5) and 35% for compounds  quantitated by internal standard (Method 1613, Section 7.6.1.2).

7.3.3    Calibration Verification

      Method 1613 Section 14.3  addresses calibration  verification.  Concentration limits  for calibration
verification (VER) are required to insure that acceptable instrument response (measured as concentration)
is maintained over time. Specifications for upper and  lower concentration limits were developed using
laboratory-submitted VER analyses. These specifications are detailed in Table 9.  Submitted data were then
tested against the developed specifications to determine the percentage of compounds that would fall
outside the limits.  Table 7  of Method  1613 will  be updated to include these specifications.  The statistical
approach is detailed in Appendix C.
                                                17
                                                        085

-------
7.3.4    Relative Retention Time

      The relative retention time (RRT) of the native compound versus the labeled analog is a critical
specification for  isotope  dilution  quantitation.   Current  relative  retention time (RRT) specifications
contained in Table 2 of Method 1613 were developed prior to applying isotope dilution to the analysis of the
2,3,7,8-substituted dioxin and furan isomers, and these specifications were extrapolated from the RRT of
chlorinated compounds in Method 1625. The extrapolated RRT specifications for native compounds of
approximately 0.999 -1.001 have proven to be too narrow. Method 1613 looks at a much narrower range of
analytes, and uses a slower temperature program for GC.  As a result, the analytes are spread out across the
chromatogram further, and the differences between the retention times of the native and labeled compounds
increase.  Therefore, the  RRT limits will be adjusted according to the specfications in Table 10  of this
report, and will vary by compound.  Appendix F details the statistical analyses used in developing these
specifications.

7.3.5    Initial Precision and Accuracy

      Method 1613 requires the analysis of four initial precision  and accuracy  aliquots (EPR, Method
Section 8.2) prior to the analysis of any samples.  The arithmetic average (mean) and standard deviation (sd)
of the four results generated for each labeled and native compound are computed per  the Method.  The
mean and sd must meet the limits specified in the Method in order to insure acceptable precision and bias
for subsequent analyses.  As part of this study, limits were  constructed for accuracy and precision. Table
11A details the EPR accuracy limits, expressed  as mean percent recovery, and Appendix C provides  the
derivation of these limits.  Table 11B provides the the IPR precision limits expressed as sd of the recoveries,
and Appendix D provides the derivation of these limits.  The existing limits in the Method will be updated to
relfect the limits generated in this study.

7.3.6    Ongoing Precision and Accuracy

      In each batch of samples analyzed by the laboratory, one ongoing precision and accuracy aliquot is
analyzed (OPR, Method 1613 Section 14.5). Mean recovery is evaluated against specified limits to ensure
analytical control. Mean percent recovery specifications developed during this study are presented in Table
12. The derivation of these limits is given in Appendix C  The existing limits  in the Method will be updated
to relfect the limits generated in this study.

7.3.7    Labeled Compound Recovery in EPA Samples

      The data for the recoveries of labeled compounds  from the various field validation samples  are
presented in Table 13.  This table contains the data from all three laboratories, including those data from
Lab 1 discussed above which exhibited significantly lower recoveries. Table 14 contains the recovery data for
only Lab 2 and Lab 3.  As can be seen by comparing these two tables, the means,  medians, and standard
deviations of the recoveries increased significantly when the data from Lab 1 were  screened out of the
population.  The differences  in the  extraction and analyses  procedures that  these data represent  are
indicative of the improvements made to Method 1613 during the field vlaidation studies.

      Table 15 contains the limits generated for recovery of labled compounds from field samples. These
limits were generated from the field samples results for Lab 2 and Lab 3 only, as discussed above.  Generally
speaking, the ranges of recoveries for each compound  derived from these data  are within the current
specifications (25-150%) of Method 1613, with some predicted recoveries above the current limits.

      These limits for labled compound recovery in field samples will be incorporated into the Method, and
re-evaluated as more and more data are collected by the Agency using the latest version of the procedures.
                                                18


                                                            06G

-------
8.    Conclusions                                                                 "	'

      Method 1613 was developed by OW for the analysis of PCDDs and PCDFs in effluents in other
matrices such as solids and sludges.  The data presented here demonstrate that the method is sufficiently
rugged to be used by a variety of laboratories. The precision and accuracy of the method meet the needs of
the Agency for a survey method for dioxins and furans. The method contains an extensive QA/QC program
that allows the data user to evaluate the quality of individual sample results. This QA/QC program was built
into the method from the start, not simply added on to an existing method.  The benefits of this approach
were evident during the development of the final method, where data on labeled compound recoveries were
instrumental in evaluating various sample extraction and clean up procedures.

      For 2,3,7,8-TCDD and 2,3,7,8,-TCDF, Method 1613 is approximately 200 times more sensitive than
Method 613, the  existing method promulgated under Section 304(h).  Validation data for these two
compounds from the analyses of over 300 field samples indicate acceptable levels of precision and bias for
these compounds. Therefore, Method 1613 is proposed as a nationwide Alternative Test  Procedure under
Section 304(h) for the analysis of 2,3,7,8-TCDD and 23,7,8-TCDF.

      Given that no method for the analysis of the penta- through octachlorinated dioxins and furans has
been proposed or promulgated under Section 304(h), Method 1613 is proposed as the nationwide primary
test procedure for these analytes.

      Finally, the results of the evaluation presented here and the public comments received during the
course of this evaluation have been used to revise the July 1989 version of Method 1613 for nationwide use
in support of effluent guideline development and NPDES monitoring.

References

1.     "Ambient Water Quality Criteria for 2,3,7,8 - Tetrachloro-dibenzo-p-dioxinn, USEPA, OWRS,
      Washington DC 20460, EPA 440/5-84-007, February 1984.

2.     Tondeur, Yves, "Method 8290: Analytical Procedures and Quality Assurance for Multimedia Analysis
      of Polychlorinated Dibenzo-p-dioxins and Dibenzorurans by High-Resolution Gas
      Chromatography/High-Resolution Mass Spectrometry," USEPA, EMSL-Las Vegas, Nevada, June
      1987.

3.     "Measurement of 23,7,8-Tetrachlorinated Dibenzo-p-dioxin (TCDD) and 23,7,8-Tetrachlorinated
      Dibenzofuran (TCDF) in Pulp, Sludges, Process Samples and Wastewaters from Pulp and Paper
      Mills", Wright State University, Dayton OH  45435, June 1988.

4.     "NCASI Procedures for the Preparation and Isomer Specific Analysis of Pulp and Paper Industry
      Samples for 2,3,7,8-TCDD and 23,7,8- TCDF", National Council of the Paper Industry for Air and
      Stream Improvement, 260 Madison Av, New York NY  10016, Technical Bulletin No. 551, Pre-
      release Copy, July 1988.

5.     "Analytical Procedures and Quality Assurance Plan for the Determination of PCDD/PCDF in Fish",
      U.S. Environmental Protection Agency, Environmental Research Laboratory, 6201 Congdon Blvd.,
      Duluth MN 55804, April 1988.

6.     Yves Tondeur, "Proposed GC/MS Methodology for the Analysis of PCDDs and PCDFs in Special
      Analytical Services Samples", Triangle Laboratories, Inc., 801-10 Capitola Dr, Research Triangle
      Park NC 27713, January 1988; updated by personal communication September 1988.

7.     Lamparski, L.L., and Nestrick, TJ., "Determination of Tetra-, Hexa-, Hepta-, and
      Octachlorodibenzo-p-dioxin Isomers in Paniculate Samples at Parts per Trillion Levels",  "Anal.
      Chem." 52,2045-2054 (1980).
                                              19

                                                        08 V

-------
8.    Lamparski, L.L., and Nestrick, TJ., "Novel Extraction Device for the Determination of Chlorinated
      Dibenzo-p-dioxins (PCDDs) and Dibenzofurans (PCDFs) in Matrices Containing Water", Personal
      Communication, July 1988.

9.    Patterson, D.G., et. al. "Control of Interferences in the Analysis of Human Adipose Tissue for 2,3,7,8-
      Tetrachlorodibenzo-p-dioxin", "Environ. Toxicol. Chem.," 5,355-360 (1986).

10.    Stanley, John S., and Sack, Thomas M., "Protocol for the Analysis of 2,3,7,8-Tetrachlorodibenzo-p-
      dioxin by High-Resolution Gas Chromatography/High-Resolution Mass Spectrometry", USEPA,
      Environmental Monitoring Systems Laboratory, Las Vegas  NV 89114, EPA 600/4-86-004, January
      1986.

11.    "Method 613 - 23,7,8-Tetrachlorodibenzo-p-dioxin", 40 CFR136 (49 FR 43234), October 26,1984.

12.    "Guidelines and Format for EMSL-Cincinnati Methods", USEPA, EMSL, Cincinnati, OH 45268,
      EPA-600/8-83-020, August 1983.

13.    "Guidelines for Selection and Validation of USEPA's Measurement Methods", USEPA, ORD,
      OADEMQA, Draft, August 1987.

14.    Eynon, B.P., Maxwell, C and Valder, A., "Interlaboratory Validation of U.S. Environmental
      Protection Agency Method 1625A", EPA Contract 68-01-6192, July  1984.

15.    Hoaglin, D.C, Mosteller, F., and Tukey, J.S., "Understanding Robust and Exploratory Data Analysis",
      John Wiley and Sons, New York, pp 37-39,1983.

16.    Milliken, G.A., Johnson, D.E., "Analysis of Messy Data", Lifetime Learning Publications, California,
      pp 238-241,1984.

17.    Wilson, A.L., "Approach for Achiecing Comparable Results from a Number of Laboratories", The
      Analyst, Vol. 104 No. 1237, p 273,1979.

18.    "USEPA Development Document for Existing Source Pretreatment Standards for the Electroplating
      Point Source Category", EPA Document 440/1-79/003, August 1979.
                                              20
                                                          088

-------
U S E P A
           INDUSTRIAL  TECHNOLOGY
                 METHOD 1613 PERFORMANCE EVALUATION
                   DIVISION
                              TABLE 3
                      METHOD DETECTION
      LIMIT
COMPOUND

2378-TCDD
12378-PECDD
123478-HXCOO
123678-HXCDD
1234678-HPCDD
OCDD
2378-TCOF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCDF
1234678-HPCDF
1234789-HPCDF
COMPOUND
^ 123789-HXCDD
X-C OCDF
NUMBER
DBS
6
6
7
6
5
7
6
6
6
6
7
6
6
7
3
NUMBER
OBS
5
6
                   QUANTITATION=ISOTOPE DILUTION
                           MEAN
                           26.3
                           27.6
                           31.3
                           26.1
                           52.3
                          195.6
                           27.2
                           27.7
                           30.6
                           38.8
                           28.0
                           25.4
                           25.9
                           67.9
                           70.2
 STD
 DEVIATION

    0.8
    2.0
    1.7
    2.6
    4.9
   97.4
    1.4
    1.7
    1.7
    2.4
    0.5
    1.3
    1.9
    6.0
                                                T-VALUE
 3.365
 3.365
 3.143
 3.365
 3.747
 3.143
 3.365
 3.365
 3.365
 3.365
 3.143
 3.365
 3.365
 3.143
 6.965
                   QUANTITATION=INTERNAL STANDARD 	

                          MEAN    STD          T-VALUE
                          22.4
                          75.8
STD
DEVIATION

   2.1
   8.2
3.747
3.365
 METHOD
 DET LIMIT

     2.7
     6.7
     5.3
     8.7
    18.2
   306.2
     4.7
     5.8
     5.7
     8.1
     1.7
     4.3
     6.4
    19.0
    20.3
METHOD
DET LIMIT

   8.0
  27.6

-------
USEPA  INDUSTRIAL  TECHNOLOGY
                 METHOD 1613 PERFORMANCE EVALUATION
                   DIVISION
COMPOUND

2378-TCDD
12378-PECDD
123478-HXCDD
123678-HXCDO
1234678-HPCDD
OCDD
2378-TCDF
12378-PECOF
23478-PECDF
123470-HXCDF
123678-HXCOF
123789-HXCDF
234678-HXCDF
1234678-HPCDF
1234789-HPCDF
COMPOUND
o
^- 2378-TCDD-C13
__• 2378-TCDD-CL37
v— 12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
123789-HXCDD
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
NO OF
OBSERVATINS
21
19
21
21
22
18
22
20
21
22
21
18
22
19
20
NO OF
OBSERVATINS

22
22
22
21
20
22
22
22
20
21
21
21
20
22
17
22
22
22
                              TABLE 5
   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                        SUMMMARY STATISTICS
                         ALL LABS COMBINED

                   QUANTITATION=ISOTOPE DILUTION
                          MEAN
 88.5
 96.6
 92.2
 93.9
 92.2
102.
 91.
 96.
 95.5
 97.7
 92.3
MEDIAN
                            A
                            .3
                            .5
                          97.7

                          W.2
                          96.8
  85.0
  97.0
  93
  90
  93
 102
  94.2
  98.0
  95.0
  96.2
  92.0
  98.0
  98.0
 100.0
  98.5
                   qUANTITATION=INTEHNAL STANDARD

                           MEAN    MEDIAN
                           56.4
                           57.0
                           76.6
                           72.5
                           72.6
                           91.7
                           82.5
                           80.6
                           50.3
                           70.1
                           72.9
                           63.4
                           67.8
                           72.7
                           69.4
                           67.6
                           73.6
                           97.8
   67.0
   56.5
   71.1
   66.2
   68.0
   90.0
   83.8
   83.5
   58.5
   72.0
   72.0
   63.1
   63.3
   69.9
   67.1
   62.3
   75.5
  103.6
STD
DEVIATION
14.4
5.0
9.2
10.7
12.9
10.3
19.6
5.6
5.4
13.6
5.9
4.5
11.9
4.8
7.5
STD
DEVIATION
22.0
25.8
30.3
15.4
17.5
17.6
13.7
25.3
19.7
26.3
25.2
13.4
15.3
16.1
15.8
14.0
16.9
24.2
MINIMUM

66.0
87.0
76.0
78.0
72.0
82.0
55.0
85.0
86.0
73.0
82.0
89.0
79.0
89.0
61.0
MINIMUM
19.0
21.0
26.0
54.0
37.4
64.0
60.0
34.9
16.0
36.0
38.0
46.0
47.0
49.0
49.0
48.0
46.6
56.0
MAXIMUM

118.0
105.0
105.0
114.0
120.0
124.0
125.0
104.0
105.0
123.0
103.0
108.0
128.0
108.0
108.0
MAXIMUM
86.2
. 95.1
140.0
101.0
97.7
133.0
113.0
120.0
77.3
129.0
133.0
98.0
99.0
106.0
109.0
95.8
101.0
150.0

-------
U S E P A
INDUSTRIAL  TECHNOLOGY
      METHOD 1613 PERFORMANCE EVALUATION
                                                     DIVISION
                              TABLE 6
   ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                        SUMMMARY STATISTICS
                         ALL LABS COMBINED
COMPOUND

2378-TCDD
12378-PECDD
123478-HXCDD
123678-HXCDD
1234678-HPCDD
OCDD
2378-TCDF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCOF
1234678-HPCDF
1234789-HPCDF
------- muuti
NO OF
OBSERVATINS
48
47
45
48
42
47
47
44
46
47
47
45
47
48
47
A i M i Aun-
MEAN

90.6
91.5
92.1
92-. 8
91.7
95.1
92.5
92.1
92.7
93.7
91.0
90.9
91.0
95.7
95.1
Aauiurc uj
MEDIAN

89.5
93.0
90.5
90.5
94.0
96.0
92.5
92.0
92.5
95.0
90.0
93.0
93.0
96.0
96.0
STD
DEVIATION
13.8
9.2
11.2
13.1
7.2
8.7
12.0
7.6
8.6
9.4
9.3
7.8
9.4
9.8
10.3
MINIMUM

58.0
76.0
67.0
70.0
80.0
80.0
66.0
77.0
74.0
77.0
74.0
75.0
70.0
75.0
80.0
MAXIMUM

125.0
115.0
115.0
125.0
105.0
116.0
125.0
108.0
111.0
112.0
110.0
105.0
110.0
118.0
120.0

COMPOUND
o
is. 2378-TCDD-C13
2378-TCDD-CL37
12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
123789-HXCDD
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
NO OF MEAN MEDIAN STD
OBSERVATINS
50
49
50
46
48
48
49
49
50
49
47
50
49
49
35
50
48
44

55.2
57.4
69.9
73.1
75.2
88.6
69.4
55.8
55.6
65.1
64.4
68.4
70.6
69.5
69.8
63.4
66.2
93.2

58.0
60.0
68.0
71.0
71.0
88.0
66.0
54.0
57.5
62.0
59.0
70.5
70.0
70.0
70.0
61.5
68.0
92.0
DEVIATION
22.8
24.2
28.0
13.4
15.1
12.5
16.8
18.4
21.0
23.5
23.9
15.1
14.8
14.5
14.0
13.5
17.4
9.8
MINIMUM

11.0
9.6
10.0
49.0
45.0
69.0
36.0
11.0
11.0
35.0
5.7
37.0
37.0
37.0
42.0
38.0
30.9
77.5
MAXIMUM

96.0
113.0
135.0
101.0
109.0
115.0
108.0
97.0
91.0
115.0
115.0
103.0
101.0
100.0
96.0
93.0
97.0
119.5

-------
                             USEPA  INDUSTRIAL  TECHNOLOGY
                                              METHOD  1613  PERFORMANCE  EVALUATION
                                                           DIVISION
                                                           TABLE  7
                                         PRECISION AND  RECOVERY OF  LABELED
                                                     SUMMARY STATISTICS
                                                     BY  INO CATC MATRIX)
                                                     COMPOUNDS
                              IND_CAT=PULP +  PAPER  MATRIX=AQUEOUS QUANTITATION=INTERNAL STANDARD
                       COMPOUND
                       2378-TCDD-C13
                       12378-PECDD-C13
                       123478-HXCDD-C13
                       123678-HXCDD-C13
                       1234678-HPCDD-C13
                       OCDD-C13
                       2378-TCDF-C13
                       12378-PECDF-C13
                       23478-PECDF-C13
                       123478-HXCDF-C13
                       123678-HXCDF-C13
                       123789-HXCDF-C13
                       1234678-HPCDF-C13
                       1234789-HPCDF-C13
                                            NO OF
                                            OBS
                                                     MEAN
                                                            MEDIAN
1 31.0
1 36.0
1 41.0
1 32.0
1 24.0
1 20.0
1 26.0
31.0
34.0
34.0
31.0
33.0
20.0
1 22.0
31.0
36.0
41.0
32.0
24.0
20.0
26.0
31.0
34.0
34.0
31.0
33.0
20.0
22.0
                                                STD
                                                DEVIATION
                                                                                   MINIMUM    MAXIMUM
31.0
36.0
41.0
32.0
24.0
20.0
26.0
31.0
34.0
34.0
31.0
33.0
20.0
22.0
31.0
36.0
41.0
32.0
24.0
20.0
26.0
31.0
34.0
34.0
31.0
33.0
20.0
22.0
                               IND_CAT=PULP  + PAPER MATRIX=SOLID QUANTITATION=INTERNAL STANDARD
o
                       COMPOUND
2378-TCDD-C13
2378-TCDD-CL37
12378-PECDD-C13
123478-HXCDO-C13
123678-HXCDD-C13
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCOF-C13
                                           NO OF
                                           OBS
                                                    MEAN
                                                            MEDIAN
                                                STD
                                                DEVIATION
                                                                                   MINIMUM
                                                                                              MAXIMUM
31
31
31
31
31
31
31
31
31
31
31
31
31
31
31
31
56.2
57.8
85.5
80.5
72.5
58.1
48.5
51.8
73.6
85.3
72.9
68.7
70.7
66.8
55.2
60.6
59.0
60.0
84.0
79.0
74.0
57.0
46.0
52.0
71.0
87.0
70.0
68.0
70.0
67.0
55.0
61.0
14.7
15.1
13.2
10.3
11.6
8.7
7.6
14.4
10.2
12.4
9.7
10.6
9.8
8.7
8.8
8.7
27.0
29.0
63.0
66.0
51.0
41.0
33.0
28.0
53.0
63.0
58.0
52.0
55.0
45.0
40.0
39.0
79.0
83.0
113.0
113,0
112.0
90.0
72.0
79.0
92.0
110.0
104.0
102.0
103.0
87.0
85.0
89.0

-------
       U S E P A
                  INDUSTRIAL  TECHNOLOGY
                        METHOD 1613 PERFORMANCE EVALUATION
                             DIVISION
                                     TABLE  7
                   PRECISION AND RECOVERY OF LABELED
                               SUMMARY STATISTICS
                                BY IND CATC MATRIX)
                       COMPOUNDS
	 IND_CAT=PESTICIDES MATRIX=AQUEOUS qUANTITATION=INTERNAL STANDARD 	

 COMPOUND             NO OF    MEAN    MEDIAN    STD           MINIMUM    MAXIMUM
 2378-TCDD-C13
 2378-TCDD-CL37
 12378-PECDD-C13
 123478-HXCDD-C13
 123678-HXCDD-C13
 1234678-HPCDD-C13
 OCDO-C13
 2378-TCDF-C13
 J2378-PECDF-CJ3
 23478-PECDF-C13
 123478-HXCDF-C13
 123678-HXCDF-C13
 123789-HXCDF-C13
 234678-HXCDF-C13
 1234678-HPCDF-C13
 1234789-HPCDF-C13
                      NO OF
                      OBS
                  STD
                  DEVIATION
8
8
8
8
8
8
8
8
8
8
8
8
7
7
8
8
41.0
58 .'3
60.4
54.8
54.9
51.5
48.0
38.8
49.9
54.0
48.0
53.3
61.0
58.4
48.4
53.1
43.0
59.0
64.5
57.5
54.0
52.0
48.0
41.0
53.5
58.i
52.0
56.0
59.0
55 0
50.5
54.0
18.9
17.1
24.1
20.1
21.1
23.1
20.6
18.5
20.1
20.9
18.0
20.4
8.3
7.9
22.1
22.8
11.0
32.0
8.0
9.0
15.0
17.0
13.0
10.0
8.0
7.0
8.0
9.0
55.0
53.0
8.0
13.0
71.0
82.0
69.0
75.0
87.0
90.0
74.0
71.0
75.0
79.0
69.0
62.0
79.0
75.0
87.0
90.0
          IND_CAT=PESTICIDES MATRIX=SOLID C|UANTITATION=INTERNAL STANDARD
 COMPOUND
 2378-TCDD-C13
 2378-TCDD-CL37
 12378-PECDD-C13
 123478-HXCDD-C13
 123678-HXCDD-C13
 1234678-HPCDD-C13
 OCDD-C13
 2378-TCDF-C13
 12378-PECDF-C13
 23478-PECDF-C13
 123478-HXCDF-C13
 123678-HXCDF-C13
 123789-HXCDF-C13
 234678-HXCDF-C13
 1234678-HPCDF-C13
 1234789-HPCDF-C13
                      NO OF
                      OBS
MEAN
        MEDJAN
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
76.0
78.0
64.0
85.0
59.0
71.0
68.0
69.0
60.0
54.0
63.0
68.0
66.0
67.0
58.0
67.0
76.0
78.0
64.0
85.0
59.0
71.0
68.0
69.0
60.0
54.0
63.0
68.0
66.0
67.0
56.0
67.0
                  STD
                  DEVIATION
                               MINIMUM
                                76.0
                                78.0
                                64.0
                                85.0
                                59.0
                                71.0
                                68.0
                                69.0
                                60.0
                                54.0
                                63.0
                                68.0
                                66.0
                                67.0
                                58.0
                                67.0
MAXIMUM
 76.0
 78.0
 64.0
 85.0
 59.0
 71.0
 68.0
 69.0
 60.0
 54.0
 63.0
 68.0
 66.0
 67.0
 58.0
 67.0
o

-------
                            USEPA   INDUSTRIAL  TECHNOLOGY
                                             METHOD  1613  PERFORMANCE EVALUATION
                                      DIVISION
                                                         TABLE  7
                                        PRECISION AND RECOVERY OF  LABELED
                                                    SUMMARY STATISTICS
                                                    BY  IND CAT(MATRIX)
                                COMPOUNDS
                     	  IND_CAT=PETROLEUM  REF. MATRIX=AQUEOUS QUANTITATION=INTERNAL STANDARD 	

                      COMPOUND             NO  OF    MEAN    MEDIAN    STD          MINIMUM    MAXIMUM
                      2378-TCDD-C13
                      2378-TCDD-CL37
                      12378-PECDD-C13
                      123478-HXCDD-C13
                      123678-HXCDD-C13
                      1234678-HPCDO-C13
                      OCDD-C13
                      2378-TCDF-C13
                      12378-PECDF-C13
                      23478-PECDF-C13
                      123478-HXCDF-C13
                      123678-HXCDF-C13
                      123789-HXCDF-C13
                      234678-HXCDF-C13
                      1234678-HPCDF-C13
                      1234789-HPCDF-C13
NO OF
OBS
STD
DEVIATION
17
17
17
17
17
17
17
17
17
17
17
17
17
15
17
17
62.1
56.8
62.7
76.4
70.5
59.6
46.6
69.0
64.1
64.0
63.3
64.3
60.7
69.0
52.6
52.8
62.0
59.1
63.1
79.9
76.9
64.1
51.6
62.2
59.2
60.0
68.0
68.3
58.4
69.4
58.3
53.0
18.8
19.5
11.9
27.6
14.2
16.6
16.4
22.2
12.4
14.3
13.5
14.8
15.8
10.4
14.4
15.0
23.6
21.0
34.3
41.7
42.8
17.5
10.0
34.7
39.5
37.3
38.7
39.7
34.9
52.0
11.4
11.9
103.0
88.9
82.5
151.0
92.4
84.7
68.5
112.0
95.1
101.0
81.6
86.7
90.2
87.5
67.0
74.0
o
                     	 IND_CAT=PETROLEUM REF. MATRIX=OTHER QUANTITATION=INTERNAL STANDARD 	

                     COMPOUND             NO OF    MEAN    MEDIAN    STD          MINIMUM    MAXIMUM
                     2378-TCDD-C13
                     2378-TCDD-CL37
                     12378-PECDD-C13
                     123678-HXCDD-C13
                     123678-HXCDD-C13
                     1234678-HPCDD-C13
                     OCDD-C13
                     2378-TCDF-C13
                     12378-PECOF-C13
                     23478-PECDF-C13
                     123478-HXCDF-C13
                     123678-HXCDF-C13
                     1237tt9-HXCDF-C13
                     234678-HXCDF-C13
                     1234678-HPCDF-C13
                     1234789-HPCDF-C13
NO OF
OBS
STD
DEVIATION
16
16
16
16
16
16
16
16
16
16
15
16
16
14
16
16
51.2
45.3
53.3
68.5
66.7
58.1
38.3
60.0
56.6
56.0
70.9
66.6
60.2
64.2
55.5
55.2
48.6
42.2
43.4
68.2
64.3
55.6
33.9
56.4
57.0
55.6
66.6
63.3
50.3
63.5
54.3
51.4
24.6
22.8
26.7
26.6
26.7
19.0
15.2
19.4
21.6
23.9
17.9
25.9
21.1
15.5
22.5
22.1
7.6
5.8
7.1
4.4
7.3
27.2
14.6
31.3
18.5
3.5
46.4
3.9
22.7
42.4
7.0
24.1
95.1
83.8
98.1
103.0
103.0
88.0
66.0
104.0
102.0
106.0
96.0
106.0
91.0
65.4
84.0
86.0

-------
                            U S E P A
INDUSTRIAL  TECHNOLOGY
      METHOD 1613 PERFORMANCE EVALUATION
                                                                                DIVISION
                                                          TABLE  7
                                        PRECISION  ANO  RECOVERY OF  LABELED
                                                    SUMMARY  STATISTICS
                                                    BY  IND  CAT(MATRIX)
                                    COMPOUNDS
                     	  IND_CAT=PETROLEUM  REF.  MATRIX=SOLID  QUANTITATION=INTERNAL STANDARD

                      COMPOUND
                      2378-TCDD-C13
                      2378-TCDD-CL37
                      12378-PECDD-C13
                      123478-HXCDD-C13
                      123678-HXCDD-C13
                      1234678-HPCDD-C13
                      OCDD-C13
                      2378-TCDF-C13
                      12378-PECDF-C13
                      23476-PECDF-C13
                      123478-HXCDF-C13
                      123678-HXCDF-C13
                      123789-HXCDF-C13
                      234678-HXCDF-C13
                      1234678-HPCDF-C13
                      1234789-HPCDF-C13
NO OF
OBS
2
2
1
2
1
1
2
2
2
2
2
2
2
2
2
2
MEAN

51.3
46.8
68.4
76.8
65.0
41.0
45.3
55.6
64.4
59.5
65.7
57.6
56.1
60.1
46.8
38.9
MEDIAN

51.3
46.8
68.4
76.8
65.0
41.0
45.3
55.6
64.4
59.5
65.7
57.6
56.1
60.1
46.8
38.9
STD
DEVIATION
11.0
7.4
,
14.4
.
.
30.7
12.2
10.7
2.1
13.2
0.6
11.1
9.8
14.4
10.0
MINIMUM

43.5
41.6
68.4
66.7
65.0
41.0
23.6
47.0
56.9
58.0
56.4
57.2
48.3
53.2
36.6
31.8
MAXIMI

59.0
52.0
68.4
87.0
65.0
41.0
67.0
64.3
72.0
61.0
75.0
58.0
64.0
67.0
57.0
46.0
o

-------
                           USEPA  INDUSTRIAL  TECHNOLOGY  DIVISION
                                            METHOD 1613 PERFORMANCE EVALUATION

                                                         TABLE 7
                                       PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                                   SUMMARY STATISTICS
                                                    BY IND_CAT(MATRIX)

                           IND_CAT=SUPERFUND/POTM MATRIX=AQUEOUS QUANTITATIOH=INTERNAL STANDARD
                     COMPOUND             NO OF    MEAN    MEDIAN    STD          MINIMUM    MAXIMUM
                                          DBS                        DEVIATION

                     2378-TCDD-C13         14      23.7     18.5       17.3         3.0       54.0
                     2378-TCDD-CL37        14      36.8     36.0       16.5         7.0       64.0
                     12378-PECDD-C13       14      43.6     37.5       21.5        12.0       75.0
                     123478-HXCDD-C13      14      52.8     59.5       23.4        20.0       82.0
                     123678-HXCDD-C13      14      51.6     61.5       23.0        20.0       78.0
                     1234678-HPCDD-C13     14      52.1     68.0       27.6        14.0       80.0
                     OCDO-C13              14      46.9     59.0       26.0        11.0       73.0
                     2378-TCDF-C13         14      24.7     19 0       16.8         4.0       51.0
                     12378-PECDF-C13       14      43.8     41.5       22.8        14.0       79.0
                     23476-PECDF-C13       14      52.6     51.5       23.4        19.0       88.0
                     123478-HXCDF-C13      14      45.8     52.0       20.9        15.0       70.0
                     123678-HXCDF-C13      14      48.7     58.5       23.0        18.0       77.0
                     123789-HXCDF-C13      14      50.8     59.5       21.9        20.0       74.0
                     234678-HXCDF-C13       9      69.8     70.0       11.4        56.0       93.0
                     1234678-HPCDF-C13     14      46.4     51.5       25.6        12.0       77.0
                     1234789-HPCDF-C13     14      55.4     69.0       30.8        14.0       93.0
                            IND_CAT=SUPERFUND/POTW MATRIX=SOLID QUANTITATION=INTERNAL STANDARD
                     COMPOUND             NO OF    MEAN    MEDIAN    STD          MINIMUM    MAXIMUM
£-5                                       OBS                        DEVIATION

"^                  2378-TCDD-C13          3      77.3     79.0        3.8        73.0        80.0
C"                  2378-TCDD-CL37         3      79.3     79.0        1.5        78.0        81.0
                     12378-PECDD-C13        3      93.3     90.0        6.7        £9.0        101.0
                     123478-HXCDD-C13       3      69.0     67.0       15.1        55.0        85.0
                     123678-HXCDD-C13       3      70.3     66.0        7.5        66.0        79.0
                     1234676-HPCDD-C13      3      74.0     69.0        9.5        68.0        85.0
                     OCDD-C13               3      62.7     77.0       10.7        76.0        95.0
                     2378-TCDF-C13          3      75.3     76.0        2.1        73.0        77.0
                     12376-PECDF-C13        3      85.0     79.0       11.3        78.0        98.0
                     23478-PECDF-C13        3      86.0     84.0        3.5        84.0        90.0
                     123478-HXCDF-C13       3      61.3     58.0        5.8        58.0        68.0
                     123678-HXCDF-C13       3      64.7     61.0        9.1        58.0        75.0
                     123789-HXCDF-C13       3      60.3     55.0        9.2        55.0        71.0
                     234678-HXCDF-C13       3      64.7     62.0        6.4        60.0        72.0
                     1234678-HPCDF-C13      3      60.7     57.0        8.1        55.0        70.0
                     1234789-HPCDF-C13      3      76.0     70.0       H.3        69.0        89.0

-------
                            U S E P A
INDUSTRIAL  TECHNOLOGY
      METHOD 1613 PERFORMANCE EVALUATION
                                                                                DIVISION
                                                          TABLE  7
                                        PRECISION AND  RECOVERY OF LABELED
                                                    SUMMARY STATISTICS
                                                    BY  IND CATCMATRIX)
                                    COMPOUNDS
                     	 IND_CAT=NATL SEWAGE SLUD  MATRIX=SOLID  QUANTITATIOH=INTERNAL STANDARD

                      COMPOUND
                      2378-TCDD-C13
                      2378-TCDD-CL37
                      12378-PECDD-C13
                      123478-HXCDD-C13
                      123678-HXCDD-C13
                      1234678-HPCDD-C13
                      OCDD-C13
                      2378-TCDF-C13
                      12378-PECDF-C13
                      23478-PECDF-C13
                      123478-HXCDF-C13
                      123678-HXCDF-C13
                      123789-HXCDF-C13
                      234678-HXCDF-C13
                      1234678-HPCDF-CI3
                      1234789-HPCDF-C13
NO OF
OBS
200
199
201
201
201
201
197
201
201
196
194
201
200
122
199
201
MEAN

64.0
64.9
58.7
70.3
70.5
71.9
51.9
57.2
57.7
46.4
65.2
66.7
67.3
79.3
55.8
64.1
MEDIAN

76.5
78.0
64.0
83.0
84.0
77.0
55.0
67.0
62.0
47.0
75.0
78.0
75.0
78.5
62.0
74.0
STD
DEVIATION
32.6
35.5
31.5
36.7
34.5
33.4
29.0
25.6
25.6
26.9
26.3
32.8
25.9
11.2
24.1
30.6
MINIMUM

3.0
2.6
3.0
2.8
5.7
9.6
2.2
3.5
6.8
2.0
8.0
3.2
10.0
49.0
4.7
6.2
MAXIMU1

119.0
139.0
134.0
137.0
122.0
132.0
118.0
111.0
120.0
121.0
116.0
117.0
118.0
106.0
111.0
119.0
o
-vl
c;

-------
 USEPA  INDUSTRIAL  TECHNOLOGY
                  METHOD 1613 PERFORMANCE EVALUATION
DIVISION
                               TABLE 8
CALIBRATION LINEARITY LIMITS ~ INTERNAL STANDARD AND ISOTOPE DILUTION
__ 	 _____ 	 ____ 	 ___ 	 ____________ 	 MUATIIi.1
COMPOUND


2378-TCDD
12378-PECDD
123478-HXCDD
123678-HXCDD
1234678-HPCDD
OCDD
2378-TCDF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCDF
1234678-HPCDF
1234789-HPCDF
AI_UN-_3UIU
DEGREES
OF
FREEDOM
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
ire UILUI J.UN •
CV a 95X
LIMIT

11.43
12.15
15.07
13.59
13.44
10.05
12.46
9.11
8.40
11.63
12.90
12.05
11.04
24.19
25.24
7. OUT OF
PRED_LMT

0.00
0.00
0.00
0.00
0.00
50.00
0.00
0.00
0.00
0.00
0.00
50.00
0.00
0.00
0.00

COMPOUND DEGREES CV 3 95X


-J 2378-TCDD-C13
2378-TCDD-CL37
12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
123789-HXCDD
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
OF
FREEDOM
14
14
14
14
14
14
14
14
14
14
14
14
14
13
14
14
14
14
LIMIT

6.21
10.29
35.89
13.88
22.31
12.21
12.76
18.47
15.85
18.76
17.53
14.74
14.52
14.47
13.94
26.98
13.97
16.51
7. OUT OF
PRED_LMT

0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
50.00
0.00
0.00
0.00
0.00
0.00

-------
USEPA  INDUSTRIAL  TECHNOLOGY
                 METHOD 1613 PERFORMANCE EVALUATION

                              TABLE 9
              ONGOING CALIBRATION VERIFICATION LIMITS
                              ALL LABS
DIVISION
COMPOUND


2378-TCDD
12378-PECDD
123478-HXCDD
123678-HXCDD
1234678-HPCDD
OCDD
2378-TCDF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCDF
1234678-HPCDF
1234789-HPCDF
COMPOUND
O

CO 2378-TCDD-C13
2378-TCDD-CL37
12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
123789-HXCDD
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCOF-C13
123789-HXCUF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
NO OF
OBS

11
11
11
11
11
11
11
11
11
11
11
11
11
11
i A i xun-Aa
SPIKE
LEVEL

10
50
50
50
50
100
10
50
50
50
50
50
50
50
uiurc UXI.UIA
LOWER
95X
PRED_LMT
8.6
44.2
37.6
39.7
41.6
87.5
8.8
46.7
47.2
41.5
40.5
45.9
44.1
43.1
UPPER
95X
PRED_LMT
11.6
56.6
66.5
63.0
60.2
114.4
11.3
53.5
53.0
60.2
61.7
54.5
56.7
58.0
11 50 43.6 57.3
NO OF SPIKE LOWER UPPER
OBS LEVEL 95X 95X

11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11

100
10
100
100
100
50
100
200
100
100
100
100
100
100
100
100
100
100
PRED.LMT
90.0
8.6
80.6
76.1
84.0
42.6
82.2
164.2
87.7
81.8
83.0
85.2
85.0
89.5
85.7
88.5
89.0
83.9
PRED_LMT
111.2
11.6
124.0
131.3
119.1
58.7
121.6
243.6
114.0
122.3
120.5
117.4
117.7
111.7
116.7
113.1
112.4
119.2
X OUT OF
PRED.LMT

0.00
0.00
0.00
0.00
0.00
0.00
0.00
9.09
0.00
0.00
0.00
0.00
9.09
0.00
9.09
7. OUT OF
PRED.LMT

9.09
9.09
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
9.09
9.09
0.00
9.09
18.18
0.00

-------
USEPA  INDUSTRIAL  TECHNOLOGY
                 METHOD 1613 PERFORMANCE EVALUATION
DIVISION
                              TABLE 10
 RELATIVE RETENTION TIMES — INTERNAL STANDARD AND ISOTOPE DILUTION
                              ALL LABS
COMPOUND


2378-TCDD
12378-PECOD
123478-HXCDD
123678-HXCDD
1234678-HPCDD
OCDD
2378-TCDF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCDF
1234678-HPCDF
1234789-HPCDF
COMPOUND

C£- 2378-TCDD-C13
12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
123789-HXCDD-C13
123789-HXCOD
1234678-HPCDD-C13
OCDO-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
NO OF
DBS

95
96
96
96
96
96
95
96
96
96
96
96
69
96
96
NO OF
DBS

94
94
94
94
94
94
94
94
94
94
94
94
94
94
69
94
94
96
<*UAn 1 J. 1 A 1
MEAN


1.001
1.001
1.000
1.001
1.001
1.000
1.001
0.997
1.000
1.000
0.999
1.003
1.000
0.976
•lun-jauiurc i
STD
DEVIATION

0.003
0.005
0.000
0.003
0.006
0.002
0.003
0.030
0.000
0.005
0.010
0.025
0.000
0.017
0.988 0.035
Ol lAMTTTATTnM-TMTFDMAI
MwAn i x i M i £ur(— xn i cnnAi.
MEAN STD
DEVIATION

1.015
1.281
0.988
0.990
1.000
1.001
1.074
1.162
0.963
1.231
1.265
0.970
0.973
1.005
0.985
1.063
1.086
1.004

0.008
0.056
0.005
0.006
0.000
0.006
0.019
0.043
0.012
0.053
0.054
0.009
0.013
0.004
0.006
0.041
0.023
0.004
LOWER
95X
PRED_LMT
0.993
0.987
0.999
0.992
0.986
0.996
0.993
0.918
1.000
0.986
0.973
0.937
1.000
0.930
0.896
LOWER
95X
PRED.LMT
0.993
1.134
0.974
0.975
1.000
0.986
1.023
1.050
0.931
1.091
1.123
0.947
0.940
0.993
0.971
0.953
1.024
0.995
UPPER
95X
PRED_LMT
1.009
1.016
1.001
1.009
1.016
1.005
1.009
1.076
1.001
1.015
1.025
1.068
1.001
1.022
1.079
UPPER
95X
PRED_LMT
1.036
1.428
1.002
1.006
1.000
1.016
1.125
1.275
0.994
1.371
1.408
0.992
1.006
1.017
1.000
1.172
1.148
1.013
'/. OUT OF
PRED.LMT

4.21
1.04
3.13
4.17
3.13
3.13
4.21
1.04
3.13
1.04
1.04
1.04
1.45
0.00
0.00
X OUT OF
PRED_LMT

0.00
0.00
0.00
1.06
0.00
2.13
0.00
0.00
0.00
0.00
0.00
1.06
1.06
1.06
0.00
0.00
0.00
0.00
                                                                                                                      !•   ~.:-7
                                                                                                                      A     H

-------
USEPA  INDUSTRIAL  TECHNOLOGY  DIVISION
                 METHOD 1613 PERFORMANCE EVALUATION

                             TABLE HA
   INITIAL PRECISION AND RECOVERY — LABELED AMD NATIVE COMPOUNDS
                    START-UP LIMITS FOR ACCURACY
                              ALL LABS
COMPOUND 95 X LOWER
LIMIT
2378-TCDD 38.67
12378-PECDD 94.87
123478-HXCDD 61.80
123678-HXCDD 66.43
1234678-HPCDO 54.91
OCDO 69.37
2378-TCDF 31.56
12378-PECDF 88.54
e3478-PECDF 90.69
123478-HXCDF 51.32
123678-HXCDF 78.20
123789-HXCDF 75.85
234678-HXCDF 54.80
1E34678-HPCDF 79.12
1234789-HPCDF 73.32
ft U1LUI1UN —
95 X UPPER
LIMIT
205.96
100.72
140.35
131.82
153.00
154.63
267.82
108.17
100.46
183.94
108.86
125.85
171.53
124.23
129.83
X OUT OF
PRED_LMT
0.00
60.00
0.00
0.00
0.00
0.00
0.00
20.00
20.00
0.00
0.00
0.00
0.00
0.00
0.00

O COMPOUND 95 X LOWER 95 X UPPER
C 5 LIMIT
1 2378-TCDD-C13 26.00
2378-TCDD-CL37 31.06
12378-PECDD-C13 20.98
123478-HXCDD-C13 50.75
123678-HXCDD-C13 48.82
123789-HXCDD 45.39
1234678-HPCDD-C13 75.16
OCDD-C13 46.15
2378-TCDF-C13 16.57
12378-PECDF-C13 15.15
23478-PECDF-C13 17.47
123478-HXCDF-C13 51.78
123678-HXCDF-C13 44.86
123789-HXCDF-C13 60.57
234678-HXCDF-C13 52.58
1234678-HPCDF-C13 55.78
1234789-HPCDF-C13 41.24
OCDF 82.88
LIMIT
121.20
104.58
261.19
108.33
127.96
181.84
90.73
139.76
211.08
331.51
317.78
81.95
116.22
87.34
98.78
81.71
129.06
116.03
X OUT OF
PRED_LMT
0.00
20.00
0.00
0.00
0.00
0.00
40.00
0.00
0.00
0.00
0.00
20.00
0.00
20.00
0.00
20.00
0.00
60.00

-------
USEPA  INDUSTRIAL  TECHNOLOGY  DIVISION
                 METHOD 1613 PERFORMANCE EVALUATION

                             TABLE 11B
   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                   START-UP LIMITS FOR PRECISION
                              ALL UBS
COMPOUND

2378-TCDO
12378-PECDD
123478-HXCDD
123678-HXCOD
1234678-HPCDD
OCOD
2378-TCDF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCDF
1234678-HPCOF
1234789-HPCDF
n-j.au lure uiuu
95 7. UPPER
LIMIT
13.13
8.43
11.02
11.72
12.73
12.63
19.87
9.29
8.49
12.50
7.26
7.21
10.18
7.06
8.32
X OUT OF
PRED_LMT
0.00
20.00
20.00
20.00
0.00
20.00
0.00
20.00
20.00
0.00
20.00
20.00
20.00
20.00
20.00

COMPOUND 95 X UPPER X OUT OF

o
GO 2378-TCDD-C13
L^. 2378-TCDD-CL37
12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
123789-HXCDD
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
LIMIT

45.92
55.61
38.27
24.47
42.54
18.77
23.58
45.14
53.43
31.24
32.48
23.57
29.79
26.03
29.30
21.86
22.80
45.15
PRED_LMT

0.00
0.00
0.00
20.00
20.00
20.00
0.00
0.00
0.00
20.00
20.00
0.00
0.00
0.00
0.00
0.00
20.00
0.00
                                                                                                                    ,

-------
U S E P A  INDUSTRIAL  TECHNOLOGY  DIVISION
                 METHOD 1613 PERFORMANCE EVALUATION

                             TABLE 12
   ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                  ONGOING QUALITY ASSURANCE LIMITS
                              ALL LABS
	 	 	 	 	 	 	 ——---..----—--- HUMPH i i«i iun-iouiu
COMPOUND 95 X LOWER
LIMIT
2378-TCDD 59.49
12378-PECDD 71.34
123478-HXCDO 70.15
123678-HXCDD 66.61
1234678-HPCDD 74.98
OCOD 75.58
2378-TCDF 66.09
12378-PECDF 73.43
23478-PECDF 75.61
123478-HXCDF 73.86
123678-HXCDF 69.65
123789-HXCDF 74.20
234678-HXCDF 71.40
1234678-HPCDF 74.89
95 •/. UPPER
LIMIT
142.30
116.22
120.83
128.70
113.54
118.74
127.44
114.64
113.81
117.64
117.65
111.37
120.08
120.98
1234789-HPCOF 73.81 121.13
COMPOUND 95 7. LOWER 95 7. UPPER
O LIMIT LIMIT
CO
•r; 2378-TCDD-C13 29.36

-------
                             U S E P A
INDUSTRIAL  TECHNOLOGY
      METHOD 1613 PERFORMANCE EVALUATION
                          DIVISION
                                                          TABLE  13
                                         PRECISION AND  RECOVERY  OF LABELED
                                                     SUMMARY STATISTICS
                                                     ALL LABS COMBINED
                                    COMPOUNDS
                                                QUANTITATION=INTERNAL  STANDARD
                    COMPOUND
                    2378-TCDO-C13
                    2378-TCOO-CL37
                    12378-PECDD-C13
                    123478-HXCDO-C13
                    123678-HXCDD-C13
                    1234678-HPCDD-C13
                    OCDD-C13
                    2378-TCOF-C13
                    12378-PECDF-C13
                    23478-PECDF-C13
                    123478-HXCDF-C13
                    123678-HXCDF-C13
                    123789-HXCDF-C13
                    234678-HXCDF-C13
                    1234678-HPCDF-C13
                    1234789-HPCDF-C13
 NO OF
 OBSERVATINS

     293
     291
     293
     294
     293
     293
     290
     294
     294
     289
     286
     294
     292
     204
     292
     294
                                                        MEAN
59.8
61.1
61.2
70.4
69.0
67.2
50.4
55.5
59.1
53.1
64.6
65.4
65.7
74.0
54.8
61.7
                                                                MEDIAN
64.0
64.0
66.0
78.0
75.0
68.0
51.0
61.0
62.0
57.0
70.0
70.
6?.
74.
58.
                         65.0
STD
DEVIATION
30.3
31.8
29.6
32.9
30.7
30.2
26.0
24.8
23.9
27.3
25.4
29.1
23.5
12.7
22.2
27.7
MINIMUM

3.0
2.6
3.0
.2.8
5.7
9.6
2.2
3.5
6.8
2.0
8.0
3.2
10.0
42.4
4.7
6.2
MAXIMUM

119.0
139.0
134.0
151.0
122.0
132.0
118.0
112.0
120.0
121.0
116.0
117.0
118.0
106.0
111.0
119.0
o
CO
Co
                                                                                                                                                    ^
                                                                                                                                               tE.T

-------
o
CO
                            USEPA  INDUSTRIAL  TECHNOLOGY   DIVISION
                                             METHOD 1613 PERFORMANCE  EVALUATION

                                                         TABLE 1*
                                        PRECISION AND RECOVERY OF LABELED   COMPOUNDS
                                                    SUMMARY STATISTICS
                                                  LAB 2 AND LAB 3 COMBINED
                                               QUANTITATION=INTERNAL STANDARD
                   COMPOUND             NO OF          MEAN    MEDIAN    STD           MINIMUM    MAXIMUM
                                        OBSERVATINS                      DEVIATION

                   2378-TCDD-C13            216        71.6     77.5       25.1          3.0        119.0
                   2378-TCDD-CL37           214        74.2     78.0       25.7          5.8        139.0
                   12378-PECDO-C13          215        72.6     75.0       23.2          7.1        134.0
                   123478-HXCDD-C13         216        85.3     87.9       22.9          4.4        151.0
                   123678-HXCDD-C13         215        82.4     85.0       22.1          7.3        122.0
                   1234678-HPCDD-C13        215        78.0     77.0       26.0         14.0        132.0
                   OCDD-C13                 215        59.5     59.0       21.2         10.0        118.0
                   2378-TCOF-C13            216        64.6     68.0       20.3          4.0        112.0
                   12378-PECDF-C13          216        67.4     69.0       18.9          8.0        120.0
                   23478-PECDF-C13          215        62.8     63.0       22.0          3.5        121.0
                   123478-HXCDF-C13         215        75.1     77.0       17.9          8.0        116.0
                   123678-HXCDF-C13         216        77.9     80.0       21.1          3.9        117.0
                   123789-HXCDF-C13         215        74.8     76.0       17.8         20.0        118.0
                   234678-HXCDF-C13         204        74.0     74.2       12.7         42.4        106.0
                   1234678-HPCDF-C13        214        62.9     64.3       17.6          7.0        111.0
                   1234789-HPCDF-C13        216        72.6     75.0       21.3         11.9        119.0

-------
o
oo
                                USEPA   INDUSTRIAL  TECHNOLOGY  DIVISION
                                                METHOD  1613  PERFORMANCE EVALUATION

                                                             TABLE  15
                                            PRECISION AND RECOVERY OF LABELED COMPOUNDS
                                                     QUALITY ASSURANCE LIMITS
                                                  QUANTITATION=INTERNAL STANDARD
                                    COMPOUND             95 X LOWER    95 X UPPER    X OUT OF
                                                         LIMIT         LIMIT         PRED_LMT

                                    2378-TCDD-C13          33.09         168.23        9.72
                                    2378-TCDD-CL37         27.97         209.43        4.19
                                    12378-PECDD-C13        36.92         152.56        8.33
                                    123478-HXCDD-C13       52.11         145.05        8.80
                                    123678-HXCDD-C13       44.91         156.80        6.02
                                    1234678-HPCDD-C13      29.84         198.48        4.63
                                    OCDD-C13               25.94         138.15        6.48
                                    2378-TCDF-C13          41.31         110.93       14.81
                                    12378-PECOF-C13        41.59         116.59       10.65
                                    23478-PECOF-C13        29.14         131.21        7.41
                                    123478-HXCDF-C13       48.76         122.80        7.87
                                    123678-HXCDF-C13       41.92         149.47        5.09
                                    123789-HXCDF-C13       37.68         152.32        4.63
                                    234678-HXCDF-C13       49.41         110.13        7.87
                                    1234678-HPCDF-C13      33.93         121.66        5.09
                                    1234789-HPCDF-C13      28.41         192.74        4.63

-------
                                  Appendix A

                                   Outliers
To examine the data for outlying values,  Hoaglin et al.  (1983)  provides  an
approach based on the interquartile range.   When observations  came  from  a
normal distribution,  points outside the following range  are  rejected
      M  +      - tt>   •   IQR
         -  2 N (.75)

      where ,
      M   = median
      IQR = interquartile range
      N(l - a) = (1 - a)th percentile of normal  distribution

The data were screened for outlier values using  the above method.   This
method of screening was applied to all three types of data sets (IPR,  OPR and
EPA- type Samples) on the  logarithms of the amounts.   A total of 29
observations were identified as outliers in IPR- type samples (722  obs) and
deleted.  From OPR-type samples,  89 observations  were deleted (1624 obs).   A
total of 245 observations were identified as outliers in EPA- type  samples
(3455 obs) and deleted.
                                         086

-------
                                  Appendix B

                       Estimation  of Variance Components


For the purpose of calculating quality control limits from the data in this
study, the variance components model assumes that the logarithm of the
percent recovery X^- measured by laboratory i and replicate j  can be written
as:

     log (Xjj)  =  fj. + EL + Ay

     where,
     i   =1,2,3 labs,
     j   = l,2,...n£ replicate measurements at lab i,

     p   = average response,
                             .                                     2
     E^  «= random interlaboratory effect with mean 0 and variance a ,  and
                                                                   III
                                                                   2
     ^11 = rand°m intralaboratory effect with mean 0 and variance a^.

The variance components analysis was performed by the Type 1 method using
PROC VARCOMP(SAS) to estimate the inter- and intralaboratory variance
components of the logarithms of the percent recovery.  The MIVQUE method also
showed comparable results.  Tables B-l through B-3 give the results of the
variance components analysis for each sample type (IPR, OPR, EPA),
re'spectively.  For each compound, the total numbers of observations, the
logarithmic mean M, the square roots of the variance components Sr.
(interlaboratory) and  S.  (intralaboratory), and the percentage of the total
variance due to interlaboratory variation is computed as:


                            100 *  f ——E-—— 1
                                            087

-------
U S E P A
           INDUSTRIAL  TECHNOLOGY
                 METHOD 1613 PERFORMANCE EVALUATION
DIVISION
                             TABLE B-l
   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
               RESULTS OF VARIANCE COMPONENT ANALYSIS
                              ALL LABS
COMPOUND

2378-TCDD
12378-PECDD
123478-HXCDD
123678-HXCDD
1234678-HPCDD
OCDD
2378-TCDF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCDF
1234678-HPCDF
1234789-HPCDF
COMPOUND
^ 2378-TCDD-C13
CO 2378-TCDD-CL37
CO 12378-PECDD-C13
123478-HXCOD-C13
123678-HXCDD-C13
123789-HXCDD
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCOF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
NO OF
OBSERVATIONS
22
19
22
21
22
19
22
21
21
22
21
21
22
21
i j. IMI iun-1
NO OF
LABS
3
3
3
3
3
3
3
3
3
3
3
3
3
3
auiurc i
LOG
MEAN
4.49
4.58
4.53
4.54
4.51
4.64
4.51
4.58
4.56
4.57
4.52
4.58
4.57
4.60
20 3 4.58
NO OF NO OF LOG
OBSERVATIONS LABS MEAN
22
22
22
22
22
22
22
22
22
22
22
22
22
22
ie
22
22
22
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
3
3
3
3.93
3.93
4.26
4.29
4.32
4.50
4.40
4.33
3.97
4.23
4.28
4.15
4.25
4.26
4.25
4.19
4.27
4.55
WITHIN
(S_A)
0.09
0.05
0.07
0.08
0.09
0.08
0.13
0.06
0.06
0.08
0.05
0.05
0.06
0.04
0.05
WITHIN
(S_A)
0.45
0.51
0.30
0.20
0.31
0.13
0.17
0.33
0.48
0.26
0.25
0.22
0.24
0.22
0.24
0.20
0.19
0.27
BETWEEN(S_E)

0.19
0.00
0.11
0.10
0.13
0.10
0.24
0.03
0.02
0.14
0.05
0.07
0.13
0.06
0.07
BETHEEN(S_E)
0.24
0.16
0.36
0.13
0.16
0.18
0.00
0.18
0.39
0.40
0.38
0.07
0.16
0.05
0.09
0.06
0.18
0.02
7. VAR DUE -
TO LAB
80.34
0.00
67.40
60.53
70.53
64.08
76.00
22.93
7.29
75.36
50.08
67.24
79.08
63.48
66.03
'/. VAR DUE
TO LAB
22.75
8.58
58.60
27.80
20.65
66.39
0.00
23.66
40.25
70.08
68.67
9.59
29.12
5.08
12.25
7.99
46.53
0.47

-------
USEPA  INDUSTRIAL  TECHNOLOGY
                 METHOD 1613 PERFORMANCE EVALUATION
DIVISION
                             TABLE B-2
   ONGOING PRECISION AND RECOVERY ~ LABELED AND NATIVE COMPOUNDS
               RESULTS OF VARIANCE COMPONENT ANALYSIS
                              ALL LABS
COMPOUND

2378-TCOD
12378-PECDD
123478-HXCDD
123678-HXCDD
1234678-HPCDD
OCDD
2378-TCDF
12378-PECDF
23478-PECDF
123478-HXCDF
123678-HXCDF
123789-HXCDF
234678-HXCDF
1234678-HPCDF
1234789-HPCDF
COMPOUND
o
C» 2378-TCDD-C13
1C 2378-TCDD-CL37
12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
123789-HXCDD
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDr-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
OCDF
— 	 	 — — — mj
NO OF
OBSERVATIONS
49
47
44
49
44
47
47
44
45
47
47
44
47
48
IMTililAllU
NO OF
LABS
3
3
3
3
2
3
3
3
3
3
3
3
3 •
3
n-isuiurt ui
LOG
MEAN
4.52192
4.51145
4.52248
4.52818
4.52470
4.55107
4.51935
4.51909
4.53005
4.53493
4.50562
4.50980
4.52824
4.55582
WITHIN
(S_AJ
0.15
0.10
0.11
0.15
0.09
0.09
0.12
0.07
0.09
0.10
0.10
0.08
0.10
0.10
47 3 4.54922 0.11
NO OF NO OF LOG WITHIN
OBSERVATIONS
42
45
49
47
49
49
50
47
44
49
48
47
47
48
35
50
47
46
LABS
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
3
3
3
MEAN
4.09185
4.04677
4.19112
4.28357
4.31097
4.48022
4.21991
4.01236
4.07703
4.11370
4.16151
4.23546
4.25858
4.23199
4.22599
4.12521
4.19957
4.54296
(S_A)
0.28
0.37
0.29
0.17
0.20
0.14
0.26
0.28
0.28
0.32
0.31
0.19
0.19
0.20
0.21
0.23
0.26
0.11
BETWEEN! (S_E)

0.10
0.04
0.03
0.00
0.00
0.04
0.06
0.05
0.00
0.01
0.05
0.03
0.04
0.03
0.00
BETMEEN((S_E)

0.11
0.14
0.38
0.12
0.11
0.02
0.00
0.12
0.00
0.22
0.28
0.01
0.00
0.03
0.00
0.00
0.05
0.06
'/. VAR.DUE
TO LAB
30.38
15.49
6.31
0.00
0.00
18.56
21.38
29.84
0.00
0.73
23.16
11.83
14.63
6.10
0.00
7. VAR DUE
TO LAB
12.91
11.97
62.77
34.40
23.34
1.19
0.00
14.04
0.00
32.58
45.28
0.15
0.00
2.66
0.00
0.00
3.20
20.84

-------
                              U S E P A
                                         INDUSTRIAL  TECHNOLOGY
                                               METHOD 1613 PERFORMANCE EVALUATION
DIVISION
                                                           TABLE B-3
                                          PRECISION AND RECOVERY OF LABELED COMPOUNDS
                                             RESULTS OF VARIANCE COMPONENT ANALYSIS
                                                            ALL LABS
COMPOUND

2378-TCDD-C13
2378-TCDD-CL37
12378-PECDD-C13
123478-HXCDD-C13
123678-HXCDD-C13
1234678-HPCDD-C13
OCDD-C13
2378-TCDF-C13
12378-PECDF-C13
23478-PECDF-C13
123478-HXCDF-C13
123678-HXCDF-C13
123789-HXCDF-C13
234678-HXCDF-C13
1234678-HPCDF-C13
1234789-HPCDF-C13
NO OF
OBSERVATIONS
195
195
194
202
202
206
202
194
197
206
199
204
204
203
205
202
HiAun-j.nii
NO OF
LABS
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
iKrwu SIM
L06
MEAN
4.31
4.34
4.32
4.47
4.43
4.34
4.09
4.21
4.24
4.12
4.35
4.37
4.33
4.30
4.16
4.30
WITHIN
(S_A)
0.27
0.27
0.24
0.21
0.21
0.30
0.30
0.23
0.21
0.34
0.17
0.21
0.18
0.17
0.23
0.22
BETWEEN
(S_E)
0.20
0.27
0.16
0.08
0.15
0.23
0.18
0.00
0.09
0.02
0.09
0.15
0.18
0.05
0.14
0.26
X VAR DUE
TO LAB
35.33
49.86
31.03
11.47
32.89
37.34
26.13
0.00
15.50
0.47
22.51
34.51
49.98
8.55
26.31
57.84
o

-------
                                  Appendix C

               Derivation  of Quality Control Limits For Accuracy
If we observe a test series X	X»j independently chosen from a normal
distribution in the unknown mean M and unknown variance a ,  the mean and
variance can be estimated by:

                         N
                  X = 1  ^S x.
                         N
                  S2 -   \ (xL-x)2
A (1-p) 100 percent confidence interval for a single independent future
observation X from the same distribution can be constructed by noting that
x - x has mean 0 and variance (1 + 1/N) a ,  hence:
                  x € (x ± tN.x (l-p/2) J(1+1/N) S)

with probability exactly (1-p),  where £„-! is the inverse cumulative t
distribution with N-l degrees of freedom.  All of the quality control limits
formulae for accuracy used in this report are extensions of this concept.

Log Normal Data

If X.; has a log normal distribution with logarithmic mean n and_logarithmic
variance a , the limits can be derived by letting Y- =  a log (X-) and
computing Y and S .  The prediction interval for the future value of
Y- = log (X-) can be exponentiated to obtain
                  X  £   (Exp [Y - tN.1(l-p/2) Jl-H/N S ]
                         (Exp [Y + tN^d-p/2) Jl+l/N Syl)

with probability 1-p.

Average of Log Normal Values

Because the start-up test is to be based on the arithmetic average of four
observations, we consider the case of predicting future x when the data are
drawn from a ^og normal  distribution, with parameters fj. and a .   For small
values of n, Xn can be assumed to have log normal distribution (see
References 1 and 2).
                                            091

-------
     Let  Y  =  log (Xn)




We derive a prediction interval for Y and then exponentiate it to produce a

prediction interval for Xn.




If X-^, X-2> Xj,  X^ have log normal distribution,


      —                             o

      Xn has mean m = exp (M + 1/2 a )



                        2 2

      and variance  =  m

                        n
where ,   n  = 4 ,  and


        K2 = [exp(a




Using converging Taylor's series on f(x) = log(x) ,
        K2 = [exp(a2 - 1]
                                                    1  k2
Y =• log (X ) will have a mean'    =  M  +   — r

          n                                2        2  n


                      2


and a variance   =   —

                     n



             2                           2
Since /* and a  are estimated by Y,, and S   as before, we have:





                  v   f ^7    i o2     i  ky 1
                      I  "   2  y     2  n  J



        2         2
where, k^ = exp (S.~)-l) has asymptotic mean zero and variance approximately:
                  kf + £1 + I fi    I]

                  n    N    4 [    nj
                                       (N-l)



                     oo            o             22
Using the fact that k  - a  for small a  to combine S^ and k^ terms^


Therefore, an approximate lOO(l-p) percent confidence interval for XR can be


computed as:





                             1  2     lkVl
              Exp   |  Yn  +  - S^	y    -  tN , (1

                          .   2  y     2  n I
              Exp     v-  ^    c2
                      Y"  +  2  y" 2  n       +  tN-l(1'P/2) ' S
         -   k'y   s2y   l          2   s4y
where,   S = —- + —- + - (I - l/n)* 	-

              n     N    2             (N-l)



Variance Components
Let  Xij = M + Ei + A.J
                                            032

-------
where,
i - 1,
j = 1,
Ei =
2, . .
2, . .
N (0, c
N (0, <
. i;
. J;
J)'
                          and
The estimates M, Sg, and Sf of the p, Sg,  and a^ respectively are

obtainable through a variance component analysis.   Since the difference X-M

has mean 0 and asymptotic variance °v + api + CTE/^ + CTA/*^ '  ancl
approximate prediction interval is given by:
M - t
              - p/2)S,  M
                                           - P/2)S
where,
S =
c2 o2
S2 S2
I IJ
It will often be necessary to know the number of degrees of freedom (d) that

can be assigned to S.  The degrees of freedom are given by Wilson (1979).


When the between laboratories mean square (BLMS) is much greater than the

within laboratory mean square (WLMS),  a' close approximation given by:
                  d
                1(1-1) [BLMS + (J-l)WLMSp


                I(BLMS)2 + (I-1)(J-1)(WLMS)2
when,  BLM < WLMS use d ~ I(J-l).


Applications


The limits for the arithmetic average (X ) of the four tests are obtained by

combining the variance components estimates to give:
            exp
                    M
                1
             +  —
                2
                             2
                             ^
                                      n
  t(d,l-p/2)
                           (S
                                 K
                                 K
                                 n
                                             N
                                        (N-L)
                                   o         2
where M, S., and Sg are as above, Kf =• exp (Sf)-l, n=4, N is the number of

observations in the study, and L is the number of laboratories.
                                          033

-------
The ongoing precison and recovery limits  are  obtained  from  the  analysis of
the OPR type samples as:
      exp
           M ±
The limits for labeled compound recoveries  are obtained  from  the analysis of
the EPA type samples as:
      exp
           M ±
SE
+ *l +
s£
L
+ »2
N
The ongoing calibration verification limits  are  obtained  from  the  analysis of
VER-type samples as:
      exp
ln(MU)  ± t(d,(l-p)/2)SA
      where MU is the corresponding spike  level.
                                           034

-------
                                                                     ,=
                                                                 a -i»/.v
                                  Appendix D

              Derivation of Quality Control Limits For Precision
We need to determine the distribution of:
                            n
                  S	L-
where n=4 and X^ are distributed log normally, the distribution of S depends
on a ,  but n can be removed from consideration by a transformation such that:
                  S'  =

                         2
Finally, for a range of a  values, the upper quantiles of S' were determined
by a simulation which was performed with SAS ,  using 10,000 replicates, and
the quantiles were estimated with PROG UNIVARIATE.

The precision limit on the standard deviation is given by:

                  exp(M)
where Q is the upper quantile at 1-p of SA and K is the approximate
correction factor for the estimation of S. ,  approximated as- (see Reference
2):
Fn-l,d(l-P)
      K
and F and C are the inverse cumulative distributions of the F and Chi-squared
distributions, respectively, and d is the df of S^.

-------
                                 Appendix E

                           Calibration Linearity

In order to calculate  the  concentration of each compound in a sample ,  a
calibration curve is applied to the peak area of the compound and of the
reference compound obtained from  the Gas Chromatograph .  This calibration
curve is constructed by the analysis of a series of calibration samples at
known concentrations .

For isotop dilution, the ratio, of the peak areas and concentrations to  those
of a labeled compound  are  used for calibration

         A/Alabel    =  f(  c/clabel)
and the unknown concentration in a sample is constructed from the area ratio
and the known level of  the standard spiked into the sample by

                             )  clabel  j
                                         i
In estimating the calibration curve, a range of calibration  samples are
used in order to evaluate the response of the instrument over its
performance range.  In this study, five points were obtained for each
laboratory, at levels specified in the method (only two labs provided data
for this).

  The simplest form of  the response curve is proportional response curve

           f(x)   =ax

The random variation in the calibration response can be assumed  to have a
proportional error structure, i.e., for repeated measurements the area ratio
is distributed around    f (C/clabel)i            as


         f(C/Clabel)(l+0/


where £ has  mean zero and variance a ^ i ! independent of C/C^abel\ In the case
of linear response curve

                                         I
        A/Alabel   =a(C/Clabel)
and rewritting in terms of the response factor RF give



       RF  =>(A/Alabel)/(c/Clabel)   -  a(l +  O-





                                            090

-------
Hence the coefficient of variation of the response factor would be
        a2a2/a   = a
a constant for all concentration ratios.

If a linear propotional calibration curve is to be fitted to a calibration
set containing values   AI/ A2,  ..., A5, Alabell, Alabei2,  ••••   Alabel5,

  cl'   C2,.-., C5,Clabell, Clabei2'  •••» Clabel5 j
, the best estimate of the calibration coefficient can be derived form the
formula for a weighted regression^ (see reference 3) as

        (Ai/Alabeli)/(Ci/Ciabeli)  - Vn V KF±  I

Where  RF is the average response factor

Linearity Specification calculation

Linearity specification is calculated based on the CAL type samples. RF was
calculated for each compound. A mild pre- screening of the RF was done.
Then, appropriate limits for testing the goddness of fit of a linear
calibration were obtained. For each compound and laboratory, the variance
among midpoint calibrations' were obtained. And this variance was
standardized by the square of the mean response factor for all calibration
points. This stadardized variance was then averaged across all laboratories,
weighted by the degrees of freedom for each laboratory, to come up with an
overall standardized variance of response factors o2|, with d total degrees
of freedom. Then for calibration, assuming the response is truely linear
with proportional error structure, the test for linearity used was to
compare the coefficient of variation of   RFlf RF2,
with
          C V Limit  = 100 ( a2 FM_lfd(.95) ) V2


where N is the number of RFs . F is the inverse of the cumilative F
distribution, d is the degrees of freedom for o2[ These values are
calculated in Table 8

These limits were applied to actual calibration data to find out how many
points are there  outside this specification.
                                  Appendix  F
                           •   •

                           Relative Retention Time
                                      ,o      °97

-------
                                 Appendix F

                          Relative Retention Time
Before analysis,  a mild pre-screening was  done  to  relative  retention time
data. Relative retention was  calculated from the absolute retention time  of
compound and its  reference compound.  Nominal scale analyses were  used for
these calculations,  the 95 percent confidence limits  for the mean of the
quantity is computed as

    x"  *

where

    T =
    SN = [ £(Xi-"x)2/N-l]1/2
                             I

and fcN-l is the inverse cumilative distribution functon of  the  t
distribution with N-l degrees of freedom.  The prediction limits are given by
    X ±
.  The limits are given in Table 10  . Also  given are  the percentage  of times
the observations fell outside  the calculated limits.
                                     11
                                           038

-------
                        References for Appendices
Eynon B.P, Maxwell C., and Valder A.  Interlaboratorv Validation  of U.S.
Environmental Protection Agency Method 1625A.  USEPA Contract  68-01-6192, June
1984.

Hoaglin, D.C., Hosteller, -P.,-and Turkey,  J.S.  Understanding  Robust  and
Exploratory Data Analysis. John Wiley and Sons,  New York,  pp  37-39,  1983.

Milliken G.A., Johnson D.E. Analysis  of Messy Data.  Lifetime  Learning
Publications, CA, pp 238-241, 1984.

Wilson, A.L. Approach for Achieving Comparable Analytical  Results From a
Number of Laboratories. The Analyst,  Vol.  104 No.  1237,  p  273, 1979.

Hunt, D.T., Wilson, A.L. The Chemical Analysis of Water. The  Royal Society  of
Chemistry, London, pp 252-254, 1986.

U.S. EPA Development Document for Existing Source Pretreatment Standards for
the Electroplating Point Source Category,  USEPA, Document  440/1-79/003,
August 1979.
                                          090

-------
                         FIGURE 1
INITIAL PRECISION AND RECOVERY OF LABELED AND NATIVE COMPOUNDS



      FREQUENCY DISTRIBUTION PLOT OF ALL LABS COMBINED
                           100

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                         COOE=2378-TCDF-C13
    150.
   130.
S
c
H
E
M
A
T
I
C  110.
F
0  90.0
R
     O
   70.0
   50.0
   30.0

-------
                                 INITIAL PRECISION AND  RECOVERY —  LABELED  AND  NATIVE  COMPOUNDS
                                     .   FREQUENCY DISTRIBUTION PLOT --  ALL  LABS COMBINED
                                                         CODE=2378-TCDF
 150.
 130.
 110.
90.0
O
70.0
50.0
30.0

-------
                                   INITIAL PRECISION AND RECOVERY -  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLCT ~ ALL LABS COMBINED
                                                         CODE=2378-TCDD-C13
    150.
   130.
S
c
H
E
M
A
T
I
C  110.
F
0  90.0
R
    O
    Go
   70.0
   50.0
   30.0

-------
                                 INITIAL  PRECISION  AND  RECOVERY  --  LABELED  AND  NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT —  ALL  LABS COMBINED
                                                     CODE=237B-TCDD-CL37
150.
130.
110.
F
0  90.0
R

V
A
H    ^-*
I
A
B    """
L  70.0
E
  o
                                                                  *- + -*
50.0
30.0

-------
                                   INITIAL PRECISION AND RECOVERY -- LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED   '
                                                           CODE=2378-TCDD
    150.
    130.
S
c
H
E
M
A
T
I
C   110.
F
0  90.0
R
     O
v
A
R
I
A    ^
B
L  70.0
E
                                                                       •f
                                                                     *	*
                                                                     I   I
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                        CODE=23478-PECDF-C13
   150.
   130.
   110.
F
0  90.0
R
    O
   70.0
   50.0
   30.0
                                                                     *-+-#

-------
                                INITIAL PRECISION AND RECOVERY —  LABELED AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED .
                                                       CODE=2347fl-PECDF
 150.
 130.
110.
90.0
                                                                     I
                                                                     I
                                                                  •f	*
                                                                  I   I
                                                                  *-+-*
                                                                  +	+
                                                                     I
  o
  ^ J

70.0
50.0
30.0

-------
                                INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                    CODE=234678-HXCDF-C13
150.
130.
110.
90.0
  O
  CC

70.0
                                                                  +	*
50.0
30.0

-------
                                   INITIAL PRECISION AND RECOVERY  --  LABELED  AND NATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL  LABS COMBINED
                                                         CODE=33
-------
                                   INITIAL PRECISION AND RECOVERY -- LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123789-HXCDF-C13
   150.
   130.
S
c
H
E
M
A
T
I
C  110.
F
0  90.0
R
   70.0
   50.0
   30.0

-------
                                INITIAL PRECISION AND RECOVERY -- LABELED AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                      CODE=123789-HXCDF
 150.
 130.
110.
90.0
                                                                  *	*
                                                                  I + I
                                                                  +	+
                                                                    I
                                                                    0
70.0
50.0
30.0

-------
                                 INITIAL  PRECISION  AND  RECOVERY  —  UBELED AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                       CODE=123789-HXCDD
150.
130.
110.
                                                                  +	1
90.0
                                                                  •f	+
70.0
50.0
30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                        CODE=12378-PECDF-C13
   150.
   130.
S
c
H
E
M
A
T
I
C  110.
   90.0
   70.0
   50.0
                                                                     •f	+
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY —  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                          CODE=12378-PECDF
   150.
   130.
   110.
                                                                       I
                                                                       I
                                                                      	*
0  90.0
R
   70.0
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY —  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                        CODE=1237fl-PECDD-C13
    150.
    130.
    110.
F
0  90.0
R

V
A      \-


<      5
B
L  70.0
E
   50.0
   30.0

-------
                                 INITIAL  PRECISION  AND RECOVERY  —  LABELED AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                       CODE=12378-PECDD
150.
130.
110.
90.0
70.0
50.0
30.0

-------
                                   INITIAL PRECISION AND RECOVERY -- LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                       CODE=123676-HXCDF-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
   90.0
   70.0
   50.0
   30.0

-------
                                 INITIAL  PRECISION  AND RECOVERY  ~  LABELED  AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT ~ ALL  LABS COMBINED
                                                      CODE=123678-HXCDF
150.
130.
110.
90.0
                                                                    *
                                                                    0
                                                                    I
                                                                    I
                                                                  +-•—+
                                                                  *-*-*
70.0
50.0
30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                       CODE=123678-HXCDD-C13
   150.
   130.
S
c
H
E
M
A
T
I
C  110.
F
0  90.0
R
   70.0
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                         CODE=12367fl-HXCDD
   150.
   130.
   110.
F
0  90.0
R
       ro
       o
   70.0
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                       CODE=1234789-HPCDF-C13
    150.
   130.
   110.
F
0  90.0
R
   FO
   70.0
                                                                     *- + -*
   50.0
   30.0

-------
                                   INITIAL  PRECISION AND RECOVERY ~  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PIOT ~ ALL LABS COMBINED
                                                         CODE=12347*9-HPCDF
   150.
   130.
   110.
F
0  90.0
R
  *
  I
  I
  I
* --- *
I + I
+ --- +
  I
                                                                       *
                                                                       *
                                                                       *
   70.0
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND  NATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123478-HXCDF-C13
    150.
   130.
S
C
H
E
M ,
A
T
I
C  110.
F
0  90.0
R
      fO
      CO
   70.0
                                                                     *-+-*
   50.0
   30.0

-------
                                    INITIAL  PRECISION AND  RECOVERY  --  LABELED  AND  NATIVE  COMPOUNDS
                                          FREQUENCY  DISTRIBUTION PLOT --  ALL  LABS COMBINED
                                                          CODE=123478-HXCDF
   150.
   130.
S
c
H
E
M
A
T
I
C  110.
                                                                     «-•»-*
F
0  90.0
R
   70.0
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123478-HXCDD-C13
    150.
   130.
   110.
0  90.0
R
V
A
R
I
A
B
L  70.0
E
ro
C/!
   50.0
   30.0

-------
                                INITIAL PRECISION AND  RECOVERY ~ LABELED AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                      CODE=1E3478-HXCDD
150.
130.
110.
90.0
I    I
I    I
*-+-*
I    I
 rc
 cr

70.0
50.0
30.0

-------
                                   INITIAL PRECISION ANO RECOVERY -- LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123467fl-HPCDF-C13
    150.
   130.
   110.
F
0  90.0
R
   70.0
                                                                     I  +
                                                                     I    I
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY -- LABELED AND  NATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                         CODE=123467fl-HPCDF
   150.
   130.
   110.
0  90.0
R
    ro
    CO

   70.0
   50.0
   30.0
                                                                     *-+-*
                                                                     +	1

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                       CODE=1234678-HPCDD-C13
    150.
    130.
   110.
F
0  90.0
R
    ro
   70.0

   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY -- LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                         CODE=1E34678-HPCDD
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
                                                                     K	*

                                                                       •f
   90.0
   70.0
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                             CODE=OCDF
   150.
F
0
R

V
A
R
I
A
B
L
E
P
R
E
C
   130.
S
C
H
E
M
A
T
I
C  110.

P
L
0
T
S
90.0
70.0
   50.0
   30.0
                                                                  +—t

-------
                                   INITIAL PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                           CODE=OCDO-C13
   150.
   130.
S
C
H
E
M
A
T
I
c  no.
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0

-------
                                   INITIAL PRECISION AND RECOVERY —  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                             CODE=OCDD
   150.
   130.
   110.
S
C
H
E
tl
A
T
I
C

P
L
0
T
S
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
P
R
E
C

   50.0
   30.0
I    I
I    I
I  + I
*	*
I    I
+	+
  I
  I
  I
  I
  0

-------
                     FIGURE 2
ONGOING PRECISION AND RECOVERY OF LABELED AND NATIVE
                    COMPOUNDS

  FREQUENCY DISTRIBUTION PLOT OF ALL LABS COMBINED
                          134

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT « ALL LABS COMBINED
                                                         COOE=2378-TCDF-C13
   150.
   130.
S
C
H
E
tl
A
T
I
C  110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
                                                                                                                                                       LO
                                                                                                                                                       CO
                                                                     «-+-*

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED ArlPWKlIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                           CODE=Z37S-TCDF
   150.
   130.
   110.
F
0  90.0
H
   70.0
   50.0
   30.0
                                                                                                                                                         CO

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                         CODE=2378-TCDD-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
   90.0
   70.0
   50.0
   30.0
                                                                                                                                                       oc

-------
                                   ONGOING PRECISION AND RECOVERY -- LABELED ATOTlATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                        CODE=2378-TCDD-CL37
   150.
   130.
   110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
                                                                                                                                                      CO
                                                                                                                                                      CO

-------
                                   ONGOING PRECISION AMD RECOVERY — LABELED AMD NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                           COOE=2378-TCDD
   150.
   130.
   110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
tt-t-*
                                                                                                                                                        CO

-------
                                   ONGOING  PRECISION AND  RECOVERY — LABELED AHJWPtlVE COMPOUNDS
                                          FREQUENCY  DISTRIBUTION PLOT —  ALL LABS COMBINED
                                                        CODE=23<*78-PECDF-C13
   150.
   130.
S
C
H
E
n
A
T
I
c  no.
F
0  90.0
R
   70.0
   50.0
   30.0
                                                                                                                                                            «•."*(

-------
                                ONGOING PRECISION AND RECOVERY  —  LABELED  AND  NATIVE  COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT --  ALL  LABS COMBINED
                                                       CODE=23<*78-PECDF
150.
130.
110.
90.0
  0
  I
  I
  I
  I
+	+
I   I
I   I
*-+-*
I   I
+	+
  I
  I
  I
  0
70.0
50.0
30.0

-------
                                ONGOING PRECISION AND RECOVERY -- LABELED AJATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                    CODE=a34678-HXCDF-C13
150.
130.
110.
90.0
70.0
50.0
30.0

-------
                                   ONGOING PRECISION AND RECOVERY.--  LABELED AND NATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LA3S COMBINED
                                                         CODE=23467,l-HXCDF
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
0  90.0
R
   70.0
   50.0
   30.0
I    I
I    I
*	*
I  > I
I    I
I    I

-------
                                   ONGOING PRECISION AND RECOVERY ~ LABELED^BT NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123789--HXCDF-C13
   150.
   130.
S
C
H
E
M
A
T
I
c  no.

p
L
o
T
S
F
0
R

V
A
R
I
A
B
L
E
90.0
70.0
                                                                  *-+-*
P
R
E
C
   50.0
   30.0

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                         CODE=123789-HXCDF
   150.
   130.
   110.
f
0  90.0
R
   70.0
   50.0
   30.0
                                                                       I
                                                                       I
                                                                     •f	1
                                                                     I   I
                                                                     *	*

-------
                                   ONGOING PRECISION AKD RECOVERY —  LABELED ATWTATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                         CODE=123789-HXCDD
   150.
   130.
   110.
F
0  90.0
R
   70.0
   50.0
   30.0

-------
                                ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                     CODE=12378-PECDF-C13
150.
130.
110.
90.0
70.0
50.0
30.0

-------
                                   ONGOING PRECISION AND  RECOVERY --  LABELED AriVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                          CODE=12378-PECDF
   150.
   130.
S
C
H
E
M
A
T
I
c  no.
F
0  90.0
R
   70.0
   50.0
   30.0
  0
  I
  I
  I
  I
4	«•
I   I
»- + -»
I   I
+	+
  I
  I
  I
  0
  0
                                                                                                                                                      CO

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                        CODE=12378-PECDD-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0

-------
                                   ONGOING PRECISION AND RECOVERY  —  LABELED  A^^IATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL  LABS  COMBINED
                                                          CODE=12378-PECDD
   150.
   130.
   110.
F
0  90.0
R
   70.0
   50.0
   30.0
•f	+
I   I
*-+-*
I   I
I   I
I   I

-------
                                   ONGOING  PRECISION AND RECOVERY  ~  LABELED  AND  NATIVE COMPOUNDS
                                          FREQUENCY  DISTRIBUTION PLOT --  ALL  LABS COMBINED
                                                       COD E=12 36 78-HXCD F-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
*-+-*

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED
                                          FREQUENCY DISTRIBUTION PLOT — ALL
                                                         COOE=133678-HXCDF
                                                                              NATIVE COMPOUNDS
                                                                          LABS COMBINED
   150.
   130.
S
C
H
E
M
A
T
I
C

P
L
0
T
S

F
0
R

V
A
R
I
A
B
L
E
110.
                                                                                                                                                     10
90.0
I    I
*- + -*
I    I
70.0
P
R
E
C
   50.0
   30.0

-------
                                   ONGOING PRECISION AND  RECOVERY  —  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123678-HXCDD-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
0  90.0
R
   70.0
   50.0
   30.0
                                                                                                                                                     CO
                                                                                                                                                     to
                                                                    4	4
                                                                    *—*
                                                                    I   I
                                                                    4	4

-------
                                ONGOING PRECISION AND RECOVERY  --  LABELED  A^WIftTIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT —  ALL  LABS COMBINED
                                                      CODE=123678-HXCDO
150.
130.
110.
90.0
I    I
I    I
I  + I
*	*
I    I
I    I
•f	+
70.0
50.0
30.0

-------
                                ONGOING PRECISION AMD RECOVERY — LABELED  AND  NATIVE COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT --  ALL  LABS COMBINED
                                                    CODE=1234789-HPCDF-C13
150.
130.
110.
                                                                                                                                                LO
                                                                                                                                                LQ
90.0
70.0
                                                                 *—*
50.0
30.0

-------
                                   ONGOING PRECISION AND RECOVERY — LABELEB^TO NATIVE COMPOUNDS
                                         FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                        COOE=1234789-HPCDF
   150.
   130.
   no.
s
c
H
E
M
A
T
I
C

P
L
0
T
S
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
P
R
E
C

   50.0
   30.0
                                                                    I    I
                                                                    I    I
                                                                    I    I
                                                                    t --- +
                                                                      I
                                                                      I
CO
LO
•H

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED AND NATIVE COMPOUND?
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                       CODE=123478-HXCDF-C13
   150.
   130.
   110.
                                                                                                                                                       LO
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0

-------
p
R
E
C
                                   ONGOING PRECISION AND RECOVERY -- LABELE^fib NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                         CODE=123478-HXCDF
   150.
   130.
   110.
S
C
H
E
M
A
T
I
C

P
L
0
T
S
F
0  90.0
R
V
A
R
I
A
B
L
E
   70.0
   50.0
   30.0
I
I
*-+-*
I
I
I
*  I
                                                                                                                                                    CO
                                                                                                                                                    LO

-------
                                   ONGOING  PRECISION AMD  RECOVERY —  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLCT ~ ALL LABS COMBINED
                                                            = ia3478-IIXCDD-C13
   150.
   130.
S
c
H
E
n
A
T
I
C  110.
f
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
I    I
I    I
I    I
*	*
+	-f

-------
                                ONGOING  PRECISION AND  RECOVERY -- LABELED TWT NATIVE COMPOUNDS

                                       FREQUENCY  DISTRIBUTION PLOT ~ ALL LABS COMBINED

                                                       CODE=123<>78-HXCDD
150.
130.
110.
90.0
I    I
I    I
I    I
I    I

«	*
•f	+
70.0
50.0
30.0

-------
                                   ONGOING PRECISION AND RECOVERY ~ LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=1234678-HPCDF-C13
   150.
   130.
S
C
H
E
tt
A
T
I
C  110.
   90.0
V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
                                                                                                                                                      CD
                                                                     *—*

-------
                                   ONGOING PRECISION AND RECOVERY  «  LABELED TfBTNATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL  LABS COMBINED
                                                         CODE=123<«678-HPCDF
   150.
   130.
   110.
F
0  90.0
R
   70.0
   50.0
   30.0
+	*
I   I
X—f-*
I   I
I   I
I   I
4.	+
                                                                                                                                                     CD

-------
                                   ONGOING PRECISION AND RECOVERY —  LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=1234678-HPCDD-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
F
0  90.0
R
   70.0
   50.0
   30.0
CJD

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED^PB NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL  LABS COMBINED
                                                         CODE=123<*678-HPCDD
   150.
   130.
S
C
H
E
M
A
T
I
C  110.

P
L
0
T
S

F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
P
R
E
C

   50.0
   30.0
I    I

-------
                                   ONGOING PRECISION AND RECOVERY — LABELED  AMD  NATIVE  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL  LABS COMBINED
                                                             CODE=OCDF
   150.
   130.
S
C
H
E
M
A
T
I
C

P
L
0
T
S

F
0
R

V
A
R
I
A
B
L
E
110.
                                                                 *-+-*
90.0
10
CD
70.0
P
R
E
C
   50.0
   30.0

-------
p
R
E
C
                                   ONGOING PRECISION AND  RECOVERY  —  LABELED^WT NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL  LABS COMBINED
                                                           CODE=OCDD-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.

P
L
0
T
S

F
0  90.0
R
V
A
R
I
A
B
L
E
70.0
   50.0
   30.0
                                                                                                                                                CD
                                                                                                                                                CD
                                                                    *—*

-------
                                   ONGOING PRECISION ANO RECOVERY -- LABELED AND NATIVE COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                             CODE=OCDD
   150.
   130.
   110.
S
C
H
E
M
A
T
I
C

P
L
0
T
S
F
0  90.0
R

V
A
R
I
A
B
L  70.0
e
p
R
E
C

   50.0
   30.0
                                                                     I    I
                                                                     I    I
                                                                                                                                                   CD

-------
                                           -,     • ";".--. '. 3
                  FIGURES
PRECISION AND RECOVERY OF LABELED COMPOUNDS



     FREQUENCY DISTRIBUTION PLOT BY LAB
                          168

-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                               FREQUENCY DISTRIBUTIOf4 PLOT — BY LAB
                                                         CODE=Z378-TCDF-C13
   150.
   130.
S
C
H
E
M
A
T
I
c  no.
   90.0
   70.0
   50.0
   30.0
  LAB
  LAB
 i    i
 i    i
 i    i
-*-+-*-
  LABI
—o —
 LAB2
                                                              LAB3

-------
                                             PRECISION AND  RECOVERY OF LABE LEO^WftPOUNDS
                                               FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                         CODE=2378-TCDD-C13
   150.
   130.
   110.
F
0  90.0
R
   70.0
   50.0
                                       I    I
                                       I    I
                                         [Ail-
—o—
 LAB2
                                                                                                      I

-------
                                             PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                                FREQUENCY DISTRIBUTE IN PLOT -- BY LAB
                                                         CODE=2378-Tt OD-CL37
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
   90.0
   70.0
   50.0
                                                                      •f	4-
                                        I    I
                                        I    I
                                       "'LAB!"
  o
—o—
 LAB2
                                                                                                      t
                                                                                                    *	
-LAB3-

-------
                                          PRECISION AND RECOVERY OF LABELED^PGHPOUNDS
                                             FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                      CODE=23<»78-PECDr-C13
150.
130.
110.
                                                                                                                                                   CM
                                                                                                                                                   r-
90.0
70.0
                                      *
                                      *
                                      0
                                      0
•f	•»•
                                                                  *-+-*
                                                                                                 *-+_*
50.0
                                                                    LAB2
                                                                                                   te3

-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                               FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                       COD E=Z346 78-HXCD F-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
   90.0
   70.0
   50.0
I    I
I    I
I    I
*-*-*
I    I
I    I
                                                                                         *-+-*
                                                  LAB2
                                                                                          LAB3

-------
                                            PRECISION AND RECOVERY OF LABELED^OMPCUNDS
                                               FREQUENCY DISTRIBUTION PLOT —' BY LAB
                                                      CODE=1237S9-HXCDF-C13
   150.
   130.
S
C
H
E
II
A
T
I
C  110.
F
0  90.0
R
   70.0
   50.0
                                       Oil-
  *
--*	
 UB2
-— I
"UBS"

-------
                                         PRECISION AND RECOVERY OF LABELED   COMPOUNDS
                                            FREQUENCY DISTRIBUTION PLOT --  BY  LAB
                                                     COOE=12378-PECDF-C13
150.
130.
110.
90.0
70.0
50.0
                                      *
                                      ft
                                                                                                                                                LO
                                                                                                                                                r-
                                                                                               ft~f-»
                                     LABI
                                                                   LAB2
 +	1
   I
   I
   I
   I
~~LAB3~

-------
                                            PRECISION AND RECOVERY OF LABELED
                                               FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                        CODE=13378-PECDD-C13
   150.
   130.
   110.
F
0  90.0
R
   70.0
   50.0
                                                                                                                                                    CD
                                                                                                   *-+-*
                                       I    I
                                       I    I
                                       I    I
                                      •'LABI'
 0
 0

 0

LABE
 •f	
   1

"U

-------
                                             PRECISION AND  RECOVERY OF LABELED  COMPOUNDS
                                                FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                        COD E=12 36 78-HXCD F-C13
   150.
   130.
   110.
                                                                                                                                                        f>
f
0  90.0
R
   70.0
   50.0
                                                                                                    +	1
                                         Ail'
                                        LAB
—0—
 LAB2
"LA!
                                                                                                       B3

-------
                                             PRECISION AND  RECOVERY Or LABELED^CflPOUNDS
                                                FREQUENCY DISTRIBUTION PLOT -- BY LAB
                                                        CODE=1E3678-HXCDD-C13
   150.
   130.
   110.
                                                                      t	1
F
0  90.0
R
   70.0
   50.0
                                                                                                    *	*
                                       I   I
                                       I   I
                                       I   I
                                       I   I
 0

LAB2
                                                                                                     tiii"
                                        LABI

-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                               FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                       CODE=1Z34789-HPCDF-C13
   150.
   130.
S
c
H
E
M
A
T
I
C  110.
F
0  90.0
R
   70.0
   50.0
•f	+





1
1
1
1
+ 1
_l_l 	



^




^ 	 H
1



i-

                                                                                                   *-+-*
                                                                                                                                                      CD
                                                                      LAB2
                                                                                                     AB3

-------
                                         PRECISION AND RECOVERY OF LABELED^fflPOUNDS
                                            FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                    COOE=123478-HXCOF-C13
150.
130.
no.
90.0
70.0
50.0
                                    *-+-*
                                                                   LAB2
                                                                                                LA83

-------
                                         PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                            FREQUENCY DISTRIBUTION PLOT — BY LAB
                                                    CODE=123478-HXCDD-C13
150.
130.
110.
90.0
70.0
50.0
1 1
I 1
1 1
1 I
	 	 1 _*-| 	


                                                                                                                                                       CO
                                     LAB
                                                                   LAB2
                                                                                                 LAB3

-------
                                            PRECISION AND  RECOVERY OF  LABELED
                                               FREQUENCY DISTRIBUTION  PLOT — BY LAB
                                                       CODE=1234676-HPCDF-C13
   150.
   130.
S
C
H
E
ft
A
T
I
c  no.
   90.0
   70.0
   50.0
                                                                                                                                                     cvj
                                                                                                                                                     oo
                                                                    *—*
                                                                                                  *-+-*
                                                                                                  +---*
*	tt-
 LAB1
                                                                    —o —
                                                                     LAB2
TAB!

-------
                                         PRECISION AND RECOVERY OF  LABELED  COMPOUNDS
                                            FREQUENCY DISTRIBUTION  PLOT  —  BY  LAB
                                                    CODE=1234678-HPCDD-C13
150.
130.
110.
                                                                                                                                                 CO
                                                                                                                                                 CO
90.0
70.0
50.0
[Ail
                                                                  UB2
                                                                                                 I
                                                                                                 I
                                                                                                 I
                                                                                                LABS"

-------
                                             PRECISION AND RECOVERY CH: LABELED
                                                FREQUENCY  DISTRIBUTION PLOT — BY
                                                            CODE=OCOD-C13
 IPOUNDS
LAB
   150.
   130.
   110.
f
0  90.0
R
   70.0
   50.0
                                                                                                                                                       00
                                                                                                    4	+
                                       •f—
                                       I
                                                                       LAB2
                                                                                                      Jii"
                                                                                                     LAB

-------
                                             ^
                   FIGURE 4
  PRECISION AND RECOVERY OF LABELED COMPOUNDS



FREQUENCY DISTRIBUTION PLOT OF ALL LABS COMBINED
                        185

-------
                                         PRECISION AND RECOVERY CP LABELED^BtlPOUNDS
                                       FREQUENCY DISTRIBUTION PLCT — ALL LABS COMBINED
                                                      CODE=2378-1CDF-C13
150.
130.
110.
                                                                                                                                                CO
90.0
70.0
50.0
30.0

-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                         CODE=2378-TCOD-C13
   150.
   130.
   110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
                                                                                                                                                   CO

-------
                                         PRECISION AND RECOVERY  OF  LABE LED^PBMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT  —  ALL  LABS COMBINED
                                                     CODE=2378-TCOD-CL37
ISO.
130.
110.
                                                                                                                                                CO
                                                                                                                                                co
90.0
70.0
50.0
30.0

-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                        CODE=23<*7fl-PECDF-C13
   150.
   130.
S
C
H
E
M
A
T
I
C  110.
   90.0
   70.0
   50.0
   30.0
                                                                                                                                                CD
                                                                                                                                                CO

-------
                                            PRECISION AND RECOVERY OF LABELED "
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=234678-HXCDF-C13
   150.
   130.
   110.
F
0  90.0
R
   70.0
   50.0
   30.0
                                                                                                                                                     O
                                                                                                                                                     CD
                                                                     *-+-*

-------
                                         PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                    COOE=123789-HXCDF-C13
150.
130.
110.
90.0
70.0
50.0
30.0

-------
                                            PRECISION AND RECOVERY OF LABELED^ElPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                        CODE=1237e-PECDF-C13
   150.
   130.
   110.
0  90.0
R
   70.0
   50.0
   30.0
                                                                                                                                                       (73

-------
                                         PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                       FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                     CODE=1Z378-PECDD-C13
150.
130.
110.
                                                                                                                                                      CO
                                                                                                                                                      a:
90.0
70.0
50.0
30.0

-------
                                            PRECISION AND RECOVERY OF LABELE0^5(1POUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123678-HXCDF-C13
   150.
   130.
S
c
H
E
M
A
T
I
C  110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
                                                                                                                                                       CD

-------
                                            PRECISION AND RECOVERY  OF  LABELED  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT  —  ALL LABS COMBINED
                                                       CODE=123678-HXCDO-C13
   150.
   130.
   110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0
                                                                                                                                                       LO
                                                                                                                                                       CO
*—*

-------
                                            PRECISION AND RECOVERY OF LABELED^RMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=1334789-HPCDF-C13
   ISO.
   130.
S
C
H
E
M
A
T
I
C  110.
F
0  90.0
R

V
A
R
I
A
B
L  70.0
E
   50.0
   30.0


-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT ~ ALL LABS COMBINED
                                                       CODEM23478-HXCDF-C13
   150.
   130.
S
C
H
E
(1
A
T
I
C  110.
   90.0
   70.0
   50.0
   30.0

-------
                                            PRECISION AND RECOVERY OF LABELEB^COMPOUHDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=123478-HXCDD-C13
   150.
   130.
S
C
H
E
M
A
T
I
C

P
L
0
T
S

F
0
R

V
A
R
I
A
B
L
E
110.
                                                                                                                                                  CO
90.0
70.0
   50.0
   30.0

-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                       CODE=1E34678-HPCDF-C13
   150.
   130.
S
c
H
E
M
A
T
I
C  110.
F
0  90.0
R
   70.0
   50.0
   30.0
                                                                                                                                                        03
                                                                     +—*

-------
                                            PRECISION AND RECOVERY OF LABELgS^OMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT — ALL LABS COMBINED
                                                       CODE=1234678-HPCDD-C13
   150.
   130.
S
C
H
E
M
A
T
I
C

P
L
0
T
S

F
0
R

V
A
R
I
A
B
L
E
110.
                                                                                                                                                   o
                                                                                                                                                   CM
90.0
70.0
                                                                     *-+-*
P
R
E
C
   50.0
   30.0

-------
                                            PRECISION AND RECOVERY OF LABELED  COMPOUNDS
                                          FREQUENCY DISTRIBUTION PLOT -- ALL LABS COMBINED
                                                           CODE=OCDD-C13
   150.
   130.
S
C
H
E
n
A
T
I
C  110.
   90.0
   70.0
   50.0
   30.0
                                                                                                                                                       O
                                                                                                                                                       CM
                                                                     *-+-*

-------
             USEPA OFFICE OF WATER
STUDY PLAN FOR THE EVALUATION OF METHOD 1613
                      May 1990
                 USEPA Office of Water
          Office of Water Regulations and Standards
           Industrial Technology Divsion (WH-552)
                 Washington, DC 20460

-------
                                  TABLE OF CONTENTS








Study Plan                                                           Section 1





Data Reporting Forms                                                 Section 2





Description of Quality Control Solutions                                   Section 3





Description of PE Spike Solutions                                        Section 4
                                          203

-------
                 SECTION 1





STUDY PLAN FOR THE EVALUATION OF EPA METHOD 1613
                   204

-------
Introduction

       The USEPA Office of Water intends to promulgate Method 1613 on a nationwide basis under the
Clean Water Act (as amended, 1987) for the determination of polychlorinated dibenzo-p-dioxins and furans
(dioxins and  furans).   The objective of this  study is  to evaluate the use  of  high resolution  gas
chromatography  mass spectrometry (HRGC/HRMS)  for  the  determination  of dioxins and furans in
industrial effluents.  The data from the study will be used to assess the performance of the method, to
determine the intra-laboratory  and inter-laboratory components of variability for  the method, and to
generate improved method specifications.  Laboratories successfully completing this study will be qualified
to perform Method 1613 for the Agency and other clients.

       Because of the problems associated with shipping  large volumes of wastewater effluent to all
participants, the Agency has prepared a spiking solution to be added to reagent water.  This spiking solution
resulted from the batch extraction  of wastewaters collected from several pulp and paper  facilities  and
combined prior to extraction.

Sample Preparation

       Each laboratory will  receive two (2) flame-sealed ampules containing the spiking solutions, along
with specific instructions  for preparing two (2) synthetic samples (the "Performance Samples") from the
solutions.

Extraction and Analysis

       Each prepared sample will be spiked with the 15 labeled analogs of the 2,3,7,8-substituted PCDDs
and PCDFs listed in Table 1 of Method 1613. After spiking, the samples will be extracted according to the
procedures, in Method 1613, filtering water samples before extraction.  The particulates collected on the
filter are to be extracted separately, using the Soxhlet-Dean Stark apparatus described in the method.  The
filtered water is extracted in a separatory funnel, and combined with the paniculate extract after solvent
exchange.

       The remainder of the analytical procedures are detailed in the method itself, and must be followed as
written.

Quality Assurance/Quality Control

       As with all the 600 and 1600 series EPA methods, there are a series of prescribed QA/QC analyses
that  are essential to the process.   This  study seeks to  evaluate these aspects of  the method as well.
Therefore, participating  laboratories must be prepared  to provide  data for  the following non-sample
analyses:

        1.     Start-up tests in  Section 8.2 (Initial Precision and Accuracy) and Ongoing Precision  and
              Accuracy (Section 14.5)

            The Initial Precision and Accuracy test consists of the analysis of four aliquots of reagent water
            spiked with the native and labeled analytes listed in Table 1 (PAR solution, Sections 6.14, and
            8.2 of the method).  Note that these analyses  are required  in addition to the two samples
            prepared from  the pulp  and paper spiking solution.   They are used  to demonstrate the
            laboratory's ability to generate data of acceptable initial precision and recovery (IPR) in a
            reference matrix  Laboratories who have generated IPR data previously through the use of
            Method 1613. may submit  those data.
                                                         £ U O

-------
            In addition, if the analyses of standards and samples for this study continue beyond a 12-hour
            shift beginning with the injection of the first calibration standard, the Ongoing Precision and
            Accuracy analysis  is required. This test demonstrates that ongoing precision and recovery
            (OPR) meets the QC limits specified in the Method (see Section 14.5). If the analyses of the
            standards and samples above can  be completed in a single 12-hour time period, no OPR
            analysis is required.

        2.     Method blank in  Section 10.4.4

            This reagent water aliquot spiked with the labeled analogs in Table 1 serves as a check on
            contamination.  One blank must  be extracted along with the Performance Samples, and
            submitted with that data.

        3.     Initial Calibration and Calibration Verification Data  in Sections 7.5 and 14.1

            A five point initial calibration must be performed using the compounds and concentrations
            listed in Table 4 of the method. Unless all the samples and blanks, including the start-up test
            aliquots, are analyzed in the 12 hour period  immediately after the initial calibration, a
            calibration verification standard must be analyzed at the begining of the 12 hour shift, as
            described in Section 14.1.

Deliverables

      The deliverables for this study consist of the following materials:

        1.     A narrative discussion of the data outlining any problems encountered during the analyses,
              and detailing all steps taken to overcome them. Additional comments on the method write-
              up and specifications are welcome.

        2.     Sample  concentration  data  calculated  according  to  the  specifications in the  method,
              recoveries of labeled compounds in all samples and standards, and calibration data, including
              response factors for all native and labeled compounds. Draft forms have been provided with
              this outline. These forms may be photocopied and completed by hand, or the laboratory may
              substitute forms in another format, so long as all data elements contained on the draft forms
              appear in the other format as well.

        3.     Raw data, in the form of selected ion current  profiles, data system reports, or handwritten
              tabulations of instrument data, containing peak areas, for all analyses, including  calibration
              standards, start-up  test aliquots, samples, and  blanks.  These data, in conjunction with the
              forms above, must be sufficient to allow an  independant reviewer to  reconstruct all
              calculations and quantitations performed by the laboratory.

        4.     Copies of log  books,  bench notes, etc., detailing the processing of samples and  sample
              extracts.

      The data described above  should be sent to:

              USEPA Sample Control Center
              P. O. Box 1407
              Alexandria, VA 22313
              USA

            or, if sent by overnight courier, use street address of:
                                                           0 *i
                                                           \j o

-------
              USEPA Sample Control Center
              300 North Lee Street
              Alexandria, VA 22314
              USA

Statistical Analysis

      Standard statistical methods for the analysis of collaborative tests will be used to analyze the data
submitted by the laboratories. Samples will be coded and their assignment to laboratories randomized so
that the identity of each sample and its contents will be unknown to the laboratory performing the analysis.
Statistical procedures (see  Youden,  WJ., Statistical Techniques for Collaborative Tests. Association of
Official Analytical Chemists,  1973) will  be used to estimate intra-laboratory and  inter-laboratory
components of variability and assess precision and accuracy for different concentration levels.

Questions

      Questions regarding participation in the study or the method in particular should be addressed to Or.
Harry McCarty, Sample Control Center, at the above address, by telephone at 703-557-5040, or by telefax at
703-683-0378.

-------
                       SECTION 2





DATA REPORTING FORMS FOR THE EVALUATION OF EPA METHOD 1613
                             :08

-------
 USEPA Industrial Technology Division
 Sample Control  Center
 P. O. Box 1407  - Alexandria, VA  22313        ,,
 703/557-5040    FTS  8-557-5040             ;1


                    COVER PAGE - PCDD/PCDF ANALYSES DATA  PACKAGE


 Lab Name: 	  Contract No.:  	  SAS No.:

 Episode No.: 	    EPA Method No.: 	    Method Issue/Rev. Date;

 Industrial Category: 	   Program:  	


               EPA Sample No.           Lab Sample  ID
Comments:  Narrative Report is attached.  	  (Yes)


I certify that  this data package  is  in  compliance with the terms and
conditions of the contract,  both  technically and for completeness, for
other than the  conditions detailed in the Narrative Report.  Release of
the data contained in this hardcopy  data package (and in the data submitted on
magnetic media,  if data  is submitted on magnetic media) , has been
authorized by the Laboratory Manager or the Manager's designee, as verified
by the following signature.


Signature:  	   Name:   	

Date:       	   Title:
                                                                        3/90
                                           2 Of)

-------
                                    USEPA — ITD                        • '        J

                                      FORM 1A                  EPA SAMPLE NO.
                          PCDD/PCDF ANALYSIS DATA SHEET	
                         Use for Sample and Blank Results
 Lab Name: 	    Episode No.:
 Contract No.: 	   SAS No.: 	    Lab Sample ID:
 Matrix (aqueous/solid/leachate): 	 Sample Wt/Vol:   •	 g or mL:  	

 Sample Receipt Date: 	          Initial Calibration Date: 	

 Ext.  Date: 	 Shift: 	                  Instrument ID:  	

 Analysis Date: 	 Time: 	                GC Column ID:  	

 Extract Volume (uL): 	             Sample Data Filename: 	

 Injection Volume (uL):  	            Blank Data Filename: 	

 Dilution Factor:  	            Cal.  Ver. Data Filename:
 Concentration Units (pg/L or ng/Kg dry weight):  	        % Solids:
                   CONCENTRATION        DETECTION       ION ABUND.     RRT
   ANALYTE             FOUND              LIMIT         RATIO (1)      (1)

 2378-TCDD          	       	       	       	
 12378-PeCDD        	       	       	       	
 123478-HxCDD       	._       	       	       	
 123678-HxCDD       	       	       	       	
 123789-HXCDD       	       	       	       	
 1234678-HpCDD      	       	       	       	
 OCDD                	       	       	       	
 2378-TCDF          	       	       	       	
 12378-PeCDF        	       	       	       	
 23478-PeCDF        	       	       	       	
 123478-HXCDF       	       	       	       	
 123678-HxCDF       	       	       	       	
 123789-HxCDF       	       	       	       	
 234678-HxCDF       	"       	       	       	
 1234678-HpCDF      	       	       	       	
 1234789-HpCDF      	       	       	       	
 OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
(1)  Contract-required limits for RRTs and ion abundance ratios are specified
    in Tables 2 and 3A, respectively, Method 1613.


                                                                         4/90

-------
                                    USEPA — ITD

                                     FORM IB                   EPA SAMPLE NO.
                   'PCDD/PCDF CONFIRMATION ANALYSIS DATA SHEET
 Lab Name: 	    Episode No.:
 Contract No.: 	   SAS No.: 	    Lab Sample ID:
 Matrix (aqueous/solid/leachate): 	 Sample Wt/Vol: 	 g or mL: 	

 Sample Receipt Date: 	          Initial Calibration Date:  	

 Ext. Date: 	 Shift: 	                  Instrument ID:  	

 Analysis Date: 	 Time: 	                GC Column ID:  	

 Extract Volume (uL): 	             Sample Data Filename:  	

 Injection Volume (uL): 	            Blank Data Filename:  	

 Dilution Factor:  	        Cal. Ver. Data Filename:  	

 Concentration Units (pg/L or ng/Kg dry weight):  	        % Solids: 	
                   CONCENTRATION        DETECTION       ION ABUND.    RRT
   ANALYTE             FOUND              LIMIT         RATIO  (1)     (1)

 2378-TCDD          	       	      	      	

 12378-PeCDD
 123478-HxCDD
 123678-HxCDD
 123789-HXCDD

 1234678-HpCDD

 OCDD

 2378-TCDF

 12378-PeCDF
 23478-PeCDF

 123478-HXCDF
 123678-HxCDF
 123789-HxCDF
 234678-HxCDF

 1234678-HpCDF
 1234789-HpCDF

OCDF
(1) Contract-required  limits  for RRTs and ion abundance ratios are specified
    in Tables 2 and 3A, respectively, Method 1613.

                                                                         4/90

                                      21.1

-------
                                                .  . -.•„.        -     ••-.-. I'- • - • ..-• •#•••••• • t

                                    DSEPA - ITD

                                      FORM 2
                        PCDD/PCDF LABELED COMPOUND AND         EPA SAMPLE1 NO.
                          CLEANUP STANDARD RECOVERIES          	
 Lab Name: 	    Episode No.:
 Contract No.: 	   SAS No.: 	    Lab Sample ID:
 Matrix (aqueous/solid/leachate) : 	 Sample Wt/Vol: 	 g or mL: 	

 Sample Receipt Date: 	          Initial Calibration Date: 	

 Ext.  Date:  	 Shift: 	                  Instrument ID: 	

 Analysis  Date: 	 Time: 	                GC Column ID: 	

 Extract Volume (uL): 	             Sample Data Filename: 	

 Injection Volume (uL):  	            Blank Data Filename: 	

 Dilution  Factor:                      Cal.  Ver.  Data Filename:
 Concentration Units (pg/L or ng/Kg dry weight):  	        % Solids:
                                                                ION
                           SPIKE           CONC.         R(%)    ABUND.    RRT
                           CONC.            FOUND         (1)    RATIO (2)  (2)
LABELED COMPOUNDS

 13C-2378-TCDD          	   	    		   	
 13C-12378-PeCDD       	.        	    	    	   	
 13C-123478-HXCDD      	•_		    	    		
 13C-123678-HXCDD      	   	    	    	   	
 13C-1234678-HpCDD     	   	    	    	   	
 13C-OCDD
 13C-2378-TCDF
 13C-12378-PeCDF
 13C-23478-PeCDF
 13C-123478-HXCDF
 13C-123678-HXCDF
 13C-123789-HXCDF
 13C-234678-HXCDF
 13C-1234678-HpCDF
 13C-1234789-HpCDF

CLEANUP STANDARD

 37C14-2378-TCDD
(1)  Contract-required limits for percent recovery  (R)  are  25-150%  (Section
    8.3.3, Method 1613).
(2)  Contract-required limits for RRTs and ion abundance  ratios  are specified
    in Tables 2 and 3A, respectively, Method 1613.
                                                                          4/90


                                        1 '>
                                        -L »„

-------
                                   USEPA - ITD             ;    •";-

                                     FORM 3A
                 PCDD/PCDF INITIAL CALIBRATION RELATIVE RESPONSES
 Lab Name
          Episode No.:
 Contract No.:
SAS No.:
 Initial Calibration Date:

 Instrument ID:  	
 CS1 Data Filename:

 CS2 Data Filename:

 CS3 Data Filename:
NATIVE ANALYTES

 2378-TCDD

 12378-PeCDD

 123478-HXCDD
 123678-HXCDD
 123789-HxCDD  (2)

 1234678-HpCDD

 OCDD

 2378-TCDF

 12378-PeCDF
 23478-PeCDF

 123478-HxCDF
 123678-HxCDF
 123789-HXCDF
 234678-HxCDF

 1234678-HpCDF
 1234789-HpCDF

 OCDF (3)
   GC Column ID:
        CS4 Data Filename:

        CSS Data Filename:
                              RELATIVE RESPONSE  (RR)
                               MEAN
                                RR
                      CS1
CS2
CS3
CS4
CSS
  Cv
(%RSD)
 (1)
(1)  For contract Cv specifications, see Section 7.5.4,  Method 1613.
(2)  Response Ratios are calculated relative to the labeled analogs of  the
    other two HxCDDs (Section 16.1.2,  Method 1613).
(3)  Response Ratios are calculated relative to the labeled analog of OCDD
    (Section 16.1.1, Method 1613).
                                                                        4/90
                                         213

-------
 Lab Name:
                  DSEPA - ITD

                    FORM 3B
 PCDD/PCDF INITIAL CALIBRATION RESPONSE FACTORS


	    Episode No.: 	
 Contract No.:
            SAS No.:
 Initial Calibration Date:

 Instrument ID:  	
 CS1 Data Filename:

 CS2 Data Filename:

 CS3 Data Filename:
              GC Column ID:
                    CS4 Data Filename:

                    CSS Data Filename:
                             RESPONSE FACTOR  (RF)
                                           MEAN
                                            RF
 LABELED COMPOUNDS

  13C-2378-TCDD

  13C-12378-PeCDD

  13C-123478-HXCDD
  13C-123678-HXCDD

  13C-1234678-HpCDD

  13C-OCDD

  13C-2378-TCDF

  13C-12378-PeCDF
  13C-23478-PeCDF

  13C-123478-HXCDF
  13C-123678-HXCDF
  13C-123789-HXCDF
  13C-234678-HXCDF

  13C-1234678-HpCDF
  13C-1234789-HpCDF

CLEANUP STANDARD

37C14-2378-TCDD
                      CS1
            CS2
CS3
CS4
CSS
 CV
(%RSD)
 (1)
(1)  For assignment of labeled compounds to internal standards,  see Table  2.   F
    contract Cv specifications,  see Section 7.6.3,  Method 1613.
                                                                         4/90
                                       214

-------
 Lab Name:
                    •'••"" .:,: ...        USBPA - ITD

                                       FORM 3C
                 PCOD/PCDF INITIAL CALIBRATION ION ABUNDANCE RATIOS
                    Episode No.:
Contract No . :
SAS No.:

Initial Calibration Date:
Instrument ID:
CS1 Data Filename:
CS2 Data Filename:
CS3 Data Filename:
NATIVE ANALYTES
2378 TCDD
12378 PeCDD
123478 HxCDD
123678 HXCDD
123789 HxCDD
1234678 HpCDD
OCDD
2378 TCDF
12378 PeCDF
23478 PeCDF
123478 HxCDF
123678 HxCDF
123789 HXCDF
234678 HxCDF
1234678 HpCDF
1234789 HpCDF
GC Column ID:
CS4 Data Filename:
CSS Data Filename:

M/Z'S ION ABUNDANCE RATIO
FORMING
RATIO (1) CS1 CS2 CS3 CS4
M/M+2
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M/M+2
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4


QC
LIMITS
CSS (2)
0.65-0.89
1.32-1.78
1.05-1.^
1.05-l.M
1.05-1.40
0.88-1.20
0.76-1.02
0.65-0.89
1.32-1.78
1.32-1.78
1.05-1.43
1.05-1.43
1.05-1.43
1.05-1.43
0.88-1.20
0.88-1.20
 OCDF
M+2/M+4
0.76-1.02
(1)  See Table  3,  Method 1613, for m/z specifications.

(2)  Ion Abundance Ratio Control Limits from Table 3A, Method  1613.
                                       91 r;i
                                       >i. J. U
                                                                           4/90

-------
                                   USEPA - ITD
 Lab Name:
                                     FORM 3D
                PCDD/PCDF INITIAL CALIBRATION ION ABUNDANCE RATIOS
           Episode No.:
 Contract No.:
SAS No.:
 Initial  Calibration Date:

 Instrument ID:       	
   GC Column ID:
 CS1  Data Filename:

 CS2  Data Filename:

 CS3  Data Filename:
 LABELED COMPOUNDS

 13C-2378-TCDD

 13C-12378-PeCDD

 13C-123478-HXCDD
 13C-123678-HXCDD

 13C-1234678-HpCDD

 13C-OCDD

 13C-2378-TCDF

 13C-12378-PeCDF
 13C-23478-PeCDF

 13C-123478-HXCDF
 13C-123678-HXCDF
 13C-123789-HXCDF
 13C-234678-HXCDF

 13C-1234678-HpCDF
 13C-1234789-HpCDF
         CS4  Data Filename:

         CSS  Data Filename:
M/Z'S
FORMING
RATIO (1) CS1
M/M+2
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M/M+2
M+2/M+4
M/M+2
M/M+2
M/M+2
M/M+2
M/M+2
M/M+2
ION ABUNDANCE RATIO QC
LIMITS
CS2 CS3 CS4 CSS (2)
0.65-0.89
1.32-1.78
1.05-1.43
1.05-1.43
0.88-1.20
0.76-1.02
0.65-0.89
1.32-1.78
1.32-1.78
0.43-0.59
0.43-0.59
0.43-0.59
0.43-0.59
0.37-0.51
0.37-0.51
(1)  See Table 3,  Method 1613,  for m/z specifications.

(2)  Ion Abundance Ratio Control Limits from Table 3A,  Method 1613
                                        216
                                                                         4/90

-------
 Lab Name:
           USEPA - ITD   '"•

             FORM 4A
PCDD/PCDF CALIBRATION VERIFICATION

               Episode No.
Contract No . :

SAS No.:

Initial Calibration Date:
Instrument ID:
VER Data Filename:
NATIVE ANALYTES
2378 TCDD
12378 PeCDD
123478 HxCDD
123678 HXCDD
123789 HXCDD
1234678 HpCDD
OCDD
2378 TCDF
12378 PeCDF
23478 PeCDF
123478 HXCDF
123678 HXCDF
123789 HXCDF
234678 HXCDF
1234678 HpCDF
1234789 HpCDF
OCDF


M/Z'S
FORMING
RATIO (1)
M/M+2
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M/M+2
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
GC Column ID:
Analysis Date:
ION QC
ABUND. LIMITS
RATIO (2)
0.65-0.89
1.32-1.78
1.05-1.43
1.05-1.43
1.05-1.43
0.88-1.20
0.76-1.02
0.65-0.89
1.32-1.78
1.32-1.78
1.05-1.43
1.05-1.43
1.05-1.43
1.05-1.43
0.88-1.20
0.88-1.20
0.76-1.02

Time:
CONC.
CONC. RANGE (3)
FOUND (ng/mL)
8.6-11.6
44.2-56.6
37.6-66.5
39.7-63.0
42.6-58.7
41.6-60.2
87.5-114.4
8.8-11.3
46.7-53.5
47.2-53.0
41.5-60.2
40.5-61.7
45.7-54.5
44.1-56.7
43.1-58.0
43.6-57.3
83.9-119.2
(1)  See Table 3, Method 1613, for m/z specifications.

(2)  Ion Abundance Ratio Control Limits as specified in Table 3A,  Method 1613.

(3)  Contract-required concentration range as specified in Table 7,  Method
    1613,  under VER.
                                       217
                                                                         4/90

-------
 Lab Name:
            USEPA -MTD

              FORM 4B

 PCDD/PCDF CALIBRATION VERIFICATION

	    Episode No.: 	
 Contract No.:
      SAS No.:
 Initial Calibration Date:

 Instrument ID: 	
        GC Column ID:
 VER Data Filename:
            Analysis Date:
Time:
LABELED COMPOUNDS
13C-2378-TCDD
13C-12378-PeCDD
13C-123478-HXCDD
13C-123678-HXCDD
13C-1234678-HpCDD
13C-OCDD
13C-2378-TCDF
13C-12378-PeCDF
130-23478-PeCDF
13C-123478-HXCDF
13C-123678-HXCDF
13C-123789-HXCDF
13C-234678-HXCDF
13C-1234678-HpCDF
13C-1234789-HpCDF
M/Z'S
FORMING
RATIO (1)
M/M+2
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M+2/M+4
M/M+2
M+2/M+4
M+2/M+4
M/M+2
M/M+2
M/M+2
M/M+2
M/M+2
M/M+2
ION QC
ABUND. LIMITS CONC.
RATIO (2) FOUND
0.65-0.89
1.32-1.78
1.05-1.43
1.05-1.43
0.88-1.20
0.76-1.02
0.65-0.89
1.32-1.78
1.32-1.78
0.43-0.59
0.43-0.59
0.43-0.59
0.43-0.59
0.37-0.51
0.37-0.51
CONC.
RANGE (3)
(ng/mL)
90.0-111.2
80.6-124.0
76.1-131.3
84.0-119.1
82.2-121.6
164.2-243.6
87.7-114.0
81.8-122.3
83.0-120.5
85.2-117.4
85.0-117.7
89.5-111.7
85.7-116.7
88.5-113.1
89.0-112.4
CLEANUP STANDARD

 37C14-2378-TCDD  (4)
                                         24.4-46.4
(1)  See Table 3, Method 1613, for m/z specifications.

(2)  Ion Abundance Ratio Control Limits from-Table 3A, Method 1613.

(3)  Contract-required concentration range, as specified in Table 7,  Method
    1613,  under VER.
(4)  No ion abundance ratio; report concentration found.
                                       218
                                                                         4/90

-------
Lab Name:
                    JJSBPA. --ITD

                        FORM 5
 PCDD/FCDF ,ST WINDOW AND ISOMER SPECIFICITY STANDARDS

	   Episode No.: 	
Contract No.:

Instrument  ID:
               SAS No.
RT Window  Data Filename:

DB-5 IS Data  Filename:
DB-225 IS Data  Filename:
                   Init. Calibration Date:

                  	    Analysis Date:

                  	    Analysis Date:

                  	  Analysis Date:
                      DB-5 RT WINDOW DEFINING STANDARDS RESULTS
     ISOMERS

   1368-TCDD  (F)
   1289-TCDD  (L)

   12479-PeCDD  (F)
   12389-PeCDD  (L)

   124679-HxCDD  (F)
   123467-HxCDD  (L)

   1234679-HpCDD  (F)
   1234678-HpCDD  (L)
         ABSOLUTE
            RT
  ISOMERS

1368-TCDF  (F)
1289-TCDF  (L)

13468-PeCDF  (F)
12389-PeCDF  (L)

123468-HXCDF (F)
123489-HXCDF (L)

1234678-HpCDF  (F)
1234789-HpCDF  (L)
                   Time:

                   Time:

                   Time:
        ABSOLUTE
           RT
   (F)  = First eluting isomer  (DB-5);  (L) = Last eluting isomer (DB-5).
                 ISOMER SPECIFICITY (IS) TEST STANDARDS RESULTS
   ISOMERS

  1234-TCDD
  1278-TCDD

  1278-TCDD
  1478-TCDD

  1478-TCDD
  1237-TCDD

  1237-TCDD
  1238-TCDD
 %  VALLEY HEIGHT
    BETWEEN
 COMPARED PEAKS (1)
 ISOMERS

1238-TCDD
2378-TCDD

2347-TCDF
2378-TCDF

2378-TCDF
1239-TCDF
% VALLEY HEIGHT
   BETWEEN
COMPARED PEAKS (1)
  (1) To meet  contract requirements,  % Valley Height Between Compared
      Peaks shall  not exceed 25% (Section 14.4.2.2,  Method 1613).
                                                                         4/90
                                         21f)

-------
                                :--    USEPA - ITO

                                      FORM 6A
                        PCDD/PCDF RELATIVE RETENTION TIMES
 Lab Name:
               Episode No.:
 Contract No.:

 Instrument ID:

 Analysis Date:
     SAS No.:
           Init. Cal. Date:
       GC Column ID:
Time:
CS3 or VER Data Filename:
                Compounds Using 13C-1234-TCDD as Internal Standard
     NATIVE ANALYTES

      2378-TCDF
      2378-TCDD
      12378-PeCDF
      23478-PeCDF
      12378-PeCDD

     LABELED COMPOUNDS

      13C-2378-TCDF
      13C-2378-TCDD
      37C-2378-TCDD
      13C-12378-PeCDF
      13C-23478-PeCDF
      13C-12378-PeCDD
   RETENTION TIME
     REFERENCE

   13C-2378-TCDF
   13C-2378-TCDD
   13C-12378-PeCDF
   13C-23478-PeCDF
   13C-12378-PeCDD
   13C-1234-TCDD
   13C-1234-TCDD
   13C-1234-TCDD
   13C-1234-TCDD
   13C-1234-TCDD
   13C-1234-TCDD
             RRT
   RRT
QC LIMITS (1)

0.993-1.009
0.993-1.009
0.918-1.076
0.999-1.001
0.987-1.016
                        0.931-0.994
                        0.993-1.036
                        1.002-1.013
                        1.091-1.371
                        1.123-1.408
                        1.134-1.428
(1)  Contract-required limits for Relative Retention Times  (RRT)  as  specified
    in Table 2, Method 1613.
                                                                         4/90

-------
                                    USEPA -  TD

                                      FORM 6B
                         PCDD/PCDF RELATIVE RETENTION TIMES
 Lab Name:
                                Episode No.:
 Contract No.:

 Instrument ID:

 Analysis Date:
                      SAS  No.:
                                  Init.  Cal.  Date:
                        GC Column ID:
                 Time:
                      .CS3  or VER Data Filename:
               Compounds Using 13C-123789-HXCDD as Internal Standard

                             RETENTION TIME
     NATIVE ANALYTES           REFERENCE            RRT
      123478-
      123678-
      123789-
      234678-
      123478-
      123678-
      123789-
      1234678
      1234678
      1234789
      OCDD
      OCDF
    HxCDF
    HxCDF
    HxCDF
    HxCDF
    HxCDD
    HxCDD
    HxCDD
    -HpCDF
    -HpCDD
    -HpCDF
    LABELED COMPOUNDS
     13C
     13C
     13C
     13C
     ISC-
     ISC
     13C-
     13C-
     13C-
     13C-
-123478-
-123678-
-234678-
-123789-
-123478-
-123678-
-1234678
-1234678
-1234789
-OCDD
HxCDF
HxCDF
HxCDF
HxCDF
HxCDD
HxCDD
-HpCDF
-HpCDD
-HpCDF
            13C-123478-HXCDF
            13C-123678-HXCDF
            13C-123789-HXCDF
            13C-124678-HXCDF
            13C-123478-HXCDD
            13C-123678-HXCDD
            13C-123678-HXCDD
            13C-1234678-HpCDF
            13C-1234678-HpCDD
            13C-1234789-HpCDF
            13C-OCDD
            13C-OCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
13C-123789-HXCDD
                                                         RRT
                                                       QC LIMITS  (1)
                                   0.986
                                   0.973
                                   0.937
                                     999
                                     999
                                     992
                                   0.986
                                   0.930
                                   0.986
                                   0.896
                                   0.996
     -1.015
0.
0.
0.
       ,025
       ,068
       ,001
       .001
       ,009
       ,016
       ,022
     -1.016
     -1.
       ,079"
     -1.005
0.995-1.013
0.947-
0.940-
0.971-
0.993-
0.974-
0.975-
0.953-
1.023-
1.024-
1.050-
      •0.992
      •1.006
      1.000
      1.017
      1.002
      1.006
      1.172
      1.125
      1.148
      1.275
(1)  Contract-required limits for Relative Retention Times (RRT) as specified
    in Table 2, Method 1613.
                                                                          4/90

-------
 Lab Name:
                  USEPA - ITD

                    FORM 7A ,
 PCDD/PCDF INITIAL PRECISION AND RECOVERY (IPR)

	    Episode No.:  	
 Contract No.:
           SAS No.:
 Matrix (agueous/solid/leachate):

 Ext.  Date:   	 Shift: 	
 Repl Data Filename:
 Rep2 Data Filename:
 Rep3 Data Filename:
 Rep4 Data Filename:
                 Analysis Date:
                 Analysis Date:
                 Analysis Date:
                 Analysis Date:
                   Time:
                   Time:
                   Time:
                   Time:
                   Cone. Units (pg/L or ng/Kg dry weight):
 NATIVE ANALYTES

 2378-TCDD

 12378-PeCDD

 123478-HxCDD
 123678-HxCDD
 123789-HxCDD

 1234678-HpCDD

 OCDD

 2378-TCDF

 12378-PeCDF
 23478-PeCDF

 123478-HXCDF
 123678-HxCDF
 123789-HxCDF
 234678-HxCDF

 1234678-HpCDF
 1234789-HpCDF

OCDF
                     SPIKE
                     CONG.
             REP 1
             CONC.
             FOUND
REP 2
CONC.
FOUND
REP 3
CONC.
FOUND
REP 4
CONC.
FOUND
 X
(1)
 s
(1)
(1)  X = average concentration; s = standard deviation of the concentration.
    Contract-required limits for X and s are specified in Table 7, Method 1613

                                                                         4/90
                                             '}

-------
                                    USEPA'- ITD

                                      FORM 7B
                   PCDD/PCDF INITIAL PRECISION , AND RECOVERY  (IPR)

 Lab Name: 	   Episode No.:  	

 Contract No.: 	   SAS No.: 	

 Matrix (agueous/solid/leachate): 	

 Ext. Date: 	 Shift: 	

 Repl Data Filename: 	  Analysis Date:  	 Time:
 Rep2 Data Filename: 	  Analysis Date:  	 Time:
 Rep3 Data Filename: 	  Analysis Date:  	 Time:
 Rep4 Data Filename: 	  Analysis Date:  	 Time:

                   Cone. Units (pg/L or ng/Kg dry  weight): 	
                               REP 1    REP 2     REP  3    REP 4
                     SPIKE     CONC.    CONC.     CONC.    CONG.      X
                     CONG.     FOUND    FOUND     FOUND    FOUND      (1)
 LABELED COMPOUNDS

 13C-2378-TCDD      	  	  	

 13C-12378-PeCDD
 13C-123478-HXCDD
 13C-123678-HXCDD

 13C-1234678-HpCDD

 13C-OCDD

 13C-2378-TCDF

 13C-12378-PeCDF
 13C-23478-PeCDF

 13C-123478-HXCDF
 13C-123678-HXCDF
 13C-123789-HXCDF
 13C-234678-HXCDF

 13C-1234678-HpCDF
 13C-1234789-HpCDF

CLEANUP STANDARD

37C14-2378-TCDD
(1)  X = average concentration.  Contract-required concentration limits (X)  fo
    IPR are specified in Table 7, Method 1613.

                                                                         4/90

                                         223

-------
                                    USEPA - ITD
 Lab Name:
 Contract No.:
                    ,                  FORM 8 A
                   PCDD/PCDF ONGOING PRECISION AND RECOVERY  (OPR)
           Episode No.:
SAS No.:
 Matrix (aqueous/solid/leachate):

 Ext.  Date:     	 Shift: 	
      	    PAR Data Filename:

       Analysis Date: 	 Time:
                   Cone. Units (pg/L or ng/Kg dry weight):
                                   SPIKE
                                   CONC.
                   CONC.
                   FOUND
             NATIVE ANALYTES

              2378-TCDD

              12378-PeCDD

              123478-HxCDD
              123678-HXCDD
              123789-HxCDD

              1234678-HpCDD

              OCDD

              2378-TCDF

              12378-PeCDF
              23478-PeCDF

              123478-HxCDF
              123678-HxCDF
              123789-HxCDF
              234678-HxCDF

              1234678-HpCDF
              1234789-HpCDF

             OCDF
OPR CONC.
LIMITS (1)
                               5.9

                              35.6

                              35.1
                              33.3
                              31.8

                              37.5

                              75.6

                               6.6

                              36.7
                              37.8

                              36.9
                              34.8
                              37.1
                              35.7

                              37.4
                              36.9
      14.2

      58.1

      60.4
      64.4
      61.2

      56.8

      118.7

      12.7

      57.3
      56.9

      58.8
      58.8
      55.7
      60.0

      60.5
      60.6
                              69.5 - 127.0
(1)  Contract-required concentration limits for OPR as specified in Table 7,
    Method 1613.
                                                                         4/90

-------
                                    USEPA - ITD
 Lab Name:
                 FORM 8B
 PCDD/PCDF ONGOING PRECISION AND RECOVER*

	   Episode No.: 	
 Contract No.:
       SAS No.:
 Matrix (aqueous/solid/leachate):

 Ext.  Date:	 Shift: 	
              	    PAR Data Filename:

               Analysis Date: 	 Time:
                   Cone.  Units  (pg/L or ng/Kg dry weight):
                                   SPIKE
                                   CONG.
                            CONC.
                            FOUND
             LABELED COMPOUNDS

              13C-2378-TCDD

              13C-12378-PeCDD

              13C-123478-HXCDD
              13C-123678-HXCDD

              13C-1234678-HpCDD

              13C-OCDD

              13C-2378-TCDF

              13C-12378-PeCDF
              13C-23478-P6CDF

              13C-123478-HXCDF
              13C-123678-HXCDF
              13C-123789-HXCDF
              13C-234678-HXCDF

              13C-1234678-HpCDF
              13C-1234789-HpCDF

            CLEANUP STANDARD

              37C14-2378-TCDD
OPR CONC.
LIMITS (1)
                                       25.0

                                       25.0

                                       25.0
                                       25.0

                                       25.0

                                       50.0

                                       25.0

                                       25.0
                                       25.0

                                       25.0
                                       25.0
                                       25.0
                                       25.0

                                       25.0
                                       25.0
      150.0

      150.0

      150.0
      150.0

      150.0

      300.0

      150.0

      150.0
      150.0-

      150.0
      150.0
      150.0
      150.0

      150.0
      150.0
                                       10.0 -  60.0
(1)  Contract-required concentration limits for OPR as specified in Table 7,
    Method 1613.  Labeled compound concentration limits are based on
    required percent recovery of 25-150% (Section 14.5, Method 1613).


                                                                        4/90
                                         9 9

-------
                        SECTION 3

QUALITY CONTROL SOLUTIONS PROVIDED TO ANALYTICAL LABORATORIES
            FOR THE EVALUATION OF EPA METHOD 1613
                                 226

-------
                                          Table 1
                      Calibration Solutions - concentrations in ng/mL


Native CDDs and CDFs         CS1      CS2        CS3         CS4       CSS

          2,3,7,8-TCDD         0.5         2         10          40       200
          23,7,8-TCDF         0.5         2         10   .       40       200
        1,2,3,7,8-PeCDD         2.5        10         50         200      1000
        1,23,7,8-PeCDF         2.5        10         50         200      1000
        23,4,7,8-PeCDF         2.5        10         50         200      1000
      1,23,4,7,8-HxCDD         2.5        10         50         200      1000
      1,23,6,7,8-HxCDD         2.5        10         50         200      1000
      1,23,7,8,9-HxCDD         2.5        10         50         200      1000
      1,2,3,4,7,8-HxCDF  '       2.5        10         50         200      1000
      1,2,3,6,7,8-HxCDF         2.5        10         50         200      1000
      1,23,7,8,9-HxCDF         2.5        10         50         200      1000
      23,4,6,7,8-HxCDF         2.5        10         50         200      1000
    1,23,4,6,7,8-HpCDD         2.5        10         50         200      1000
    1,2,3,4,6,7,8-HpCDF         2.5        10         50         200      1000
    U,3A7,8,9-HpCDF         2.5        10         50         200      1000
               OCDD         5.0        20        100         400      2000
                OCDF         5.0        20        100         400      2000


Labeled Compounds

    13C12-23,7,8-TCDD        100       100        100         100       100
    13C12-23,7,8-TCDF        100       100        100         100       100
 13C12-l,2v3,7,8-PeCDD        100       100        100         100       100
 13C12-l,23,7,8-PeCDF        100       100        100         100       100
 13C12-23,4,7,8-PeCDF        100       100        100         100       100
13C12-l,2,3,4,7,8-HxCDD        100       100        100         100       100
13C12-l,23,6,7,8-HxCDD        100       100        100         100       100
13C12-l,2r3,4,7,8-HxCDF        100       100        100         100       100
13Cu-l,23,6,7,8-HxCDF        100       100        100         100       100
13C12-l,23,7,8,9-HxCDF        100       100        100         100       100
13C12-2,3,4,6,7,8-HxCDF        100       100        100         100       100
13C12-l,23,4,6,7,8-HpCDD      100       100        100         100       100
13C12-l,2,3,4,6,7,8-HpCDF      100       100        100         100       100
13C12-l,23,4,7,8,9-HpCDF      100       100        100         100       100
         13C12-OCDD        200       200        200         200       200
    37C14-23,7,8-TCDD         0.5         2         10          40       200
13C12-1,23,4-TCDD            100       100        100         100       100
13C12-l,23,7,8,9HxCDD        100       100        100         100       100

-------
                          Table 2
             Labeled Compound Spiking Solution
                                             Concentration
Labeled Compound                              (ng/mH


13C12-2,3,7,8-TCDD                                 100
13C12-2,3,7,8-TCDF                                 100
"C12-l,2,3,7,8-PeCDD                               100
"C12-l,2,3,7,8-PeCDF                               100
13c12-2,3,4,7,8-PeCDF                               100
^C12-l,2,3,4,7,8-HxCDD                             100
13C12-l,2,3,6,7,8-HxCDD                             100
1?Cl2-l'2'3'4»7.8-HxCDF                             100
^C12-l,2,3,6,7,8-HxCDF                             100
"C12-l,2,3,7,8,9-HxCDF                             100
^C12-2,3,4,6,7,8-HxCDF                             100
"C12-1,2,3,4,6,7,8-HPCDD                           100
"C12-l,2,3,4,6,7,8-HpCDF                            100
"C^-lAW&o-HpCDF                            100
13C12-OCDD                                       200

-------
                                    Table3
                     Precision and Recovery Standard Solution
Compound
2,3,7,8-TCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDD
OCDF
Concentration Tng/ml)
         40
         40
        200
        200
        200
        200
        200
        200
        200
        200
        200
        200
        200
        200
        200
        400
        400
                                         9 9

-------
                       SECTION 4

                   PE SPIKE SOLUTIONS
        FOR THE EVALUATION OF EPA METHOD 1613
NOTE:  PE spike solutions will be included in the report after
       laboratories have completed study analyses and reported
       data.
                                 230

-------
                 SUMMARY REPORT
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
         INDUSTRIAL TECHNOLOGY DIVISION
          METHOD DETECTION LIMIT STUDY
        FOR METHOD 1613 DETERMINATION OF
             2,3,7,8-TCDD AND 2,3,7,8-TCDF
                      May 1990
                  USEPA Office of Water
           Office of Water Regulations and Standards
            Industrial Technology Division (WH-552)
                  Washington, DC 20460
                       231

-------
                                    TABLE OF CONTENTS


Method Detection Limit Study                                                           1


Table 1 - Analytical Results                                                             3

Table 2 - Standard Deviation and Method Detection Limits                                   4

Attachment 1-   Appendix B to Part 136                                                   5
               Definition and Procedure for the Determination of the
               Method Detection Limit - Revision 1.11

-------
                        METHOD DETECTION LIMIT (MDL) STUDY FOR
               METHOD 1613 DETERMINATION OF 2,3,7,8-TCDD AND 2,3,7,8-TCDF


      As part of the development of Method 1613  for the analysis of PCDDs and PCDFs, a method
detection limit (MDL) study was undertaken in December  1989.  The basic design of this study was in
accordance with the procedure for determining MDLs specified in Appendix B of 40 CFR Part 136, as
published in the October 26, 1984 Federal Register (Attachment  1).   The major requirements of this
procedure are:

      o   At least seven (7) aliquots of reagent water must be spiked with the analytes of interest

      o   Spike levels should be in the range of one to five times the laboratory's estimate of the detection
          limit of each analyte

      o   Analyze all replicates and calculate a mean and standard deviation of the concentration of each
          analyte

      o   Calculate the MDL as  the standard deviation times the Students t value for (n-1) degrees of
          freedom, where n is the number of replicates.

According to 40CFR 136, the MDL is defined as the minimum concentration of a substance that can be
measured and reported with 99% confidence that the analyte concentration is greater than zero and is
determined from analysis of a sample in a given matrix containing the analyte.

      For the purposes of this study, the laboratory chose to analyze eight  replicate samples instead of the
minimum of seven.

Estimated Detection Limits

      As noted above, the Federal Register procedure requires that each replicate is spiked with a solution
containing target analytes at a concentration between one and five times the laboratory's estimated detection
limit. The specifications in 40 CFR 136 also list four ways in which to determine an estimate of the detection
limit of the method. The first of these  options is to estimate the concentration value that corresponds to an
instrument signal-to-noise ratio of 2.5 to 5.0.  Using this criteria,  the laboratory calculated an estimated
detection limit of 25 ppq for 2,3,7,8-TCDD and 2,3,7,8-TCDF as follows.

      o   The lowest calibration solution (CC1) in Method 1613 has a 2,3,7,8-TCDD concentration of 0.5
          ng/mL and presents a peak with a signal to noise ratio of at least  10.

      o   Each sample analyzed by Method 1613 has a final volume of 20 uL.  By multiplying the final
          volume of the  sample times the concentration of TCDD  in the CC1, it was determined that the
          sample with a  concentration  equivalent to that of the lowest calibration standard would have a
          final TCDD concentration of 10 pg/L or 10 ppq.  This concentration is equal to  the "Minimum
          Level" described in Method 1613, and is derived in the same fashion.

      o   The laboratory equated the Minimum  Level in Method 1613 with the American Chemical
          Society's (ACS) concept of the Limit of Quantitation (LOQ). The ACS further defines a Limit
          of Detection (LOD) as  approximately one third of the LOQ.  Thus, the laboratory set their
          estimated detection limit as one third the Minimum Level, or 3.3 ppq. This figure represents the
          estimated detection limit at 100% recovery, and should yield an instrumental signal of at least
          2.5 times the background noise.

      o   Since Method  1613 specifies analyte quantitation by isotope dilution and allows data acceptance
          when labeled compound recovery is as low as 25%, the estimated detection limit was adjusted by
          the laboratory to account for the worst-case recovery. Thus 3.3 ppq was divided by 0.25, and a
          worst case estimated detection limit was calculated as 13.2 ppq.

                                               1
                                   O '<  f)

-------
Spike Levels

      The Federal Register specifies that each of the replicates be spiked with each analyte to yield a
concentration between one and five times the estimated detection limit. The laboratory chose to use spike
solutions containing approximately twice the estimated detection limit of 13.2 ppq.  Thus, the 2,3,7,8-TCDD
and 2,3,7,8-TCDF isomers were both spiked at 25 ppq.

Analytical Results and Calculation of Variance, Standard Deviation and Method Detection Limit

      Table 1 provides the analytical results for 2,3,7,8-TCDD and 2,3,7,8-TCDF in each of the eight
replicates. The Federal Register requires that the Method Detection Limit for each analyte be calculated as
the standard deviation of the replicate analyses multiplied by the  Students t value for (n-1)  degrees of
freedom, where n equals the number of replicates.  Table 2 provides the mean concentrations of both
isomers from the replicate analyses along with standard deviations and the calculated MDLs. The calculated
MDL for 2,3,7,8-TCDD is 5.6 ppq and the calculated MDL for 2,3,7,8-TCDF is 1.7 ppq.

      The MDL values of 5.6 ppq and 1.7 ppq  for TCDD and TCDF in Table 2 are below the "minimum
levels" of 10 ppq listed in Method 1613  for these  isomers.  As  indicated above, the minimum levels
correspond to the concentration in a sample equivalent to the concentration of each analyte in the lowest of
the calibration standards, assuming 100% recovery of the labeled compounds added to the sample and used
for quantification by isotope dilution.  Thus, the MDL data presented here indicate  that the method  is
capable of measuring 2,3,7,8-TCDD and 2,3,7,8-TCDFs at levels at least as low as the  minimum levels
specified in the method description.
                                        234
                                                2

-------
         Table 1
     Analytical Results
ISOMER
2378-TCDD
12378-PeCDD
123478-HxCDD
123678-HxCDD
123789-HxCDD
1234678-HpCDD
OCDD
2378-TCDF
12378PeCDF
23478-PeCDF
123478-HxCDF
123678-HxCDF
234678-HxCDF
123789-HxCDF
1234678-HpCDF
1234789-HpCDF
OCDF
SMX1R
33.1
24.2
33.8
59.6
30.2
802
8250
33.3
31.4
31.2
48.9
37.9
43.3
41.8
174
88.6
833
SMX2R
32.6
26.2
33.7
28.3
29.8
51.2
153
34.3
30.7
33.4
44.6
33.9
35.1
39.0
76.3
84.1
70.4
SMX3R
30.9
22.7
36.0
27.9
21.2
53.0
169
33.3
28.9
31.4
44.6
35.3
34.5
34.9
75.8
83.9
80.3
SMX4R
30.3
22.4
33.2
29.0
28.3
109
967
32.4
28.8
31.5
40.5
33.8
34.2
39.5
82.0
73.2
123
SMX5R
32.0
24.1
32.4
29.4
28.5
51.6
161
33.3
30.3
31.4
42.8
34.0
34.6
38.1
75.8
82.2
87.1
SMX6R
27.1
23.1
33.7
28.6
19.6
82.0
581
32.8
30.1
32.0
42.8
30.9
34.6
35.3
76.6
78.5
136
SMX7R
31.6
22.5
28.7
26.2
23.1
53.1
201
33.1
29.9
31.0
41.5
31.0
32.9
36.7
72.7
82.1
58.5
SMX8R
31.7
23.1
34.1
33.8
30.1
208
2110
33.6
30.6
29.3
43.3
34.1
36.1
37.6
101
84.7
249
235

-------
                Table 2
Standard Deviation and Method Detection Limits
Isomer
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,23,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
23,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDD
No. of
Observations
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
Mean
31.2
23.5
33.2
32.8
26.3
176.2
1574
33.3
30.1
31.4
43.6
33.9
37.9
35.7
91.8
82.2
204.7
Standard
Deviation
1.9
1.3
2.1
11.0
4.3
258.4
2780
0.6
0.9
1.1
2.5
2.2
2.3
3.2
34.4
4.6
261.0
Student's
r-Value
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
2.998
Method
Detection Limit
5.6
3.8
6.2
33.0
13.0
774.8
8333
1.7
2.7
3.4
7.6
6.7
6.8
9.6
103.1
13.8
782.5
            236

-------
                               ATTACHMENT 1
                           APPENDIX B TO PART 136
DEFINITION AND PROCEDURE FOR THE DETERMINATION OF THE METHOD DETECTION LIMIT
                                REVISION 1.11
                  (EXCERPTED FROM 40 CFR 136, OCTOBER 26,1984)

-------
   198    ~    Federal Register /'VoL 49. No. 209 / Friday.-October 26. 1984 7 Roles and Regulations
   Appendix B to Part 138—Definition utd
   Procedure for the Determination of the
   Method Detection Unit—Revision Ul
   Definition
     The method detection limit (MDL) U
   defined a* the minimum concentration of a
   substance that can be measured and reported
   with 99% confidence that the analyte
   concentration is greater than zero and U
   determined from analysis of a sample in a
   given matrix containing the analyte.
   Scope and Application
     This procedure is designed for applicability
   to a wide variety of sample types ranging
   from reagent (blank) water containing
   analyte to wastewater containing analyte.
   The MDL for an analytical procedure may
   vary as a function of sample type. The
   procedure requires a complete, specific, and
   well defined analytical method. It is essential
   that all sample processing steps of the
   analytical method be included in the
   determination of the method detection limit.
     The MDL obtained by this procedure is
   used to judge the  significance of a single
   measurement of a future sample.
    The MDL procedure was designed for
   applicability to a  broad variety of physical
   and chemical methods. To accomplish this.
   the procedure was made device- or
   instrument-independent.
  Procedure
    1. Make an estimate of the detection limit
  using one of the following:
    (a) The concentration value that
  corresponds to an instrument signal/noise in
  the range of 2.5 to 5.
    (b) The concentration equivalent of three
  times the standard deviation of replicate
  instrumental measurements  of the analyte in
  reagent water.
    (c) That region of the standard curve where
  there is a significant change  in sensitivity.
  i.e.. a break in the  slope of the standard
  curve.
    (d) Instrumental limitations.
    It is recognized that the experience of the
  analyst is important to this process.
  However, the analyst must include the above
  considerations in the initial estimate of the
  detection limit.
   2. Prepare reagent (blank) water that is as
  free of analyte as possible. Reagent or
 interference free water is defined as a water
 sample in which analyte and interferent
 concentrations are not detected at the
 method detection limit of each analyte of
 interest Interferences are defined as
 systematic errors in the measured analytical
 signal of an established procedure caused by
 the presence of interfering species
 (interferentL The interferent concentration is
 presupposed to be normally distributed in
 representative samples of a given matrix.
   3. (a) If the MDL is to be determined in
 reagent (blank) water, prepare a laboratory
 standard (analyte in reagent water} at a
 concentration which is at least equal to or in
 the same concentration range as the
 estimated method detection limit
 (Recommend between 1 and 5 times the
estimated method detection limit) Proceed to
Step 4.
     (b) If the MDL is to be determined in
   another sample matrix analyze the •ample. If
   the measured tare) of the analyte is in the
   recommended range of one to fiVe time* the
   estimated detection Hnrit proceed to Step 4.
     If the measured level of analyte isles* Utan
   the estimated detection limit add a known
   amount of analyte to bring the lerd of
   analyte between one and five times the
   estimated detection limit
     If the measured level of analyte is greater
   than five times the estimated detection limit
   there are two options.
     (1) Obtain another sample with a lower
   level of analyte in the same matrix if
   possible.
     (2) The sample may be used as is for
   determining the method detection limit if the
   analyte level does not exceed 10 times the
   MDL of the analyte in reagent water. The
   variance of the analytical method changes as
   the analyte concentration increases from the
   MDL hence the MDL determined under these
   circumstances may not truly reflect method
   variance at lower analyte concentrations.
    4. (a) Take a minimum of seven aliquots of
   the sample to be used to calculate the method
   detection limit and process each through the
   entire analytical method. Make all
  computations according to the defined
  method with final results in the method
  reporting units. If a blank measurement is
  required to calculate the measured level of
  analyte. obtain a separate blank
  measurement for each sample aliquot
  analyzed. The average blank measurement is
  subtracted from the respective sample
  measurements.
    (b) It may be economically and technically
  desirable to evaluate the estimated method
  detection limit before proceeding with 4a.
  This will: (1) Prevent repeating this entire
  procedure when the costs of analyses are
  high and (2) insure that the procedure is being
  conducted at the correct concentration. It is
  quite possible that an inflated MDL will be
  calculated from data obtained at many times
  the real MDL even though the level of analyte
  is less than five times the calculated method
  detection limit. To insure that the estimate of
  the method detection limit is a good estimate.
  it is necessary to determine that a lower
  concentration of analyte will not result in a
 significantly lower method detection limit.
 Take two aliquots of the sample to be used to
 calculate the method detection limit and
 process each through the entire method.
 including blank measurements as described
 above in 4a. Evaluate these data:
   (1) If these measurements indicate the
 sample is in desirable range for
 determination of the MDL. take five
 additional aliquots and proceed. Use all
 seven measurements for calculation of the
 MDL
   (2) If these measurements indicate the
 sample is not in correct range, reestimate the
 MDL obtain new sample as in 3 and repeat
 either 4a or 4b.
  5. Calculate the variance (S*) and standard
deviation (S) of the replicate measurements.
as follows:
  where:
  X,: i=l to n. = are the analytical results in
      the final method reporting units obtained
      from the n sample aliquots and I refers
      to the sum of the X values from i = l to n.
    6. (a) Compute the MDL as follows:
        MDL =
                         Ual.l-* • O.«t)  (S)
where:
  MDL
           the method detection limit
    U.-i.].. . .«») = the students' I value
      appropriate for a 99% confidence level
      and a standard deviation estimate with
      n-1 degrees of freedom. See Table.
    S = standard deviation of the replicate
      analyses.
    (b) The 95% confidence interval estimates
  for the MDL derived in 6a are computed
  according to the following equations derived
  from percentiles of the chi square over
  degrees of freedom distribution (,'/df)-
    LCL = 0.64 MDL
    UCL = 2.20 MDL
    where: LCL and UCL are the lower and
      upper 95% confidence limits respectively
      based on seven aliquots.
    7. Optional iterative procedure to verify the
  reasonableness of the estimate of the MDL
  and subsequent MDL determinations.
    (a) If this is the initial attempt to compute
  MDL based on the estimate of MDL
  formulated in Step 1. take the MDL as
  calculated in Step 6, spike in the matrix at the
  calculated MDL and proceed through the
  procedure starting with Step 4.
   (b] If this is the second or later iteration of
  the MDL calculation, use S' from the current
 MDL calculation and S1 from the previous
 MDL calculation to compute the F-ratio. The
 F-ran'o is calculated by substituting the larger
 S'into the numerator SA- and the others into
 the denominator SV The computed F-ratio is
 then compared with the F-ratio found in the
 table which is 3.05 as follows: if S'A/
 S*B<3.05. then compute the pooled standard
 deviation by the following equation:
             -I-T-1)
if S'A/S1.>3.05. respike at the most recent
calculated MDL and process the samples
through the procedure starting with Step 4. If
the most recent calculated MDL does not
permit qualitative identification when
samples are spiked at that level report the
MDL as a concentration between the current
and previous MDL which permits qualitative
identification.
  (c) Use the SMM as calculated in 7b to
compute the final MDL according to the
following equation:
                                                    238

-------
               Federal Register /  Vol 49.'No. 209 /  Friday. October 26, 1984 / Rules and Regulations
                                                                                 199
  MDL
  where 2381 i< equal to lot. .-. = .99).
    (d) The 95* confidence limits for MDL
  derived in 7c are computed according to the
  following equations derived from precentiles
  of the chi squared over degrees of freedom
  distribution.
    LO.=072MDL
    UCL-1.65 MDL
  where LCL and UCL are the lower  and upper
  95% confidence limits respectively  baaed on
  14 aliquots.

    TABLES Of STUDENTS' t VALUES AT THE 99
         PERCENT CONFIDENCE LEVEL
NUR*MT of rapiotts
7
B , . ..
o
1ft
11 ,, ,
Iff <
9*
?tt ......
}T
fft
Oft ,, , ,
-
"•ST
frwXxn
(0.1)
6
7
8
9
10
10
20
a
30
60
00

1... -.
3.143
2.998
2.896
2.821
1764
2.6Q2
2.528
2.485
2.457
2-390
2.326

 Reporting
   The analytical method used must be
 specifically identified by number or title and
 the MDL for each analyte expressed in the
 appropriate method reporting units. If the
 analytical method permits options which
 affect the method detection limit, these
 conditions must be specified with the MDL
 value. The sample matrix used to determine
.the MDL must also be identified with MDL
fcralue. Report the mean analyte level with the
IMDL and indicate if the MDL procedure was
 iterated. If a laboratory standard or a sample
 that contained a known amount analyte was
 used for this determination, also report the
 mean recovery.
   If the level of analyte in the sample was
 below the determined MDL or does not
 exceed 10 times the MDL of the analyte in
 reagent water, do not report a value for the
 MDL.

 Appendix C to Part 136—Inductively
 Coupled Plasma—Atomic Emission
 Spectrometric Method for Trace Element
 Analysis of Water and Wastes Method
 200.7
 1. Scope and Application
  1.1  This method may be used for the
 determination of dissolved, suspended, or
 total elements in drinking water, surface
 water, and domestic and industrial
 wastewaters.
  1.2 Dissolved elements are determined in
filtered and acidified samples. Appropriate
steps must be taken in all analyse* to ensure
that potential interferences are taken into
account This is especially true when
dissolved solids exceed 1500 mg/L, (See

  j jOI>Total elements are determined after
appropriate digestion procedure* are
oerformed. Since digestion technique*
       Se dissolved solid, content of the .
  sample*, appropriate step* most be taken to
  correct for potential interference effect*. (See
  sections.)
    1.4 Table 1 KsU elements for which this
  method applies along with recommended
  wavelengths and typical estimated
  instrumental detection limits using
  conventional pneumatic nebulization. Actual
  working detection limits are sample
  dependent and as the sample matrix varies.
  these concentrations may also vary. In time,
  other elements may be added as more
  information becomes available and as
  required-
    1.5 Because of the differences between
  various makes and models of satisfactory
  instruments, no detailed instrumental
  operating instructions can be provided.
  Instead, the analyst is referred to the
  instruction provided by the manufacturer of
  the particular instrument.

  2. Summary of Method
    2.1  The method describes a technique for
  the simultaneous or sequential multielement
  determination of trace elements in solution.
  The basis of the  method is die measurement
  of atomic emission by an optical
  spectroscopic technique. Samples are
  nebulized and the aerosol that is produced is
  transported to the plasma torch where
  excitation occurs. Characteristic atomic-line
  emission spectra are produced by a radio-
  frequency inductively coupled plasma (1CP).
  The spectra are dispersed by a grating
  spectrometer and the intensities of the lines
  are monitored by pbotomoltiplier tubes. The
  photocurrents from the photomultiplier tubes
  are processed and controlled by a computer
  system. A background correction technique is
  required to compensate for variable
  background contribution to the determination
  of trace elements. Background must be
  measured adjacent to analyte lines on
  samples during analysis. The position
  selected for the background intensity
  measurement, on either or both sides of the
  analytical line, will be determined by the
  complexity of the spectrum adjacent to the
  analyte line. The  position used must be free
  of spectral interference and reflect the same
  change in background intensity as occurs at
  the analyte wavelength measured.
  Background correction is not required in
 cases of line broadening where a background
 correction measurement would actually
 degrade the analytical result The possibility
 of additional interferences named in 5.1 (and
 tests for their presence as described in 5-2)
 should also be recognized and appropriate
 corrections made.

 3. Definitions
  3.1  Dissolved—Those elements which
 will pass through a 0.45 fim membrane filter.
  3.2  Suspended—Those elements which
 are retained by a 0.45 /im membrane filter.
  3.3  Total—The concentration determined
 on an unfiltered sample following vigorous
 digestion (Section  93). or the sum of the
 dissolved plus suspended concentrations.
 (Section 11 phis 9-2L
  3.4  Total recoverable—The coccentntion
determined on an unfiltered •ampb following
treatment with hot. dilute mineral acid
(Section 9.4).
    15  Instrumental detection limit—The
  concentration equivalent to a signal, due to
  the analyte. which is equal to three times the
  standard deviation of a series often replicate
  measurements of a reagent blank signal at
  the same wavelength.
    M  Sensitivity—The slope of the
  analytical curve. i.e- functional relationship
  between emission intensity and
  concentration.
    3J  Instrument check standard—A
  multielement standard of known
  concentration* prepared by the analyst to
  monitor and verify instrument performance
  on a daily basis. (See 7.6.1)
    3JJ  Interference check sample—A
  solution containing both interfering and
  analyte elements of known concentration
  that can be used to verify background and
  interelement correction factors. (See 7.6L2.)
    3.9  Quality control sample—A solution
  obtained from an outside source having
  known, concentration values to be used to
  verify the calibration standards. (See 7.6J)
    3.10  Calibration standards—A series of
  known standard solutions used by the
  analyst for calibration of the instrument (i.e_
  preparation of the analytical curve). (See 7.4)
    3.11  Linear dynamic range—The
  concentration range over which the
  analytical carve remains linear.
    3.12  Reagent blank—A volume of
  deionized. distilled water containing the
  same acid matrix as the calibration standards
  carried through the entire analytical scheme.
  (See 7 52)
    3.13   Calibration blank—A volume of
  deionized. distilled water acidified with
  HNCs and Ha (See 7.5.1)
    3.14  Method of standard addition—The
  standard addition technique involves the use
  of the unknown and the unknown plus a
  known amount of standard. (See 10.8.1.)

  * Safety
   4.1  The toxicity of carcinogeniciry 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 laboratory is repsonsible for
 maintaining a current awareness file of
 OSHA regulations regarding the safe
 handling of the chemicals specified in  this
 method. A reference file of material data
 handling sheets should also be made
 available to all personnel involved in the
 chemical analysis. Additional references to
 laboratory safety are available and have
 been identified <'**  "••-* "•» for the
 information of the analyst

 ^Interferences
  5.1 Several types of interference effects
 may contribute to inaccuracies in the
 determination of trace elements. They can be
 summarized a* follows:
  5.1.1 Spectral interferences can be
 categorized a* (1) overlap of a spectral line
from another element; (2) unresolved overlap
of molecular band spectra; (3) background
cuutriuutioA If out continuous or
recombination phenomena; and (4)
background contribution from stray light from
                                                             23H

-------
           !   DRAFT
                                                    Master 2-2
                                                    24 May 1987
                    METHOD 8290

      ANALYTICAL PROCEDURES AND  QUALITY ASSURANCE

               FOR MULTIMEDIA ANALYSIS

                         OF



          POLYCHLORINATED DIBKNZO-p~DIOXINS

                        AND

            POLYCHLORINATED DIBENZOFURANS
                         BY
HIGH-RESOLUTION GAS CHROMATOGRAPHY/HIGH-RESOLUTION MASS

                    SPECTROMETRY



                 (Exhibits D and E)



                         by

                    Yves Tondeur

                     June 1987



                  Project Officer

                 Werner F. Beckert
      Environmental Monitoring Systems Laboratory
             Las  Vegas, Nevada 89193-3478
                         240

-------
                                                                   Draft
                                                   ':               24 May 1987
                                    DP^  p. r*^T>    ;

                                    rtAFl     ;
                                     NOTICE



     This document is a preliminary draft.   It  has  not  been  formally released


by the University of Nevada Environmental Research  Center or the U.S. Environ-


mental Protection Agency, and it should not at  this stage be construed  to


represent University or Agency policy.   It  is circulated for comments on its


technical merit and policy implications.
                                    241

-------
                                                                  Draft
                                                                  24 May 1987
                               DRAFT
                                   FOREWORD


     In January 1986,  the Environmental Protection Agency published  an analy-

tical protocol, Protocol for the Analysis of 2,3,7,8-Tetrachlorodibenzo-p-

Dioxin (TCDD) by High-Resolution Gas Chroraatography/High-Resolution  Mass

Spectroraetry (HRGC/HRMS) (EPA 600/4-86-004), aimed at the determination of part-

per-trillion and sub-part-per-trillion levels of 2,3,7,8-TCDD and  of total TCDD

in soil, sediment and  aqueous samples.  The January 1986 document  was intended

to be a stepping stone for the realization of a more comprehensive method that

would include all the  polychlorinated dibenzodioxin (PCDD) and polychlorinated

dibenzofuran (PCDF)  congeners present in a broader spectrum of environmentally

significant matrices.

     The present report constitutes a draft addressing the analytical proce-

dures (Exhibit D) and  quality assurance (Exhibit E, quality assessment and

control) requirements  sections of the future analytical protocol for the

analysis of PCDDs and  PCDFs by HRGC/HRMS; i.e., Method 8290.   At times, refer-

ence to other exhibits (e.g., Exhibit C) are made, even though these sections

have not been prepared.  The format used for this report is similar  to the

format used for other  EPA TCDD protocols.  Figures and tables are, however,

grouped at the end of  Exhibit D.  A final version of Method 8290 is  expected

following peer review  of this draft report and the completion of the single-

laboratory evaluation.  Elements included in this Method 8290 have been taken

from a variety of sources, such as the EPA Region VII low-resolution mass

spectrometry (LRMS)  TCDD protocol, the aforementioned high-resolution mass

                                      ii
                                    942
                                    '- - ^C 4-M4

-------
spectrometry TCDD protocol',
                               DRAFF
                              Draft
                              24 May 1987
Method 8280 
-------
                                                                    Draft
                                                                    24 May 1987
                               TABLE OF CONTENTS
Foreword	ii
Abbreviations and Symbols	v

Analytical Methods (Exhibit D)
1.  Scope and Application	D-l
2.  Summary of the Method	D-2
3.  Definitions	D-5
4.  Interferences	D-10
5.  Safety	D-ll
6.  Apparatus and Equipment	D-16
7.  Reagents and Standard Solutions	D-23
8.  System Performance Criteria	D-27
9.  Calibration	D-32
10. Quality Assessment/Quality Control Procedures	D-40
11. Sample Preservation	D-41
12. Extraction and Cleanup Procedures	D-44
13. Analytical Procedures  	 D-59
14. Calculations	D-63
    APPENDIX A: PROCEDURE FOR THE COLLECTION,  HANDLING, ANALYSIS, AND
                REPORTING REQUIREMENTS OF WIPE TESTS PERFORMED WITHIN THE
                LABORATORY	D-71
    APPENDIX B: STANDARDS TRACEABILITY PROCEDURE 	 D-75
    APPENDIX C: SIGNAL-TO-NOISE DETERMINATION METHOD 	 D-81

Figures	D-84
Tables	D-93

Quality Assurance Requirements (Exhibit E)
1.  Summary of QA/QC Analyses	E-l
2.  Quality Assessment/Quality Control	E-2
3.  Laboratory Evaluation Procedures 	 E-12
                                       iv
                                     244

-------
                             DRAFT
                                                                   Draft
                                                                   24 May 1987
\i
                       LIST OF ABBREVIATIONS AND SYMBOLS

A         —   Integrated Ion abundance
ADC       —   Analogue-to-digltal  conversion
AX-21     —   Type of carbon adsorbent
C         —   Concentration
CDC       —   Center for Disease Control
CDWG      —   Chlorinated Dioxins  Workgroup
0 C       —   Degree centigrade
13C       —   Carbon-13 labeled
cm        —   Centimeter
DB-5      —   Type of fused-silica capillary column
DS        —   Data system
EDL       —   Estimated detection  limit
EMPC      —   Estimated maximum possible concentration
EMSL-LV   --   Environmental Monitoring System Laboratory, Las Vegas
EPA       —   Environmental Protection Agency
g         —   Gram
GC        —   Gas chromatography or gas chromatograph
GC/MS     —   Gas chromatography/mass spectrometry
HEPA      —   High-efficiency partlculate absorbant
HpCDD     —   Heptachlorodibenzodioxin
HpCDF     —   Heptachlorodlbenzofuran
HRGC/HRMS —   High-resolution gas  chromatography/high-resolution
               mass spectrometry
HxCDD     —   Hexachlorodibenzodioxin
HxCDF     —   Hexachlorodibenzofuran
IFB       —   Invitation for Bid
IS        —   Internal Standard
KD        —   Kuderna-Danish
L         —   Liter
MB        —   Method blank
MCL       —   Method calibration limit
mL        —   Milliliter
mm        —   Millimeter
M/AM      —   Mass spectrometer resolving power
MS        —   Matrix spike
MSD       —   matrix spike duplicate
OCDD      —   Octachlorodibenzodioxin
OCDF      —   Octachlorodlbenzofuran
OSHA      —   Occupational Safety  and Health Administration
PCB       —   Polychlorinated blphenyl
PCDD      —   Polychlorlnated dibenzodioxln
PCDPE     —   Polychlorinated diphenyl ether
PCDF      —   Polychlorlnated dlbenzofuran
PE        —   Performance evaluation
                                      O A C
                                      •c --i O

-------
                                                                    Draft
                                                                    24 May 1987
PEM
PeCDD
PeCDF
PFK
PS
ppm
ppt
Q
QA
QA/QC
rpm
RPD
RRF
RRF
RRT
RS
S
SAS
SES
SICP
SIM
SMO
S/N
SOP
SP-2330
Still-
 bottom
TCDD
TEF
V
v/v
W
WTE
uL
Performance evaluation material
Pentachlorodlbenzodloxln
Pentachlorodibenzofuran
Pe rfluoroke rosene
Plcogram
Part per million
Part per trillion
Amount of substance
Quality Assurance or Quality Assessment
Quality Assessment/Quality Control
Revolutions per minute
Relative percent difference
Relative response factor
Mean relative response factor
Relative retention time
Recovery standard
EPA reference standard solution
Special Analytical Service
Site evaluation sheet
Selected ion current profile
Selected ion monitoring
Sample Management Office
Signal-to-noise ratio
Standard Operating Procedure
Type of fused-sllica capillary column

Name of a matrix that is used as a noun
Tetrachlorodibenzodloxin
Toxicity Equivalency Factor
Volume
Volume/volume
Weight or laboratory working standard
Wipe test experiment
Microllter
                                       vi
                                      248

-------
DRAFT
                         Draft
                         24 May 1987
ANALYTICAL METHODS
  (EXHIBIT D)
    247

-------
                                                                    Draft
                                                                    24 May 1987
                                   EXHIBIT  D


1.    Scope and Application



1.1  This method provides procedures  for the detection and quantitative measure-

     ment of 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD),  polychlorinated

     dibenzo-p-dioxins  (tetra-  through  octachlorinated homologues;  PCDDs),  and

     polychlorinated dibenzofurans  (tetra-  through octachlorinated  homologues;

     PCDFs) in a variety of  environmental matrices and at part-per-trillion

     (ppt) concentrations.   The analytical  method calls for the  use of high-

     resolution gas  chromatography  and  high-resolution mass spectrometry (HRGC/

     HRMS) on purified  sample extracts.   Table  1 lists the various  sample types

     covered by this analytical protocol, the 2,3,7,8-TCDD-based method calibra-

     tion limits (MCLs) and  other germane information.  Analysis of a one-tenth

     aliquot of the  sample permits  measurement  of concentrations up to 10 times

     the  upper MCL (Table 1).   Samples  containing concentrations of specific

     congeneric analytes (PCDDs and PCDFs)  considered within the scope of this

     method that are greater than the upper MCL must  be analyzed by a protocol

     designed for such  concentration  levels.  An optional method for reporting

     the  analytical  results  using a 2,3,7,8-TCDD toxicity equivalency factor

     (TEF) is described.
                                      D-l
                                        248

-------
1.2  The sensitivity of this method is dependent upon the level of interferences




     within a given matrix.  Actual limits of detection and quantification will




     be provided based on the single- or multi-laboratory evaluation of this




     protocol, and on examining the data gathered by the Sample Management




     Office (SMO) from Special Analytical Services (SAS) performed over the




     past few years.








1.3  This method is designed for use by analysts who are experienced with




     residue analysis and skilled in high-resolution gas chromatography/high-




     resolution mass spectrometry (HRGC/HRMS).








1.4  Because of the extreme toxicity of many of these compounds, the analyst




     must take the necessary precautions to prevent exposure to materials known




     or believed to contain PCDDs or PCDFs.  It is the responsibility of the




     laboratory personnel to ensure that safe handling procedures are employed.








2.   Summary of the Method








2.1  This procedure uses matrix-specific extraction, analyte-specific cleanup,




     and high-resolution capillary column gas chroraatography/high-resolution




     mass spectrometry (HRGC/HRMS) techniques.








2.2  If interferences are encountered, the method provides selected cleanup




     procedures to aid the analyst in their elimination.  A simplified




     analysis flow chart is shown in Figure 1.
                                      D-2
                                        2 4 fl

-------
2.3  A specified amount (see Table 1)  of soil,  sediment,  fly ash,  water,


     sludge (Including paper pulp),  still-bottom,  fuel oil,  chemical reactor


     residue,  fish tissue,  or human adipose tissue Is spiked with  a solution


     containing specified amounts of each of the nine Isotopically (^C^)


     labeled PCDDs/PCDFs listed In Column 1 of  Table 2.   The sample Is then

                                            o
     extracted according to a matrix-specific extraction  procedure.  The  extrac-


     tion procedures are:  a) toluene  (or benzene) Soxhlet extraction for soil,


     sediment  and fly ash samples; b)  methylene chloride  liquid-liquid extrac-


     tion for water samples; c) toluene (or benzene) Dean-Stark extraction  for


     fuel oils and aqueous  sludges;  d) toluene  (or benzene)  extraction for


     still-bottoms; e) hexane/methylene chloride Soxhlet  extraction for fish


     tissue and paper pulp; and f) methylene chloride extraction for human


     adipose tissue.  The decision for the selection of an extraction procedure


     for chemical reactor residue samples is based on the appearance (consistency,


     viscosity) of the samples.  Generally, they can be handled according to


     the procedure used for still-bottom (or chemical sludge) samples.





2.4  The extracts are submitted to an.acid-base washing treatment  and dried.


     Following a solvent exchange step, the residue Is cleaned up  by column


     chromatography on neutral alumina and carbon on Celite 545®.   The extract


     from adipose tissue is treated with silica gel impregnated with sulfuric


     acid before chromatography on acidic silica gel, neutral alumina, and


     carbon on Celite 545®.  Fish tissue and paper pulp are subjected to  an


     acid wash treatment only prior to chromatography or  neutral alumina  and


   •  carbon/Celite.  The preparation of the final extract for HRGC/HRMS


     analysis  is accomplished by adding, to the concentrated carbon column



                                      D-3
                                      25

-------
     eluate, 10 to 50 uL uL (depending on the matrix type) of a tridecane


     solution containing 50 pg/uL of each of the two recovery standards


     13C12-1,2,3,4-TCDD and 13C12-1,2,3,7,8,9-HxCDD (Table 2).  The former is


     used to determine the percent recoveries of tetra- and pentachlorinated


     PCDD/PCDF congeners while the latter is used for the determination of

                                   L.
     hexa-, hepta- and octa-chlorinated PCDD/PCDF congeners percent recoveries.




2.5  One to two uL of the concentrated extract are injected into an HRGC/HRMS


     system capable of performing selected ion monitoring at resolving powers


     of at least 10,000 (10 percent valley definition).




2.6  The identification of OCDD and nine of the fifteen 2,3,7,8-substituted


     congeners (Table 3), for which a *3C-labeled standard is available in the


     sample fortification and recovery standard solutions (Table 2), is based


     on their elution at their exact retention time (-1 to -4-3 seconds from the


     respective internal or recovery standard signal) and the simultaneous


     detection of the two most abundant ions in the molecular ion region.  The


     remaining six 2,3,7,8-substituted congeners (i.e., 2,3,4,7,8-PeCDF;


     1,2,3,4,7,8-HxCDD; 1,2,3,6,7,8-HxCDF; 1,2,3,7,8,9-HxCDF; 2,3,4,6,7,8-HxCDF,


     and 1,2,3,4,7,8,9-HpCDF), for which no carbon-labeled internal standards


     are available in the sample fortification solution, and all other identified


     PCDD/PCDF congeners are identified by their relative retention times


     falling within their respective PCDD/PCDF retention time windows, as estab-


     lished by using a GC column performance evaluation solution, and the


     simultaneous detection of the two most abundant ions in the molecular


     ion region.  The identification of OCDF is based on its retention time



                                      D-4
                                    251

-------
                 1 ^
     relative to   Cj2~OCDD and the simultaneous detection of the two most




     abundant ions  in the molecular Ion region.   Confirmation is based on a




     comparison of  the ratio of the Integrated Ion abundance of the molecular




     Ion species to their theoretical abundance  ratio.








2.7  Quantification of the individual congeners, total PCDDs and total PCDFs is




     achieved in conjunction with the establishment of a multipoint (seven




     points) calibration curve for each homologue, during which each cali-




     bration solution is analyzed once.








3.   Definitions








3.1  Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans




     (PCDFs):  Compounds (Figure 2) that contain from one to eight chlorine




     atoms.  The fifteen 2,3,7,8-substltuted PCDDs (totaling 75) and PCDFs




     (totaling 135) are shown in Table 3.  The number of isomers at different




     chlorination levels is shown in Table 4.








3.2  Homologous series:  Defined as a group of chlorinated dibenzodioxins or




     dibenzofurans  having a specific number of chlorine atoms.








3.3  Isomer:  Defined by the arrangement of chlorine atoms within an




     homologous series.  For example, 2,3,7,8-TCDD is a TCDD isomer.








3.4  Congener:  Any isomer of any homologous series.
                                      D-5
                                      O !T

-------
3.5  Internal Standard:  An Internal standard is a   c^-labeled analogue of a




     congener chosen from the compounds listed in Table 3 and of OCDD.   Internal




     standards are added to all samples including method blanks and quality con-




     trol samples before extraction, and they are used to measure the concentra-




     tion of the analytes.  Nine internal standards are used in this method.




     There is one for each of the dioxin and furan homologues (except for OCDF)




     with the degree of chlorination ranging from four to eight.








3.6  Recovery Standard:  Recovery standards (two) are used to determine the




     percent recoveries for PCDDs and PCDFs.  The 13Cj2-l,2,3,4-TCDD is used to




     measure the percent recoveries of the tetra- and pentachlorinated dioxins




     and furans while ^^Cj2~l»2,3,7,8,9-HxCDD permits the recovery determination




     of the hexa-, hepta- and octachlorinated homologues.  They are added to




     the final sample extract before HRGC/HRMS analysis.  Furthermore,  ^^C-.^'




     1,2,3,7,8,9-HxCDD is used for the identification of the unlabeled analogue




     present in sample extracts (this exhibit, Section 2.6).









3.7  High-Resolution Concentration Calibration Solutions (Table 5):  Solutions




     (tridecane) containing known amounts of 17 selected PCDDs and PCDFs, nine




     internal standards (  C^'labeled PCDDs/PCDFs), and two carbon-labeled




     recovery standards (this exhibit, Section 3.6); the set of seven solutions




     is used to determine the instrument response of the unlabeled analytes




     relative to the internal standards and of the internal standards relative




     to the recovery standards.
                                      D-6
                                       253

-------
3.8  Sample Fortificatior
                            DRAFT
solution (isooctane) containing
     the nine  internal standards, which is used to  spike all samples before




     extraction  and cleanup.








3.9  Recovery  Standard Solution (Table 2):  A tridecane solution containing the




     two recovery  standards, which is added to the  final sample extract before




     HRGC/HRMS analysis.








3.10  Field Blank:  A portion of a sample representative of the matrix under




      consideration, which is free of any PCDDs/PCDFs.








3.11  Laboratory Method Blank:  A blank prepared in the laboratory and carried




      through  all  analytical procedure steps except the addition of a sample




      aliquot  to the extraction vessel.








3.12  Rinsate:   A  portion of solvent used to rinse  sampling equipment.  The




      rinsate  is analyzed to demonstrate that samples were not contaminated




      during sampling.








3.13  GC Column  Performance Check Mixture:  A tridecane solution containing a




      mixture  of selected PCDD/PCDF standards including the first and last




      eluters  for  each homologous series, which is  used to demonstrate continued




      acceptable performance of the capillary column (i.e., < 25 percent valley




      separation of 2,3,7,8-TCDD from all the other 21 TCDD isomers) and to




      define the homologous PCDD/PCDF retention time windows.
                                     D-7

-------
3.14  Performance Evaluation Materials:  Representative sample portions




      containing known amounts of certain unlabeled PCDD/PCDF congeners (in




      particular the ones having a 2,3,7,8-substitution pattern).   Representa-




      tive interferences may be present.  PEMs are obtained from the EPA EMSL-LV




      and submitted to potential contract laboratories, who must analyze these




      and obtain acceptable results before being awarded a contract for sample




      analyses (see 1FB Pre-Award Bid Confirmations).  PEMSs are also included




      as unspecified ("blind") quality control (QC) samples in any sample batch




      submitted to a laboratory for analysis.








3.15  Relative Response Factor:  Response of the mass spectrometer to a known




      amount of an analyte relative to a known amount of an internal standard.








3.16  Estimated Level of Method Blank Contamination:  The response from a signal




      occurring in the homologous PCDD/PCDF retention time windows, at any of




      the masses monitored, is used to calculate the level of contamination




      in the method blank, as described in Section 14 (this exhibit).  The




      results from such calculations must be reported along with the data




      obtained on the samples belonging to the batch associated with the method




      blank.








      Reporting a method blank contamination level for any of the 2,3,7,8-




      substituted congeners except OCDD and OCDF that exceeds 10 percent




      of the desired detection limit would invalidate the results and require




      automatic sample reruns (Exhibit C) for all positive samples found in




      that batch of samples.  A positive sample is defined as a sample found to






                                      D-8

-------
      contain at least one 2,3,7,8-substituted PCDD/PCDF congener (except OCDD




      and OCDF).  A valid method  blank run is  an analysis during which all




      internal standard signals  are characterized by S/N of at least 10:1.








3.17  Sample Rerun:  Extraction  of  another portion of the sample followed by




      extract cleanup and extract analysis.








3.18  Extract Reanalysis:  Analysis by HRGC/HRMS of another aliquot of the




      final extract.








3.19  Mass Resolution Check:   Standard method  used to demonstrate a static




      resolving power of 10,000  minimum (10 percent valley definition).








3.20  Method Calibration Limits  (MCLs):  For a given sample size, a final




      extract volume, and the lowest and highest concentration calibration




      solutions, the lower and upper MCLs delineate the region of quantification




      for which the HRGC/HRMS system was calibrated with standard solutions.








3.21  HRGC/HRMS Method Blank  (MB):   This additional QC check analysis corresponds




      to a 2-uL injection of  the  method blank  extract into the GC column and a




      complete (tetra- through octachlorinated congeners) HRGC/HRMS analysis.




      Such a QC check is required following a  calibration run and before the




      daily analysis of the first sample extract.  Acceptable HRGC/HRMS method




      blanks (see this exhibit,  Section 3.16,  for guidelines) must be obtained




      before sample extracts  can  be analyzed.
                                      D-9
                                      25G

-------
3.22  Matrix Spike (MS):  A sample which is spiked with a known amount of the




      matrix spike fortification solution (this exhibit, Section 3.24) prior




      to the extraction step.  The recoveries of the matrix spike compounds are




      determined; they are used to estimate the effect of the sample matrix




      upon the analytical methodology.








3.23  Matrix Spike Duplicate (MSB):  A second portion of the same sample as




      used in the matrix spike analysis and which is treated like the matrix




      spike sample.








3.24  Matrix Spike Fortification Solution:  Solution used to prepare the MS and




      MSD samples.  It contains all unlabeled analytes listed in Table 5 at con-




      centrations corresponding to the HRCC 3.  The solution also contains all




      internal standards used in the sample fortification solution at concen-




      trations as shown in Table 2.








4.    Interferences








4.1   Solvents, reagents, glassware and other sample processing hardware may




      yield discrete artifacts or elevated baselines that may cause misinter-




      pretation of the chromatographic data (see references 1 and 2 at the




      end of this Section).  All of these materials must be demonstrated to




      be free from Interferents under the conditions of analysis by running




      laboratory method blanks.  Analysts should avoid using PVC gloves.
                                      D-10

-------
4.2   The use of high-purity reagents and solvents helps minimize interference




      problems.  Purification of solvents by distillation in all-glass systems




      may be necessary.








4.3   Interferents co-extracted from the sample will vary considerably from




      matrix to matrix.  PCDDs and PCDFs are often associated with other




      interfering chlorinated substances such as polychlorinated biphenyls




      (PCBs), polychlorinated diphenyl ethers (PCDPEs),  polychlorinated




      naphthalenes, and polychlorinated xanthenes that may be found at con-




      centrations several orders of magnitude higher than the analytes of




      interest.  Retention times of target analytes must be verified using




      reference standards.  These values must correspond to the retention time




      windows established in Section 8.1.3 (this exhibit).  While certain clean-




      up techniques are provided as part of this method, unique samples may




      require additional cleanup steps to achieve lower detection limits.








4.4   A high-resolution capillary column (60 m DB-5) is  used to resolve as many




      PCDD and PCDF isomers as possible; however, no single column is known to




      resolve all isomers.  The use of several capillary columns will, in fact,




      be necessary during the determination of the toxicity equivalency factors




      (TEFs) (this exhibit, Section 14.7).








References:




     1.      "Control of Interferences in the Analysis of Human Adipose Tissue




            for 2,3,7,8-Tetrachlorodlbenzo-p-dioxin".  D. G. Patterson et al.,




            Environ. Toxicol. Chem. 5, 355-360 (1986).






                                      D-ll
                                        258

-------
     2.     "Protocol for the Analysis of 2,3,7,8-TCDD by HRGC/HRMS".




            J. S. Stanley and T. M. Sack, EPA 600/4-86-004.








5.     Safety








5.1   The following safety practices are exerpted directly from EPA Method 613,




      Section 4 (July 1982 version) and amended for use in conjunction with




      this method.








      Other PCDDs and PCDFs containing chlorine atoms in positions 2,3,7,8 are




      known to have toxicities comparable to that of 2,3,7,8-TCDD.  The




      analyst should note that finely divided dry soils contaminated with PCDDs




      and PCDFs are particularly hazardous because of the potential for inhala-




      tion and ingestion.  It is recommended that such samples be processed in




      a confined environment, such as a hood or a glove box.   Laboratory




      personnel handling these types of samples should also wear masks fitted




      with charcoal filter absorbent media to prevent inhalation of dust.








5.2   The toxicity or carcinogenicity of each reagent used in this method is




      not precisely defined; however, each chemical compound should be treated




      as a potential health hazard.  From this viewpoint, exposure to these




      chemicals must be kept to a minimum by whatever means available.  The




      laboratory is responsible for maintaining a current awareness file of




      OSHA regulations regarding the safe handling of the chemicals specified




      in this method.  A reference file of material safety data sheets should




      also be made available to all personnel involved in the chemical analysis.






                                      D-12

-------
      Additional references to laboratory safety are given in references 1-3




      (see end of Section 5,  this  exhibit).   Benzene and 2,3,7,8-TCDD have been




      identified as suspected human or mammalian carcinogens.








5.3   Each laboratory must develop a strict  safety program for the handling of




      2,3,7,8-TCDD.  The laboratory practices listed below are recommended.








5.3.1  Contamination of the laboratory will  be minimized by conducting most of




       the manipulations in a hood.








5.3.2  The effluents of sample splitters for the gas chromatograph and roughing




       pumps on the HRGC/HRMS system should  pass through either a column of ac-




       tivated charcoal or be bubbled through a trap containing oil or high-




       boiling alcohols.








5.3.3  Liquid waste should be dissolved in methanol or ethanol and irradiated




       with ultraviolet light at a wavelength less than 290 nm for several days




       (use F 40 BL lamps or equivalent). Using this analytical method, analyze




       the liquid wastes and dispose of the  solutions when 2,3,7,8-TCDD can no




       longer be detected.








5.4  Some of the following precautions were  issued by Dow Chemical U.S.A.




     (revised 11/78) for safe handling of 2,3,7,8-TCDD in the laboratory and




     amended for use in conjunction with this method.








5.4.1  The following statements on safe handling are as complete as possible on






                                      D-13

-------
       the basis of available toxicological information.  The precautions for




       safe handling and use are necessarily general in nature since detailed,




       specific recommendations can be made only for the particular exposure




       and circumstances of each individual use.  Assistance in evaluating the




       health hazards of particular plant conditions may be obtained from




       certain consulting laboratories and from State Departments of Health or




       of Labor, many of which have an industrial health service.  The 2,3,7,8-




       TCDD isomer is extremely toxic to certain kinds of laboratory animals.




       However, it has been handled for years without injury in analytical and




       biological laboratories.  Techniques used in handling radioactive and




       infectious materials are applicable to 2,3,7,8-TCDD.








5.4.1.1  Protective Equipment:  Throw-away plastic gloves, apron or lab coat,




         safety glasses and laboratory hood adequate for radioactive work.








5.4.1.2  Training:.  Workers must be trained in the proper method of removing




         contaminated gloves and clothing without contacting the exterior




         surfaces.








5.4.1.3  Personal Hygiene:  Thorough washing of hands and forearms after each




         manipulation and before breaks (coffee, lunch, and shift).








5.4.1.4  Confinement:  Isolated work area, posted with signs, segregated glass-




         ware and tools, plastic-backed absorbent paper on benchtops.








5.4.1.5  Waste:  Good technique includes minimizing contaminated waste.






                                      D-14
                                      OO
61

-------
         Plastic bag liners should be used in waste cans.








5.4.1.6  Disposal of Hazardous  Wastes:   Refer to the November 7,  1986 issue of




         the Federal Register on Land Ban Rulings for details concerning the




         handling of dioxin-containing wastes.








5.4.1.7  Decontamination:   Personnel - any mild soap with plenty  of scrubbing




         action.  Glassware,  tools and surfaces - Chlorothene NU  Solvent (Trade-




         mark of the Dow Chemical Company) is the least toxic solvent shown to




         be effective.   Satisfactory cleaning may be accomplished by rinsing




         with Chlorothene,  then washing with any detergent and water.  Dish




         water may be disposed  to the sewer after percolation through a char-




         coal bed filter.   It is prudent to minimize solvent wastes because




         they require special disposal through commercial sources that are




         expensive.








5.4.1.8  Laundry:  Clothing known to be contaminated should be disposed with




         the precautions described under "Disposal of Hazardous Wastes".




         Laboratory coats  or other clothing worn in 2,3,7,8-TCDD  work area may




         be laundered.   Clothing should be collected in plastic bags.  Persons




         who convey the bags and launder the clothing should be advised of the




         hazard and trained in  proper handling.  The clothing may be put into a




         washer without contact if the launderer knows the problem.  The washer




         should be run through  one full cycle before being used again for other




         clothing.
                                      D-15

-------
5.A.1.9  Wipe Tests:  A useful method of determining cleanliness of work




         surfaces and tools is to wipe the surface with a piece of filter




         paper, extract the filter paper and analyze the extract.








         NOTE:  Appendix A describes a procedure for the collection, handling,




         analysis, and reporting requirements of wipe tests performed within




         the laboratory.  The results and decision making processes are based




         on the presence of 2,3,7,8-substituted PCDD/PCDFs.








5.4.1.10  Inhalation:  Any procedure that may produce airborne contamination




          must be carried out with good ventilation.  Gross losses to a venti-




          lation system must not be allowed.  Handling of the dilute solutions




          normally used in analytical and animal work presents no significant




          inhalation hazards except in case of an accident.








5.4.1.11  Accidents:  Remove contaminated clothing immediately, taking precau-




          tions not to contaminate skin or other articles.  Wash exposed skin




          vigorously and repeatedly until medical attention is obtained.








References:




     1.   "Carcinogens - Working with Carcinogens", Department of Health,




          Education, and Welfare, Public Health Service, Center for Disease




          Control.  National Institute for Occupational Safety and Health.




          Publication No. 77-206, August 1977.








     2.   "OSHA Safety and Health Standards, General Industry", (29 CFR 1910),






                                      D-16
                                        9«
283

-------
           Occupational Safety and Health Administration,  OSHA 2206  (revised




           January 1976).








     3.    "Safety in Academic Chemistry Laboratories",  American Chemical  Society




          Publication,  Committee on Chemical  Safety (3rd  Edition,  1979.)








6.   Apparatus and Equipment








6.1  High-Resolution Gas Chromatograph/High-Resolution  Mass  Spectrometer/Data




     System (HRGC/HRMS/DS).








6.1.1  The GC must be equipped for temperature programming,  and all  required




       accessories must be available,  such as syringes, gases, and capillary




       columns.  The GC injection port must be designed for  capillary




       columns.  The use of  splitless  injection techniques is  recommended.




       On-column 1-ul injections can be used  on the 60-m  DB-5  column.  The  use




       of a moving needle injection port is also acceptable.   When using  the




       method described in this protocol, a 2-uL injection volume  is used




       consistently (i.e., the injection volumes for all  extracts, blanks,




       calibration solutions and the performance check  samples are 2 uL).




       One-uL injections are allowed;  however, laboratories  are encouraged  to




       remain consistent throughout the analyses by using the  same injection




       volume at all times.








6.1.2  Gas Chromatograph/Mass Spectrometer (GC/MS)  Interface—The  GC/MS interface




       components should withstand 350° C.  The interface must be  designed  so






                                      D-17
                                        284

-------
       that the separation of 2,3,7,8-TCDD from the other TCDD isomers achieved

       in the gas chromatographic column is not appreciably degraded.  Cold

       spots or active surfaces (adsorption sites) in the GC/MS interface can

       cause peak tailing and peak broadening.  It is recommended that the GC

       column be fitted directly into the mass spectrometer ion source without

       being exposed to the ionizing electron beam.  Graphite ferrules should

       be avoided in the injection port because they may adsorb the PCDDs and

       PCDFs.  Vespel1" or equivalent ferrules are recommended.




6.1.3  Mass Spectrometer—The static resolving power of the instrument must be

       maintained at a minimum of 10,000 (10 percent valley).  The mass spec-

       trometer must be operated in a selected ion monitoring (SIM) mode with

       a total cycle time (including the voltage reset time) of one second or

       less (this exhibit, Section 9.1.4.1).  At a minimum, the ions listed in

       Table 6 for each of the five SIM descriptors must be monitored.  Note

       that with the exception of the last descriptor (OCDD/OCDF), all the

       descriptors contain 10 ions.  The selection (Table 6) of the molecular

       ions M and M+2 for 13c-HxCDF and 13c-HpCDF rather than M+2 and M+A (for

       consistency) is to eliminate, even under high-resolution mass spectrometric

       conditions, interferences occuring in these two ion channels for samples

       containing high levels of native HxCDDs and HpCDDs.  It is important to

       maintain the same set of ions for both calibration and sample extract

       analyses.  The selection of the lock-mass ion is left to the performing

       laboratory.  The recommended mass spectrometer tuning conditions (this

       exhibit, Section 8.2.3) are based on the groups of monitored ions shown

       in Table 6.


                                      D-18
                                         I f-\ [—
                                         OO

-------
6.1.4  Data System—A dedicated data system Is employed to control the rapid




       multiple ion monitoring process and to acquire the data.   Quantification




       data (peak areas or peak heights) and SIM traces (displays of intensities




       of each ion signal being monitored including the lock-mass ion as a




       function of time) must be acquired during the analyses and stored.




       Quantifications may be reported based upon computer-generated peak areas




       or upon measured peak heights (chart recording).  The data system must




       be capable of acquiring data at a minimum of 10 ions in a single scan.




       It is also recommended to have a data system capable of switching to




       different sets of ions (descriptors) at specified times during an HRGC/




       HUMS acquisition.  The data system should be able to provide hard copies




       of individual ion chromatograms for selected gas chromatographic time




       intervals.  It should also be able to acquire mass-spectral peak profiles




       (this exhibit, Section 8.2.4) and provide hard copies of peak profiles




       to demonstrate the required resolving power.  The data system should




       also permit the measurement of noise on the base line.








       NOTE:  The detector ADC zero setting must allow peak-to-peak measurement




       of the noise on the base line of every monitored channel and allow for




       good estimation of the instrument resolving power.  In Figure 3, the




       effect of different zero settings on the measured resolving power is shown,








6.2  GC Column








     In order to have an isomer-specific determination for 2,3,7,8-TCDD and to




     allow the detection of OCDD/OCDF within a reasonable time interval in one






                                      D-19
                                      266

-------
     HRGC/HRMS analysis, the 60-m DB-5 fused-silica capillary column is recom-



     mended.  Minimum acceptance criteria must be demonstrated and documented



     (this exhibit, Section 8.1).  At the beginning of each 12-hour period



     (after mass resolution is demonstrated) during which sample extracts or



     concentration calibration solutions will be analyzed, column operating



     conditions must be attained for the required separation on the column to



     be used for samples.  Operating conditions known to produce acceptable



     results with the recommended column are shown in Table 7.






6.3  Miscellaneous Equipment and Materials







     The following list of items does not necessarily constitute an exhaustive



     compendium of the equipment needed for this analytical method.







6.3.1  Nitrogen evaporation apparatus with variable flow rate.






6.3.2  Balances capable of accurately weighing to 0.01 g and 0.0001 g.






6.3.3  Centrifuge.






6.3.4  Water bath, equipped with concentric ring covers and capable of being



       temperature-controlled within + 2° C.






6.3.5  Stainless steel or glass container large enough to hold contents of



       one-pint sample containers.
                                      D-20
                                         B*~1
                                          /

-------
6.3.6  Glove box.








6.3.7  Drying oven.








6.3.8  Stainless steel spoons and spatulas.








6.3.9  Laboratory hoods.








6.3.10  Pipets, disposable, Pasteur,  150 mm  long x 5  mm ID.








6.3.11  Pipets, disposable, serological, 10  mL,  for the preparation  of  the




        carbon column specified in Section 7.1.2.








6.3.12  Reacti-vial,  2 mL,  silanized  amber glass.








6.3.13  Stainless steel meatgrinder with a 3- to 5-mm hole  size  inner plate.








6.3.14  Separatory funnels, 125 mL.








6.3.15  Kuderna-Danish concentrator,  500 mL,  fitted with 10-mL concentrator




        tube and three-ball Snyder column.








6.3.16  Teflon1" boiling chips (or equivalent), washed with  hexane  before  use.









6.3.17  Chromatographic column,  glass,  300 mm x  10.5  mm,  fitted  with Teflon




        stopcock.






                                     D-21
                                        268

-------
6.3.18  Adaptors for concentrator tubes.








6.3.19  Glass fiber filters.








6.3.20  Dean-Stark trap, 5 or 10 mL,  with T-joints,  condenser and  125-mL flask.








6.3.21  Continuous liquid-liquid extractor.








6.3.22  All-glass Soxhlet apparatus,  500-mL  flask.








6.3.23  Glass funnels, sized to hold  170  mL  of  liquid.








6.3.24  Desiccator.








6.3.25  Solvent reservoir (125 mL), Kontes;  12.35 cm diameter (special order




        item), compatible with gravity carbon column.








6.3.26  Rotary evaporator with a temperature-controlled water bath.








6.3.27  High-speed tissue homogenizer, equipped with an EN-8 probe or




        equivalent.








6.3.28  Glass wool, extracted with methylene chloride, dried and stored in a




        clean glass jar.








        NOTE:  Reuse of glassware should  be  minimized to  avoid  the risk of






                                      D-22
                                     26!)

-------
                                                          _DRAFT     •
        contamination.  All glas-sware""ttra-t-i« ..reused must bV-aerupuiot:
                                         - • ..  _   -"-*•-.»_,/ (
        cleaned  as  soon as possible after use,  applying the following procedure:
        Rinse  glassware with the last solvent used  in  it, then with high-purity
        acetone  and hexane.  Wash with hot detergent water.  Rinse with copious
        amounts  of  tap water and several portions of distilled water.  Drain,  dry
        and heat in a muffle furnace at 400°  C  for  15  to 30 minutes.  Volumetric
        glassware must not be heated in a muffle  furnace.  Some thermally stable
        materials (such as PCBs) may not be removed by heating in a muffle
        furnace.  In these cases, rinsing with  high-purity acetone and
        hexane may  be substituted for muffle-furnace heating.  After the
        glassware is dry and cool, rinse it with  hexane and store it inverted
        or capped with solvent-rinsed aluminum  foil in a clean environment.


7.  Reagents and Standard Solutions


7.1  Column Chromatography Reagents


7.1.1  Alumina,  neutral, Super 1, Woelm®, 80/200  mesh.  Store in a sealed
       container at room temperature in a desiccator over self-indicating
       silica  gel.


7.1.2  Carbopak  C (80 to 100 mesh, Supelco 1-1025)  and Celite 545® (Supelco).
       Preparation  of the Carbopak C/Celite 545®  column:  Thoroughly mix
       3.6 g Carbopak C (80 to 100 mesh) and  16.4 g Celite 545® in a 40-mL
       vial.  Activate the mixture at 130° C  for  6  hours, then store it in a
       desiccator.  Cut off both ends of a 10-mL  disposable serological pipet

                                     D-23
                                          270

-------
       to give a 4-inch long column.  Fine-polish both ends and flare,  if




       desired.  Insert a glass-wool plug at one end,  then pack the  column with




       0.64 g of the activated Carbopak C/Celite 545® mixture to form a 2-cm




       long absorbant bed.  Cap the packing with another glass-wool  plug.








7.2  Reagents








7.2.1  Sulfuric acid, concentrated, ACS grade, specific gravity 1.84.








7.2.2  Potassium hydroxide, ACS grade, 20 percent (w/v) in distilled water.








7.2.3  Sodium chloride, analytical reagent, 5 percent (w/v) in distilled




       water.








7.2.4  Potassium carbonate, anyhdrous, analytical reagent.








7.3  Desiccating Agent








7.3.1  Sodium sulfate, granular, anhydrous; use as such.








7.4  Solvents








7.4.1  High-purity, distilled-in-glass or highest available purity:   methylene




       chloride, hexane, benzene, methanol, tridecane, isooctane, toluene,




       cyclohexane, and acetone.
                                      D-24

-------
7.5  Calibration Solutions








/.5.1  High-Resolution Concentration Calibration Solutions (Table 5) — Seven




       triaecane solutions containing unlabeied (totaling 17) and carbon-labeied




       (.totaling Li} fUJUs and t-cuts at known concentrations used to canorate




       tne instrument.  The concentration ranges are nomoiogue dependent, witn




       the lowest values associated with the tetra- and pentachlorinated




       dioxins and furans (2.5 pg/uL; and the highest for the octachlorinated




       congeners (1000 pg/uL;.








/.5.2  These high-resolution concentration calibration solutions may be obtained




       from the Quality Assurance Division, US EPA, Las Vegas, Nevada.   However,




       additional secondary standards must be obtained from commercial sources,




       and solutions must be prepared in the contractor laboratory.  Trace-




       ability (Appendix B) of standards must be verified against EPA-supplied




       standard solutions.  Such procedures will be documented by laboratory




       standard operating procedures (SOP) as required in IFB Preaward Bid




       Confirmations, part 2.f.(4).  It Is the responsibility of the laboratory




       to ascertain that the calibration solutions received (or prepared) are




       indeed at the appropriate concentrations before they are used to analyze




       samples.  A recommended traceablllty procedure for PCDD/PCDF standards




       is described in Appendix B.








7.5.3  Store the concentration calibration solutions in 1-mL minivlals at




       room temperature in the dark.
                                      D-25

-------
7.6  GC Column Performance Check Solution








     This solution contains the firstand last-eluting isomers for each homolo-




     gous series from tetra- through hepta-chlorinated congeners.  The solution




     also contains a series of other TCDD isomers for the purpose of documenting




     the chromatographic resolution.  The *^Cj2~2,3,7,8-TCDD is also present.




     The laboratory is required to use tridecane as the solvent and adjust the




     volume so that the final concentration does not exceed 100 pg/uL per




     congener.  Table 8 summarized the qualitative composition (minimum




     requirement) of this performance evaluation solution.








     NOTE:  The use of a PCDD/PCDF-containing fly-ash extract is allowed but




     the qualitative equivalency of the fly-ash extract to the EPA solution




     should be demonstrated for each fly-ash extract.








7.7  Sample Fortification Solution








     This isooctane solution contains the nine internal standards at the nominal




     concentrations that are listed in Table 2.  The solution contains at least




     one carbon-labeled standard for each homologous series, and it is used to




     measure the concentrations of the native substances.  (Note that ^c-  -QCDF




     is not present in the solution.)








7.8  Recovery Standard Solution








     This tridecane solution contains two recovery standards (*-*C  -1,2,3,4-




     TCDD and 13C12-1,2,3,7,8,9-HxCDD) at a nominal concentration of 50 pg/uL




                                      D-26
                                     273

-------
     per compound.  Ten to titty UL ot this solution will be spiked into each


     sample extract before the tinal concentration step and tiKU(J/HKMS analysis.





b.  System Performance Criteria





    System performance criteria are presented beiow.  me laboratory may use the


    recommended GC column described in Section 6.2 (.this exhibit;.  It must be


    documented that ail applicable system performance criteria specified in


    Section 8.1 (this exhibit; were met before analysis ot any sample Is per-


    formed,  lable / provides recommended lie conditions tnat can be used to


    sacisiy cne required criteria,  figure 4 provides a typical i^-nour analysis


    sequence wnereoy tne response factors and mass spectrometer resolving


    power checks must be performed at the beginning and the end ot each iz-hour


    period ot operation.  A. tJC column performance check is only required at the


    beginning of each 12-hour period during which samples are analyzed.  An


    HRGC/HRMS method blank run (this exhibit, Section 3.21) is required between


    a calibration run and the first sample run.   The same method blank extract


    may thus be analyzed more than once if the number of samples within a batch


    requires more than 12 hours of analyses.





8.1  GC Column Performance





8.1.1  Inject 2 uL (this exhibit, Section 6.1.1) of the column performance


       check solution (this exhibit, Section 7.6) and acquire selected ion


       monitoring (SIM) data as described in Section 6.1.3 (this exhibit) within


       a total cycle time of £ 1 second (this exhibit, Section 9.1.4.1).



                                      D-27
                                            f~l  I
                                            i 4

-------
8.1.2  The chromatographic separation between 2,3,7,8-TCDD  and  the  peaks  repre-




       senting any other TCDD isomers must be resolved  with a valley of < 25




       percent (Figure 5), where








                             Valley Percent  =  (x/y)  (100)








       x = measured as In Figure 5 from the 2,3,7,8-closest TCDD  elutlng




           Isomer, and








       y - the peak height of 2,3,7,8-TCDD.








       It Is the responsibility of the laboratory to verify the conditions




       suitable for the appropriate resolution of 2,3,7,8-TCDD  from all other




       TCDD isomers.  The GC column performance check  solution  also contains the




       known first and last PCDD/PCDF eluters under the conditions  specified In




       this protocol.   Their retention times are used  to determine  the eight




       homologue retention time windows that are used  for qualitative  (this




       exhibit, Section 13.4.1) and quantitative purposes.   All peaks  (that




       Includes 13C12-2,3,7,8-TCDD) must be labeled and identified  on  the




       chromatograms.   Furthermore, all first eluters  of a  homologous  series




       must be labeled with the letter F, and all last  eluters  of a homologous




       series must be labeled with the letter L (Figure 5 shows an  example of




       peak labeling for TCDD isomers).  Any individual selected  ion current




       profile (SICP)  (for the tetras, this would be the SICP for m/z  322 and




       m/z 304) or the reconstructed homologue Ion current  (for the tetras,




       this would correspond to m/z 320 + m/z 322 + m/z 304 + m/z 306)






                                      D-28
                                      275

-------
       constitutes an acceptable form of data presentation.   An SICP for




       the labeled compounds (e.g. ,  m/z  334 for labeled  TCDD) is also required.








8.1.3  The retention times  for the switching of SIM ions characteristic  of  one




       homologous series to the next higher homologous  series must be indicated




       in the SICP.  Accurate switching  at  the appropriate  times is absolutely




       necessary for accurate monitoring of these compounds.   Allowable  toler-




       ance on the daily verification with  the GC performance check solution




       should be better than 10 seconds  for the absolute retention times of all




       the components of the mixture.  Particular caution should be excercised




       for the switching time between the last tetrachlorinated congener (i.e.,




       1,2,8,9-TCDD) and the first pentachlorinated congener  (i.e., 1,3,4,6,8-




       PeCDF), as these two compounds elute within 15 seconds of each other on




       the 60-m DB-5 column.  A laboratory  with a GC/MS  system that is not




       capable of detecting both congeners  (1,2,8,9-TCDD and  1,3,4,6,8-PeCDF)




       within one analysis  must indicate in the case narrative of its report




       which congener (only one is permitted) was missed.








8.2  Mass Spectrometer Performance








8.2.1  The mass spectrometer must be operated in the electron ionlzatlon mode.




       A static resolving power of at least 10,000 (10 percent valley defini-




       tion) must be demonstrated at appropriate masses  before any analysis is




       performed (this exhibit,  Section  13).   Static resolving power checks




       must be performed at the beginning and at the end of each 12-hour period




       of operation.  However,  it is recommended that a  visual check (i.e.,






                                      D-29
                                        276

-------
       documentation is not required) of the static resolution be made by using

       the peak matching unit before and after each analysis.  Corrective

       actions must be implemented whenever the resolving power does not meet

       the requirement.



8.2.2  Chromatography time for PCDDs and PCDFs exceeds the long-term mass

       stability of the mass spectrometer.  Because the instrument is operated

       in the high-resolution mode, mass drifts of a few ppm (e.g., 5 ppm in

       mass) can have serious adverse effects on the instrument performances.

       Therefore, a mass-drift correction is mandatory.  To that effect, it is

       recommended to select a lock-mass ion from the reference compound (PFK

       is recommended) used for tuning the mass spectrometer.  The selection of

       the lock-mass ion is dependent on the masses of the ions monitored

       within each descriptor.  Table 6 offers some suggestions for the lock-

       mass ions.  However, an acceptable lock-mass ion at any mass between the

       lightest and heaviest ion in each descriptor can be used to monitor and

       correct mass drifts.  The level of the reference compound (PFK) metered

       into the ion chamber during HRGC/HRMS analyses should be adjusted so

       that the amplitude of the most intense selected lock-mass ion signal

       (regardless of the descriptor number) does not exceed 10 percent of the

       full-scale deflection for a given set of detector parameters.  Under

       those conditions, sensitivity changes that might occur during the

       analysis can be more effectively monitored.



       NOTE:  Excessive PFK (or any other reference substance) may cause noise

       problems and contamination of the ion source resulting in an increase in


                                      D-30
                                           2r-i -•)
                                           I (

-------
       downtime for source cleaning.








8.2.3  By using a PFK molecular leak,  tune  the  Instrument  to  meet  the  minimum




       required resolving power of  10,000  (10 percent  valley) at m/z 304.9824




       (PFK) or any other reference signal  close  to  m/z  303.9016  (from TCDF).




       By using the peak matching unit and  the  aforementioned PFK  reference




       peak, verify that the  exact  mass of  m/z  380.9760  (PFK) is within 5  ppm




       of the required value.   Note that the selection of  the low- and high-mass




       ions must be such that they  provide  the  largest voltage jump performed




       in any of the five mass descriptors  (Table 6).








8.2.4  Documentation of the instrument resolving  power must  then be accomplished




       by recording the peak  profile of the high-mass  reference signal (m/z




       380.9760) obtained during the above  peak matching experiment by using




       the low-mass PFK ion at m/z  304.9824 as  a  reference.   The minimum




       resolving power of 10,000 must be demonstrated  on the  high-mass ion




       while it is transmitted at a lower  accelerating voltage than the low-mass




       reference ion, which Is transmitted  at full sensitivity.  The format  of




       the peak profile representation (Figure  6) must allow  manual determina-




       tion of the resolution, i.e.,  the horizontal  axis must be a calibrated




       mass scale (amu or ppm per division).  The result of  the peak width




       measurement (performed at 5  percent  of the maximum, which corresponds to




       the 10-percent valley  definition) must appear on  the hard copy  and




       cannot exceed 100 ppm  at m/z 380.9760 (or  0.038 amu at that particular




       mass).
                                      D-31
                                       278

-------
9.  Calibration
                                    DRAFT
9.1  Initial Calibration








     Initial calibration is  required before any samples are  analyzed for PCDDs




     and PCDFs.   Initial calibration is also required if any routine calibration




     (this exhibit,  Section  9.3) does not meet the required  criteria listed in




     Section 9.4 (this  exhibit).








9.1.1  All seven high-resolution concentration calibration solutions listed in




       Table 5 must  be  used  for the initial calibration.








9.1.2  Tune the  Instrument with PFK as described in Section  8.2.3  (this exhibit).








9.1.3  Inject 2  uL of the GC column performance check solution (this exhibit,




       Section 7.6)  and acquire SIM mass spectral data as described earlier In




       Section 8.1 (this exhibit).  The total cycle time must be £ 1 second.




       The laboratory must not perform any further analysis  until  it is demon-




       strated and documented that the criterion listed in Section 8.1.2 (this




       exhibit)  was  met.








9.1.4  By using  the  same GC  (this exhibit, Section 6.2) and  mass spectrometer




       (this exhibit, Section 6.1.3) conditions that produced acceptable results




       with the  column  performance check solution, analyze a 2-uL  portion of




       each of the seven concentration calibration solutions once  with the




       following mass spectrometer operating parameters.






                                     D-32
                                       97Q
                                       ^ i <..»

-------
9.I.A.I  The total cycle time for data  acquisition  must  be  <  1  second.  The




         total cycle time includes the  sum  of  all the  dwell times  and voltage




         reset times.








9.1.4.2  Acquire SIM data for all the  ions  listed in  the five descriptors  of




         Table 6.








9.1.4.3  The ratio of  integrated ion current for the  ions appearing  in  Table 9




         (homologous series  quantification  ions) must  be within the  indicated




         control limits (set for each homologous series).








9.1.4.4  The ratio of  integrated ion current for the  ions belonging  to  the




         carbon-labeled internal and recovery  standards  must  be within  the




         control limits stipulated in Table 9.








         NOTE:  Sections 9.1.4.3 and 9.1.4.4 (this  exhibit) require  that  17  ion




         ratios from Section 9.1.4.3 and  11 ion ratios from Section  9.1.4.4  be




         within the specified control  limits simultaneously in one run.   It  is




         the laboratory's responsibility  to take corrective action if the  ion




         abundance ratios are outside  the limits.








9.1.4.5  For each  SICP and for each GC  signal  corresponding to the elution of a




         target analyte and  of its labeled  standards,  the signal-to-noise  ratio




         (S/N) must be better than or equal to 2.5.   Appendix C describes  the




         procedure to  be followed for  the measurement  of the  S/N from con-




         spicuously weak signals.   This measurement is required for  any GC






                                     D-33
                                          280

-------
         peak that has an apparent S/N of less than 5:1.  The result of the




         calculation must appear on the SICP above the GC peak in question.








9.1.4.6  Referring to Table 10, calculate the 17 relative response factors




         (RRF) for unlabeled target analytes [RRF(n); n « 1 to 17] relative to




         their appropriate internal standards (Table 5) and the nine RRFs for




         the labeled 13C12 internal standards [RRF(m); m = 18 to 26)] relative




         to the two recovery standards according to the following formulae:
                                  RRF(n)  =   —
                                                   Ais
                                  RRF(m)  .
                                              Qis '  Ars
         where
          Ax = sum of the integrated ion abundances of the quantification




               ions (Tables 6 and 9) for unlabeled PCDDs/PCDFs,








         Ais = sum °f tne integrated ion abundances of the quantification




               ions (Tables 6 and 9) for the labeled internal standards,








         Ars = sum of the integrated ion abundances of the quantification




               ions (Tables 6 and 9) for the labeled recovery standards,
                                      D-34

-------
         Qj_s = quantity of the internal standard injected (pg),



         Qrs = quantity of the recovery standard injected (pg),  and



          Qx = quantity of the unlabeled PCDD/PCDF analyte injected (pg).



         The RRF(n) and RRF(m) are dimensionless quantities;  the

         units used to express QIS,  Qrs anc* QX must be the same.



9.1.4.7  Calculate the RRF(n)s and their respective percent relative standard

         deviations (%RSD) for the seven calibration solutions:
                         	           7
                         RRF(n) = 1/7  I  RRF^n)  ,
          where n represents a particular PCDD/PCDF (2,3,7,8-substituted) con-

          gener (n = 1 to 17;  Table 10),  and j  is the injection number (or

          calibration solution number;  j  = 1 to 7).



9.1.4.8  The relative response factors  to be used for the determination of the

         concentration of total isomers in a homologous series (Table 10) are

         calculated as follows:



9.1.4.8.1  For congeners  that  belong to a homologous series containing only

           one isomer (e.g., OCDD and OCDF) or  only one 2,3,7,8-substituted

           isomer (Table  4;  TCDD, PeCDD,  HpCDD, and TCDF),  the mean RRF used


                                     D-35
                                         9H9
                                         *~ O <^.

-------
           will be the same as the mean RRF determined in Section 9.1.4.7  (this


           exhibit).
                                                           13
                                                             C12~°CDF as  an
NOTE:  The calibration solutions do not contain


internal standard.  This is because a minimum resolving  power of


12,000 is required to resolve the [M+6]+ ion of 13C12-OCDF  from the

[M+2]+ ion of OCDD (and [M+4]+ from 13C12-OCDF with  [M]+ of OCDD).

Therefore, the RRF for OCDF is calculated relative to
9.1.4.8.2  For congeners that belong to a homologous series  containing more
           than one 2,3,7,8-substituted isomer (Table 4),  the  mean  RRF used


           for those homologous series will be the mean of the RRFs calculated


           for all individual 2,3,7,8-substituted congeners using the equation

           below:
                                 RRF(k)
                                1   t
                                    I
                                t  n=l
RRFT
           where
                    k  =  27 to 30 (Table 10),  with 27 = PeCDF;  28  =  HxCDF;


                          29  =  HxCDD; and 30  = HpCDF,


                    t  =  total number of 2,3,7,8-substituted  isomers present  in


                          the calibration solutions (Table  5)  for each homologous


                          series (e.g., two for PeCDF, four for  HxCDF, three for


                          HxCDD, two for HpCDF).
                                      D-36
                                       283

-------
         NOTE:   Presumably,  the HRGC/HRMS response factors of different isomers

         within a homologous  series are different.  However, this analytical

         protocol will  make  the assumption that  the HRGC/HRMS responses of all

         isomers in a homologous  series that do  not have the 2,3,7,8-substitution

         pattern are the  same as  the  responses of one or more of the 2,3,7,8-

         substituted isomer(s) in that homologous series.



9.1.4.9  Relative response factors [RRF(m)] to be used for the determination

         of the percent recoveries for the nine  internal standards are calcu-

         lated  as follows:
                                RRF(m)
                                           Aism  *
                                                 *  Ars
                              	        1    7
                              RRF(m)   =   -    £   RRFj(m),
                                          7  J-l
         where:
            m = 18 to  26  (congener  type) and  j =  1 to  7  (injection number),



         Aism = sum of the  integrated  ion abundances of  the quantification ions

                (Tables 6 and  9)  for a given  internal  standard  (m =  18 to 26),



          Ars = sum of the  integrated  ion abundances of  the quantification ions

                (Tables 6 and  9)  for the appropriate recovery standard (see Table 5,

                footnotes),
                                      D-37
                                      284

-------
                                      J)RAFT    j
         Q,.., and Qtf.m = quantities of, respectfvely",~Tnii5~TECovery standard (rs)
          rs      i. s
                        and  a particular internal standard  (is = m) injected
                        (pg),

               RRF(m) = relative response factor of a  particular internal
                        standard (m) relative to an appropriate recovery
                        standard, as determined from one  injection, and


               RRF(m) = calculated mean relative response factor of a particular
                        internal standard (m) relative to an appropriate recovery
                        standard, as determined from the  seven initial calibra-
                        tion injections (j).


9.2  Criteria for Acceptable Calibration


     The criteria listed below for acceptable calibration must be met before
     the analysis is  performed.


9.2.1  The percent relative  standard deviations for the mean response factors
       [RRF(n) and RRF(m)] from each of the 26 determinations (17 for the
       unlabeled standards and 9 for the labeled reference  compounds) must be
       less than 20 percent.


9.2.2  The S/N for the GC signals present in every SICP (including the
       ones for the labeled  standards) must be > 2.5.

9.2.3  The isotoplc ratios (Table 9) must be within the specified control limits.
                                     D-38
                                    23;
O

-------
       NOTE:  If the criterion for acceptable  calibration  listed  in  Section
       9.2.1 (this exhibit)  is  met,  the  analyte-specific  RRF can then  be  con-




       sidered independent of the analyte  quantity  for  the  calibration concen-
       tration range.   The mean RRFs  will be used for all calculations  until




       the routine calibration criteria (this exhibit,  Section 9.4)  are no




       longer met.  At such time,  new mean RRFs  will be calculated  from a new




       set of injections of the calibration solutions.








9.3  Routine Calibration (Continuing  Calibration Check)








     Routine calibrations must be  performed at the beginning of a 12-hour




     period after successful mass  resolution and GC resolution performance




     checks.  A routine calibration is also required at the end of  a 12-hour




     shift.








9.3.1  Inject 2 uL of  the concentration calibration solution HRCC-3 containing




       10 pg/uL of tetra- and pentachlorinated congeners, 25 pg/uL  of hexa-




       and heptachlorinated congeners, 50 pg/uL of octachlorinated  congeners,




       and the respective internal and recovery  standards (Table 5). By using




       the same HRGC/HRMS conditions  as used in Sections 6.1.3 and  6.2  (this




       exhibit), determine and document an acceptable calibration as provided in




       Section 9.4 (this exhibit).








9.4  Criteria for Acceptable Routine  Calibration








     The following criteria must be met before further analysis is  performed.






                                      D-39
                                         23G

-------
     If these criteria are not met, corrective action must be taken.








9.4.1  The measured RRFs [RRF(n) for the unlabeled standards] obtained during




       the routine calibration runs must be within 20 percent of the  mean




       values established during the initial calibration (this exhibit,  Section




       9.1.4.7).








9.4.2  The measured RRFs (RRF(m) for the labeled standards] obtained  during




       the routine calibration runs must be within 20 percent of the  mean




       values established during the initial calibration (this exhibit,  Section




       9.1.4.9).








9.4.3  The ion-abundance ratios (Table 9) must be within the allowed  control




       limits.








9.4.4  If either one of the above criteria (this exhibit, Sections 9.4.1 and




       9.4.2) is not satisfied, the entire initial calibration process (this




       exhibit, Section 9.1) must be repeated.  If the ion-abundance  ratio




       criterion (this exhibit, Section 9.4.3) is not satisfied, refer to the




       note in Section 9.1.4.4 (this exhibit) for resolution.








       NOTE:   An initial calibration must be carried out whenever the HRCC-3,




       the sample fortification or the recovery standard solution is  replaced




       by a new solution from a different lot.








10.   Quality Assessment/Quality Control Procedures






                                      D-40
                                       287

-------
     See Exhibit E for QA/QC requirements.








11.   Sample Preservation








11.1  The sample collection, shipping,  handling,  and chain-of-custody  procedures



      are not described in this  document.   Sample collection personnel will,  to



      the extent possible, homogenize samples  in  the field before filling  the



      sample containers.  This  should minimize or eliminate the  necessity  for



      sample homogenization in  the laboratory. The analyst should make a  judg-




      ment, based on the appearance of  the  sample,  regarding the necessity for



      additional mixing.  If the sample is  clearly inhomogeneous, the  entire



      contents should be transferred to a glass or stainless steel pan for



      mixing with a stainless steel spoon or spatula before removal of a



      sample portion for analysis.








11.2  Grab and composite samples must be collected in glass containers.



      Conventional sampling practices must  be  followed.   The bottle must not  be



      prewashed with sample before collection. Sampling equipment must be free




      of potential sources of contamination.








11.3  Grinding or Blending of Fish Samples.




      If not otherwise specified by the EPA,  the  whole fish (frozen) should be



      blended or ground to provide a homogeneous  sample.  The use of a stain-



      less steel meatgrinder with a 3-  to 5-mm hole size inner plate is recom-




      mended.  In some circumstances, analysis of fillet or specific organs of



      fish may be requested by  the EPA.  If  so requested by the  EPA, the above



      whole fish requirement is  superseded.



                                      D-41







                                        288

-------
11.4  With the exception of the fish and adipose tissues,  which must  be  stored




      at -20° C, all samples must be stored at  4° C,  extracted  within 30 days




      and completely analyzed within 45 days of collection.








11.5  Phase Separation - This is a guideline for phase separation  on  very wet




      (>25 percent water) soil and sediment samples.   Place  a 50-g portion  in a




      suitable centrifuge bottle and centrifuge for 30 minutes  at  2,000  rpm.




      Remove the bottle and mark the interface  level  on the  bottle.   Estimate




      the relative volume of each phase.  With  a disposable  pipet, transfer the




      liquid layer into a clean bottle.  Mix the solid with  a stainless  steel




      spatula and remove a portion to be weighed and  analyzed (percent moisture




      determination, extraction).  Return the remaining solid portion to the




      original sample bottle (empty) or to a clean sample  bottle that is properly




      labeled, and store it as appropriate.  Analyze  the solid  phase  by  using




      only the soil and sediment method.  Take  note of and report  the estimated




      volume of liquid before disposing of the  liquid as a liquid  waste.








      CAUTION:  Finely divided soils and sediments contaminated with  PCDDs/PCDFs




      are hazardous because of the potential for inhalation  or  ingestion of




      particles containing PCDDs/PCDFs (including 2,3,7,8-TCDD).  Such samples




      should be handled in a confined environment (i.e., a closed  hood or a




      glove box).








11.6  Soil, Sediment or Paper Sludge (Pulp) Percent Moisture Determination.




      The percent moisture of soil or sediment  samples showing  detectable
                                      D-42
                                    239

-------
      levels (see note below)  of  at  least  one  2,3,7,8-substituted  PCDD/PCDF

      congener is determined according  to  the  following  recommended procedure.



      Weigh a 9.5- to 10.5-g portion of the  soil  or sediment  sample (+ 0.5 g)

      to three significant  figures.   Dry it  to constant  weight  at  100° C  in an

      adequately ventilated oven.  Allow the sample to cool  in  a desiccator.

      Weigh the dried solid to three significant  figures.  Calculate  and  report

      the percent moisture  on Form (to  be  determined).   Do not  use this solid

      portion of the sample for extraction,  but instead  dispose of it as

      hazardous waste.  The pulp  sample (10  g) should be dried  overnight  in a

      fume hood.



      NOTE:  Until detection limits  are determined  (Section  1.2, this exhibit),

      the lower MCLs (Table 1) may be used to  estimate.the minimum detectable

      levels.
                              Weight  of  wet  soil  -  Weight  of  dry  soil
         Percent  moisture   =     .                                       x   100
                                         Weight of  wet  soil
11.7  Fish Tissue  Lipid  Content  Determination



      The percent  lipid  of  fish  samples  showing  detectable  levels  (see  Section

      11.6 note; this  exhibit) of  at  least  one 2,3,7,8-substituted PCDD/PCDF

      congener is  determined  as  follows:



      Use a separate portion  (2  g)  of the ground frozen  fish sample.  Blend it


                                      D-43
                                     290

-------
      with 6 g anhydrous sodium sulfate, pour the mixture in a 1-cm i.d.

      glass column and extract the lipids by passing two 25-mL portions  of

      methylene chloride through the column and collecting the extract in a

      tared 100-mL round-bottom flask.   Concentrate the extract on a rotary

      evaporator until constant weight is attained.  The percent lipid is

      calculated using the following expression:
                         Weight of residue from extraction (in g)
       Percent lipid  =                                  	        x  100
                           Weight of fish tissue portion (in g)
      Dispose of the lipid residue as a hazardous waste if the results of the

      analysis indicate the presence of PCDOs or PCDFs.



11.8  Adipose Tissue Lipid Content Determination



      Details for the determination of the adipose tissue lipid content are

      provided in Section 12.11.3 (this exhibit).



12.   Extraction and Cleanup Procedures



12.1  Internal standard addition.  Use a portion of 1 g to 1000 g (typical sam-

      ple size requirements for each type of matrix are given in Section 12.2

      of this exhibit and in Table 1) of the sample to be analyzed.   Transfer

      the sample portion to a tared flask and determine its weight.   Except for

      adipose tissue, add an appropriate quantity of the sample fortification

      mixture (this exhibit, Section 3.8) to the sample.  All samples should be


                                      D-44
                                        291

-------
      spiked with 100 uL  of  the  sample  fortification  mixture  to  give  internal




      standard concentrations as  indicated  in Table  1.  As  an example,  for




      l^Cj2~2»3,7,8-TCDD,  a  10-g  soil sample requires the addition  of 1000 pg of




      13C12-2,3,7,8-TCDD  to  give  the requisite  100 ppt  fortification  level.  For




      the fortification of soil,  sediment,  fly  ash, water and fish  tissue




      samples, mix the 100 uL sample fortification solution with 1.5  mL ace-




      tone.   Do not dilute the isooctane  solution for the other  matrices.  The




      fortification of adipose tissue is  carried out  at the time of homogeniza-




      tion (this exhibit,  Section 12.11.2.3).








12.2  Extraction








      The extraction and  purification procedures for  biological  tissue samples




      are described in Sections  12.10 (fish tissue) and 12.11 (adipose tissue)




      of this exhibit.








12.2.1  Sludge/Fuel Oil.   Extract aqueous sludge samples by refluxing a sample




        (e.g., 2 g) with  50  mL toluene  (or  benzene)  in a 125-mL  flask fitted




        with a Dean-Stark water  separator.  Continue  refluxing the  sample




        until all the water  is removed.   Cool the sample, filter the  toluene




        (or  benzene) extract through a  glass-fiber filter,  or equivalent,  into




        a 100-mL round-bottom flask.  Rinse the filter with 10 mL toluene  (or




        benzene), and combine the extract and rinsate.  Concentrate the combined




        solutions to near dryness on a  rotary evaporator at 50°  C (toluene) or




        a Kudema-Danish  (KD) apparatus (benzene).   Use of  an inert gas to




        concentrate the extract  is also permitted.  Proceed with Section 12.2.4




        below.




                                     D-45







                                         292

-------
        NOTE:  If the labeled sludge sample dissolves  In toluene,  treat  it




        according to the instructions in Section 12.2.2  below.   If the labeled




        sludge sample originates from pulp (paper mills),  treat  it according




        to the instructions starting in Section 12.10.1  but without  the  addition




        of sodium sulfate.








12.2.2  Still-Bottom.  Extract still-bottom samples  by mixing a  sample portion




        (e.g., 1.0 g) with 10 mL toluene (or benzene)  in a small beaker  and




        filtering the solution through a glass-fiber filter  (or  equivalent)




        into a 50-mL round-bottom flask.  Rinse the  beaker and filter with 10




        mL toluene (or benzene).  Concentrate the combined toluene (or benzene)




        solutions to near dryness on a rotary evaporator at  50°  C.   A KD appa-




        ratus can be used if benzene is the extraction solvent.  Proceed with




        Section 12.2.4 below.








12.2.3  Fly Ash.   Extract fly ash samples by placing a sample portion  (e.g., 10




        g) and an equivalent amount of anhydrous sodium  sulfate  in a Soxhlet




        extraction apparatus charged with 100 mL toluene (or  benzene), and




        extract for 16 hours using a three cycle/hour schedule.  Cool and




        filter the toluene (or benzene) extract through  a glass-fiber filter




        into a 500-mL round-bottom flask.  Rinse the filter with 5 mL toluene




        (or benzene).  Concentrate the combined toluene  (or benzene) solutions




        to near dryness on a rotary evaporator (toluene) at  50°  C  or a KD




        apparatus (benzene).  Proceed with Section 12.2.4 below.








12.2.4  Transfer the residue to a 125-mL separatory  funnel using 15 mL hexane.






                                      D-46
                                       293

-------
        Rinse the flask with two  5-mL  portions  of hexane  and  add  the  rinses  to




        the funnel.   Shake  two minutes with  50  mL of  5  percent  sodium chloride




        solution, discard the aqueous  layer  and proceed with  Section  12.3




        (this exhibit).








12.2.5  Soil.  Add 10 g anhydrous sodium sulfate to the soil  sample portion




        (e.g., 10 g)  and mix thoroughly  with a  stainless  steel  spatula.  After




        breaking up  any lumps, place the soil/sodium  sulfate  mixture  in  the




        Soxhlet apparatus on top  of a  glass-wool plug (the  use  of an  extraction




        thimble is optional). Add 200 to 250 mL benzene  (or  toluene) to the




        Soxhlet apparatus and reflux for 24  hours.  The solvent must  cycle




        completely through  the system  at least  three  times  per  hour.








12.2.5.1  Transfer the extract from Section  12.2.5 to a KD  apparatus  mounted




          with a three-ball Snyder column (or to a 500-mL round-bottom flask




          for evaporating the toluene  on a rotary evaporator).








12.2.5.2  Add a Teflon™ or  an equivalent boiling chip.  Concentrate  in a 70° C




          water bath  to an  apparent volume of 10 mL.   Remove  the  apparatus from




          the water  bath and allow it  to cool for 5 minutes.








12.2.5.3  Add 50 mL hexane  and a  new boiling chip to  the  KD flask.  Concen-




          trate in a  water  bath to an  apparent  volume of  10 mL.   Remove  the




          apparatus  from the water bath  and  allow to  cool for 5 minutes.








12.2.5.4  Remove and  invert the Snyder column,  and rinse  it down  into the KD






                                      D-47
                                        294

-------
          apparatus with two 1-mL portions of hexane.   Decant  the  contents  of




          the KD apparatus and concentrator tube Into  a 125-mL separatory




          funnel.  Rinse the RD apparatus with two additional  5-mL portions of




          hexane, and add the rinsates to the funnel.   Proceed with Section




          12.3 (this exhibit).








12.2.6  Aqueous Samples.  Mark the water meniscus on the side  of the 1-L sample




        bottle for later determination of the exact sample volume.   Pour the




        entire sample (approximately 1-L) into a 2-L separatory funnel.  Proceed




        with Section 12.2.6.1 (this exhibit).








        NOTE:  A continuous liquid-liquid extractor may be used in place of a




        separatory funnel when experience with a sample from a given source




        indicates that a serious emulsion problem will result  or an emulsion is




        encountered when using a separatory funnel.  Add 60 mL methylene chloride




        to the sample bottle, seal, and shake for 30 seconds to rinse the  inner




        surface.  Transfer the solvent to the extractor.  Repeat the sample bot-




        tle rinse with an additional 50- to 100-mL portion of  methylene  chloride




        and add the rinsate to the extractor.  Add 200 to 500  mL methylene




        chloride to the distilling flask, add sufficient reagent water to  ensure




        proper operation, and extract for 24 hours.  Allow to  cool, then detach




        the distilling flask.  Dry and concentrate the extract as  described in




        Sections 12.2.6.1 and 12.2.6.2 (this exhibit).  Proceed with Section




        12.2.6.3 (this exhibit).








12.2.6.1  Add 60 mL methylene chloride to the sample bottle, seal, and shake for






                                      D-48
                                       295

-------
          30 seconds to rinse  the  inner  surface.  Transfer  the  solvent  to  the




          separatory funnel and  extract  the  sample  by  shaking the  funnel for




          two minutes with periodic  venting.  Allow the  organic layer to sepa-




          rate from the water  phase  for  a minimum of 10  minutes.   If the emul-




          sion interface between layers  is more  than one-third  the volume  of




          the solvent layer, the analyst must employ mechanical techniques  to




          complete the phase separation.  Collect the  methylene chloride into a




          KD apparatus (mounted  with a 10-mL concentrator tube) by passing  the




          sample extracts through  a  filter funnel packed with a glass-wool  plug




          and 5 g anhydrous sodium sulfate.  Repeat the  extraction twice with




          fresh 60-ml portions of  methylene  chloride.  After the third  extrac-




          tion, rinse the sodium sulfate with an additional 30  mL  methylene




          chloride to ensure quantitative transfer. Combine all extracts  and




          the rinsate in the KD  apparatus.








12.2.6.2  Attach a Snyder column and concentrate the extract on a  water bath




          until the apparent volume  of the liquid is 5 mL.   Remove the  KD




          apparatus and allow  it to  drain and cool  for at least 10 minutes.




          Remove the Snyder column,  add  50 mL hexane,  re-attach the Snyder




          column and concentrate to  approximately 5 mL.  Add a  new boiling chip




          to the KD apparatus  before proceeding  with the second concentration




          step.  Rinse the flask and the lower joint with two 5-mL portions




          of hexane and combine  the  rinsates with the  extract to give a final




          volume of about 15 mL.








12.2.6.3  Determine the original sample  volume by filling the sample bottle to






                                     D-49
                                        296

-------
          the mark with water and transferring the water to a 1000-mL graduated




          cylinder.  Record the sample volume to the nearest 5 mL.   Proceed




          with Section 12.3 (this exhibit).








12.3  Partition the extract (15 mL hexane) against 40 mL of 20 percent (w/v)




      aqueous potassium hydroxide (KOH).  Shake for two minutes.   Remove and




      discard the aqueous layer (bottom).  Repeat the base washing until no




      color is visible in the bottom layer (perform a maximum of  four base




      washings).  Strong base (KOH) is known to degrade certain PCDDs/PCDFs,




      so contact time must be minimized.








12.4  Partition the extract (15 mL hexane) against 40 mL of 5 percent (w/v)




      aqueous sodium chloride.  Shake for two minutes.  Remove and discard the




      aqueous layer (bottom).








12.5  Partition the extract against 40 mL concentrated sulfuric acid.  Shake




      for two minutes.  Remove and discard the sulfuric acid layer (bottom).




      Repeat the acid washing until no color is visible in the acid layer




      (perform a maximum of four acid washings).








12.6  Partition the extract against 40 mL of five percent (w/v) sodium chloride.




      Shake for two minutes.  Remove and discard the aqueous layer (bottom).




      Dry the extract by pouring it through a funnel containing anhydrous




      sodium sulfate and collect it in a 50-mL round-bottom flask.   Rinse the




      sodium sulfate with two 15-mL portions of hexane, add the rinsates to the




      50-mL flask, and concentrate the hexane solution to near dryness on a






                                      D-50
                                       9 O - ?
                                       &a ^

-------
      rotary evapotator (35° C water bath),  making sure  all  traces  of  toluene




      (when applicable) are removed.   (Use of  blow-down  with an  inert  gas  to




      concentrate the extract is  also permitted.)








12.7  Pack a gravity column (glass,  300 mm x 10.5  mm), fitted with  a Teflon™




      stopcock, in the following  manner:   Insert a glass-wool plug  into the




      bottom of the column.  Add  a 4-g layer of  sodium sulfate.   Add a 4-g




      layer of Woelm® Super 1 neutral alumina.   Tap the  top  of the  column




      gently.  Woelm® Super 1 neutral alumina  need not be activated or cleaned




      before use, but it should be stored in a sealed desiccator.   Add a 4-g




      layer of anhydrous sodium sulfate to cover the alumina.  Elute with  10 mL




      hexane and close the stopcock just before  exposure of  the  sodium sulfate




      layer to air.  Discard the  eluate.  Check  the column for channeling.   If




      channeling is present, discard the column.   Do not tap a wetted  column.








12.8  Dissolve the residue from Section 12.6 (this exhibit)  in 2 mL hexane and




      apply the hexane solution to the top of  the  column.  Elute with  enough




      hexane (3-4 mL) to complete the transfer of  the sample cleanly to the




      surface of the alumina.  Discard the eluate.








12.8.1  Elute with 10 mL of 8 percent (v/v)  methyleiie chloride in hexane.








12.8.2  Elute the PCDDs and PCDFs from the column  with 15 mL of  60  percent




        (v/v) methylene chloride  in hexane and collect this  fraction in a




        conical shaped (15 mL) concentrator  tube.
                                      D-51
                                         298

-------
12.9  Carbon Column Cleanup








      Prepare a Carbopak C/Celite 545® column as described in Section 7.1.2




      (this exhibit).








12.9.1  With a carefully regulated stream of nitrogen, concentrate the




        60-percent fraction (this exhibit, Section 12.8.2) to about 2 mL.




        Rinse the Carbopak C/Celite 545® with 5 raL toluene followed by 2 mL of




        75:20:5 methylene chloride/methanol/benzene, 1 mL of 1:1 cyclohexane/




        methylene chloride, and 5 mL hexane.  The flow rate should be less  than




        0.5 mL/min.  Discard the rinsates.  While the column is still wet with




        hexane, add the sample concentrate to the top of the column.   Rinse the




        concentrator tube which contained the sample concentrate twice with




        1 mL hexane and add the rinsates to the top of the column.  Elute the




        column sequentiallly with two 2-mL portions of hexane, 2 mL cyclohexane/




        methylene chloride (50:50, v/v), and 2 mL methylene chloride/methanol/




        benzene (75:20:5, v/v).  Combine these eluates; this combined fraction




        may be used as a check on column efficiency.   Now turn the column




        upside down and elute the PCDD/PCDF fraction with 20 mL toluene.




        Verify that no carbon fines are present in the eluate.








12.9.2  Concentrate the toluene fraction to about 1 mL on a rotary evaporator




        by using a water bath at 50° C.  Carefully transfer the concentrate into




        a 1-tnL minivial and, again at elevated temperature (50° C), reduce  the




        volume to about 100 uL using a stream of nitrogen and a sand bath.




        Rinse the rotary evaporator flask three times with 300 uL of a solution






                                      D-52
                                       290

-------
        of 1 percent toluene  In methylene  chloride.   Add  10 uL  for  soil, sedi-




        ment,  and water,  or 50  uL  for sludge,  still-bottom and  fly  ash of the




        tridecane recovery standard  solution.   Store  the  sample at  room tempera-




        ture in the dark.








12.10  Extraction and Purification Procedures  for Fish  and Paper Pulp Samples








12.10.1  Add 30 g anhydrous sodium sulfate to  a 10-g  portion  of a homogeneous




         fish sample (this exhibit,  Section 11.3) and mix thoroughly with a




         stainless steel  spatula.  After breaking up  any  lumps, place the




         fish/sodium sulfate  mixture in the Soxhlet apparatus on top of a glass-




         wool plug.  Add  200  mL hexane/methylene chloride (1:1) to  the Soxhlet




         apparatus and reflux for  12 hours.  The solvent  must cycle completely




         through the system at  least three times per  hour.  Follow  the same




         procedure for the dried (this exhibit, Section 11.6) paper pulp samples.








12.10.2  Transfer the fish or paper  pulp extract from Section 12.10.1 to a KD




         apparatus equipped with a Snyder  column.








12.10.3  Add a Teflon™ or an  equivalent boiling chip.  Concentrate  the extract




         in a water bath  to an  apparent volume of 10  mL.  Remove the apparatus




         from the water bath  and allow to  cool for 5  minutes.








12.10.4  Add 50 mL isooctane  and a new boiling chip to  the KD flask.  Concentrate




         in a water bath  to an  apparent volume of 5 mL.  Remove the apparatus




         from the water bath  and allow to  cool for 5  minutes.






                                      D-53
                                         300

-------
         NOTE:  The methylene chloride must have been completely removed before




         proceeding with the next step.








12.10.5  Remove and Invert the Snyder Column and rinse it into the KD apparatus




         with two 1-mL portions of hexane.  Decant the contents of the KD




         apparatus and concentrator tube into a 125-mL separatory funnel.




         Rinse the KD apparatus with two additional 5-mL portions of hexane and




         add the rinsates to the funnel.  Proceed with the cleanup according to




         the Instructions starting in Section 12.5 (this exhibit).








12.11  Extraction and Purification Procedures for Human Adipose Tissue








12.11.1  Human adipose tissue samples must be stored at -20° C from the time of




         collection until the time of analysis.  The use of chlorinated mate-




         rials during the collection of the sample must be avoided.  Samples




         are handled with stainless steel forceps, spatulas, or scissors.  All




         sample bottles (glass) are cleaned as specified in the note appearing




         in Section 6.3 (this exhibit).  Teflon"1-!ined caps should be used.








12.11.2  Adipose Tissue Extraction Procedure








12.11.2.1  Weigh to the nearest 0.01 g a 10-g portion of a frozen adipose




           tissue sample into a culture tube (2.2 x 15 cm).








          'NOTE:  The sample size may be smaller, depending on availability.




           In such a situation, the analyst is required to adjust the volume of






                                      D-54
                                         301

-------
           the internal standard solution added to the sample to meet  the  for-




           tification level stipulated  in Table 1.








12.11.2.2  Allow the adipose tissue specimen  to reach room temperature (up to 2




           hours).








12.11.2.3  Add 10 mL methylene chloride and 100 uL of the  sample fortification




           solution.   Homogenize the mixture  for approximately 1 minute  with a




           tissue homogenizer.








12.11.2.4  Allow the mixture to separate, and remove the methylene chloride




           extract  from the residual solid material with a disposable  pipet.




           Percolate the methylene  chloride through a filter funnel containing




           a clean  glass-wool plug  and  10 g anhydrous sodium sulfate.'  Collect




           the dried extract in a graduated 100-mL volumetric flask.








12.11.2.5  Add a second 10-mL portion of methylene chloride to the sample  and




           homogenize for 1 minute.   Decant the solvent, dry it, and  transfer




           it to the 100-mL volumetric  flask  (this exhibit, Section 12.11.2.4).








12.11.2.6  Rinse the culture tube with  at least two additional portions  of




           methylene chloride (10 mL each), and transfer the entire contents




           to the filter funnel containing the anhydrous sodium sulfate.  Rinse




           the filter funnel and the anhydrous sodium sulfate contents with




           additional methylene chloride (20  to 40 mL) into the 100-mL flask.




           Discard  the sodium sulfate.






                                     D-55
                                   302

-------
                                _

12.11.2.7  Adjust  the volume  to- Ebe"."rDO.=inL*vmark-witij 'inethylene chloride.




12.11.3  Adipose Tissue Lipid Content Determination




12.11.3.1  Preweigh a clean 1-dram glass vial to the nearest 0.0001 g on an

           analytical balance tared to zero.




12.11.3.2  Accurately transfer  1.0 mL of the final extract  (100 mL) from Section

           12.11.2.6 (this exhibit) to the 1-dram vial.  Reduce the volume of

           the extract on a water bath (50-60° C) by a gentle stream of

           purified nitrogen until an oily residue remains.  Nitrogen blow-down

           is continued until a constant weight is achieved.




12.11.3.3  Accurately weigh the 1-dram vial with the residue to the nearest

           0.0001 g and calculate the weight of the lipid present  in the vial

           based on the difference of the weights.




12.11.3.4  Calculate the percent lipid content of the original sample to the

           nearest 0.1 percent as shown below:
                                                  x  vext
                  Lipid Content, LC (%)  =	 x   100

                                              wat  x  val
           where
                   V/ir = weight of the lipid residue to the nearest 0.0001 g


                         calculated from Section 12.11.3.3 (this exhibit),



                                      D-56





                                           303 ^

-------
                  Vext = total volume (100 mL)  of  the  extract  in  mL  from




                         Section 12.11.2.6 (this exhibit)








                   Wat = weight of  the original adipose  tissue sample  to  the




                         nearest 0.01 g from Section 12.11.2.1 (this exhibit),




                         and
                       = volume of the  aliquot  of  the  final  extract  in mL




                         used for the quantitative measure of  the  lipid residue




                         (1.0 mL).
12.11.3.5  Record the lipid residue  measured  in Section 12.11.3.3  (this  exhibit)




           and the percent  lipid content  from Section 12.11.3.4  (this  exhibit).








12. 11. A  Adipose Tissue Extract Concentration








12.11.4.1  Quantitatively transfer the  remaining extract volume  (99.0  mL)  to a




           500-mL round-bottom flask.   Rinse  the volumetric flask  with 20  to 30




           mL of additional methylene chloride to ensure quantitative  transfer.








12.11.4.2  Concentrate the  extract on a rotary evaporator and a  water




           bath at 40°C until an oily residue remains.








12.11.5  Adipose Tissue Extract Cleanup Procedures








12.11.5.1  Add 200 mL hexane to the  lipid residue in  the 500-mL  Erlenmeyer






                                     D-57
                                          304

-------
           flask and swirl the flask to dissolve the residue.








12.11.5.2  Slowly add, with stirring, 100 g of 40-percent w/w  sulfuric-acid-




           impregnated silica gel.  Stir with a magnetic stirrer for two  hours




           at room temperature.








12.11.5.3  Allow the solid phase to settle and decant the liquid through  a




           powder funnel containing 20 g anhydrous sodium sulfate into  another




           500-mL Erlenmeyer flask.








12.11.5.4  Rinse the solid phase with two 50-mL portions of hexane.   Stir each




           rinse for 15 minutes, decant, and dry as described  under  Section




           12.11.5.3.  Combine the hexane extracts from Section 12.11.5.3




           (this exhibit) with the rinses.








12.11.5.5  Rinse the sodium sulfate in the powder funnel with  an additional




           25 mL hexane and combine this rinse with the hexane extracts from




           Section 12.11.5.4 (this exhibit).








12.11.5.6  Prepare an acidic silica column as follows:   Pack a 2-cm  x 10-cm




           chromatographic column with a glass-wool plug, add  approximately




           20 mL hexane, add 4 g silica gel and allow to settle, then add 16  g




           of 40-percent w/w sulfuric-acid-impregnated-silica  gel and allow to




           settle.  Elute the excess hexane from the column until the solvent




           level reaches the top of the chromatographic packing.  Verify  that




           the column does not have any air bubbles and channels.






                                      D-58
                                       SOS

-------
12.11.5.7  Quantitatively transfer the hexane extract from the Erlenmeyer flask




           (this exhibit, Sections 12.11.5.3 through 12.11.5.5) to the silica




           gel column reservoir.   Allow the hexane extract to percolate through




           the column and collect the eluate in a 500-mL KD apparatus.








12.11.5.8  Complete the elution by percolating 50 mL hexane through the column




           into the KD apparatus.  Concentrate the eluate on a steam bath to




           approximately 5 mL.   Use nitrogen blow-down to bring the final




           volume to about 100  uL.








           NOTE:  If the silica gel impregnated with 40-percent sulfuric acid




           is highly discolored throughout the length of the adsorbent bed,




           the cleaning procedure must be repeated beginning with Section




           12.11:5.1 (this exhibit).








12.11.5.9  The extract is ready for the alumina and carbon cleanups described




           in Sections 12.7 through 12.9.2 (this exhibit).








13.   Analytical Procedures.








13.1  Remove the sample extract or blank from storage.  With a stream of dry,




      purified nitrogen, reduce the extract volume to 10 uL or 50 uL (the




      volume of the tridecane recovery standard solution) as stipulated above




      (this exhibit, Section 12.9.2).








13.2  Inject a 2-uL aliquot of  the extract into the GC, operated under the






                                      D-59

-------
      conditions previously used (this exhibit, Section 6.2) to produce accept-


      able results with the performance check solution.






13.3  Acquire SIM data according to Section 6.1.3 (this exhibit).  Use the same


      acquisition and mass spectrometer operating conditions previously used to


      determine the relative response factors (this exhibit, Sections 9.1.4.6



      through 9.1.4.9).  Ions characteristic for polychlorinated diphenyl


      ethers are included in the descriptors listed in Table 6.  Their presence


      is to monitor their interference during the characterization of PCDFs.






      NOTE:  The acquisition period must at least encompass the PCDD/PCDF



      overall retention time window previously determined (Section 8.1, this


      exhibit).  Selected ion current profiles (SICP) for the lock-mass ions


      (one per mass descriptor) must also be recorded and included in the data


      package as deliverables.  These SICPs must be true representations of the


      evolution of the lock-mass ions amplitudes during the HRGC/HRMS run.


      (See this exhibit, Section 8.2.2 for the proper level of reference compound


      to be metered into the ion chamber.)  It is recommended to examine the



      lock-mass ion SICP for obvious basic sensitivity and stability changes


      of the instrument during the GC/MS run that could affect the measurements


      [Y. Tondeur et al., Anal. Chem. 56, 1344 (1984)].  Report any discrepancies



      in the case narrative.






13.4  Identification Criteria






      For a gas chromatographic peak to be identified as a PCDD or PCDF, it




                                      D-60
                                        o r>
                                        OU

-------
      must meet all of the following criteria:








13.4.1  Relative Retention Times.








13.4.1.1  For 2,3,7,8-substituted  congeners,  which have an isotopically labeled




          internal or recovery standard present in the sample extract  (this




          represents a total of 10 congeners  including OCDD;  Tables  2  and 3),




          the relative retention time (RRT; at  maximum peak height)  of the




          sample components (i.e., the two ions used for quantification purposes




          listed in Table 6) must  be within -1  and +3 seconds of the retention




          time of the peak for the isotopically labeled internal or  recovery




          standard at m/z corresponding to the  first characteristic  ion (of  the




          set of two; Table 6) to  obtain a positive identification of  these




          nine 2,3,7,8-substituted PCDDs/PCDFs  and OCDD.








13.A.1.2  For 2,3,7,8-substituted  compounds,  that do not have an isotopically




          labeled internal standard present in  the sample extract (this repre-




          sents a total of six congeners;  Table 3), the relative retention time




          must fall within the established homologous retention time windows by




          analyzing the column performance check solution (this exhibit,  Section




          8.1.3).  Identification  of OCDF is  based on its retention  time rela-




          tive to ^Cj2~°CDD as determined from the daily routine calibration




          results.









13.4.1.3  For non-2,3,7,8-substituted compounds (tetra through octa; totaling




          119 congeners),  the retention time  must be within the corresponding






                                      D-61
                                         308

-------
          homologous retention time windows established by analyzing the column
          performance check solution (this exhibit, Section 8.1.3).

13.4.1.4  The ion current responses for both ions used for quantitative pur-
          poses (e.g., for TCDDs: m/z 319.8465 and 321.8936) must reach maximum
          simultaneously (+ 2 seconds).

13.4.1.5  The ion current responses for both ions used for the labeled stan-
          dards (e.g., for 13C12-TCDD: m/z 331.9368 and m/z 333.9339) must
          reach maximum simultaneously (+ 2 seconds).

          NOTE:  The analyst is required to verify the presence of 1,2,8,9-TCDD
          and 1,3,4,6,8-PeCDF (this exhibit, Section 8.1.3) in the SICPs of the
          daily performance checks.  Should either one compound be missing, the
          analyst is required to report that observation with the results
          associated with the sample batch as it may indicate a potential
          problem with the ability to detect all the PCDDs/PCDFs.
                                           \
13.4.2  Ion Abundance Ratios

13.4.2.1  The integrated ion current for the two ions used for quantification
          purposes must have a ratio between the lower and upper limits
          established for the homologous series to which the peak is assigned.
          See Sections 9.1.4.3 and 9.1.4.4 (this exhibit) and Table 9 for
          details.
                                      D-62
                                         3'

-------
13.4.3  Signal-to-Noise Ratio




13.4.3.1  All ion current intensities  must be £ 2.5 times noise level for posi-


          tive identification of  a PCDD/PCDF compound or a group of coeluting


          isomers.   Appendix C describes  the procedure to be followed for the


          determination of the S/N.




13.4.4  Polychlorinated Diphenyl  Ether Interferences




13.4.4.1  In addition to the above criteria, the identification of a GC peak as


          a PCDF can only be made if no signal having a S/N >^ 2.5 is detected,


          at the same retention time (+ 2 seconds), in the corresponding PCDPE


          channel.




14.   Calculations




14.1  For gas chromatographic peaks  that  have met the criteria outlined in


      Sections 13.4.1.1 through 13.4.3.1  (this exhibit), calculate the concen-


      tration of the PCDD or PCDF compounds using the formula:




                                         Ax  x  Qls
                                    Als   x  W  x  RRF(n)
      where
                   concentration of unlabeled PCDD/PCDF congeners (or group of


                   coeluting  isomers within an homologous series) in pg/g,


                                      D-63

                                       n  -« ,}
                                       Ol(J

-------
              Ax = sum of the integrated ion abundances of the quantification

                   ions (Table 6) for unlabeled PCDDs/PCDFs,
             Ais = sum °f *-he integrated ion abundances of the quantification  ions

                   (Table 6) for the labeled internal standards,




             Qis = quantity, in pg, of the internal standard added  to  the  sample

                   before extraction,




               W «• weight, in g, of the sample (solid or liquid), and
          RRF(n) = calculated mean relative response factor for the analyte

                   [RRF(n) with n = 1 to 17;  Section 9.1.4.7,  this  exhibit].




         If the analyte is identified as one  of the 2,3,7,8-substituted  PCDDs
         or PCDFs , RRF(n) is the value calculated using the equation in Section

         9.1.4.7 (this exhibit).  However,  if it is a non-2,3,7,8-substituted

         congener, the RRF(k) value is the  one calculated using the equation in

         Section 9.1.4.8.2 (this exhibit).   [RRF(k) with k = 27 to  30.]




14.2  Calculate the percent recovery of the nine internal standards measured in

      the sample extract, using the formula:
                                                          *  Qrs
        Internal standard percent recoverr  =                      —  x  100
                                               Qis   x  Ars  x
      where

                                      D-64
                                       311

-------
                   sum of  the  integrated  Ion  abundances  of  the  quantification




                   ions (Table 6)  for  the labeled  internal  standard,
             Ars = sum of  the  integrated  ion  abundances  of  the  quantification




                   ions (Table 6)  for  the labeled  recovery  standard;  the  selection




                   of  the  recovery standard depends  on the  type of  congeners  (see




                   Table 5,  footnotes),








             Qis ° quantity,  in pg,  of the internal  standard added  to the sample



                   before  extraction,








             Qrs » quantity,  in pg,  of the recovery  standard added  to the




                   cleaned-up  sample residue  before  HRGC/HRMS analysis, and








          RRF(m) = calculated  mean relative response factor for the labeled




                   internal  standard relative to the appropriate (see Table 5,




                   footnotes)  recovery standard.   This  represents the mean
                   obtained  in  Section  9.1.4.9  (this  exhibit)  [RRF(m)  with




                   m = 18 to 26].








                   NOTE:   For human adipose  tissue, adjust the percent




                   recoveries by  adding 1  percent  to  the calculated value.








14.3  If the concentration in the 10-uL or 50-uL final extract of any  of the




      fifteen 2,3,7,8-substituted PCDD/PCDF  compounds (Table 3) exceeds the




      upper method calibration  limits  (MCL)  listed in Table 1  (e.g., 200 pg/uL






                                     D-65







                                     312

-------
      for TCDD in soil), the linear range of response versus concentration may




      have been exceeded, and, after contacting EPA/SHO, a reanalysis of the




      sample (using one tenth aliquot) should be undertaken.  The volumes of




      the internal and recovery standard solutions should remain the same as




      described for the sample preparation (this exhibit, Sections 12.1 to




      12.9.3).  For the other congeners (including OCDD), however, report the




      measured concentration and Indicate that the value exceeds the MCL.








14.4  The total concentration for each homologous series of PCDD and PCDF is




      calculated by summing up the concentrations of all positively identified




      isomers of each homologous series.  Therefore, the total should also




      include the 2,3,7,8-substituted congeners.  The total number of GC




      signals included in the homologous total concentration value must be




      specified in the report.








14.5  Sample-Specific Estimated Detection Limit








      The sample-specific estimated detection limit (EDL) is the concentration




      of a given analyte required to produce a signal with a peak height of at




      least 2.5 times the background signal level.  An EDL is calculated for




      each 2,3,7,8-substituted congener that is not identified, regardless of




      whether or not other non-2,3,7,8-substituted isomers are present.  Two




      methods of calculation can be used, as follows, depending on the type of




      response produced during the analysis of a particular sample.








14.5.1  Samples giving a response for both quantification ions (Tables 6 and 9)




        that is less than 2.5 times the background level.




                                      D-66
                                        3
13

-------
14.5.1.1  Use the  expression  for EDL  (specific 2,3,7,8-substltuted PCDD/PCDF)

          below to calculate  an EDL for ejach absent 2,3,7,8-substltuted PCDD/

          PCDF (I.e.,  S/N  < 2.5).  The background level Is determined by

          measuring the  range of the  noise  (peak  to peak) for the two quanti-

          fication ions  (Table 6) of  a particular 2,3,7,8-substituted isomer

          within an homologous series, in the region of the SICP trace

          corresponding  to the elution of the internal standard (if  the congener

          possesses an internal standard) or in the region of the SICP where

          the congener is  expected to elute by comparison with the routine

          calibration  data (for those congeners that do not have a 13c-iabeie(j

          standard), multiplying that noise height  by 2.5, and relating the

          product  to an  estimated concentration that would produce that product

          height.



          Use the  formula:
                                                        2.5  x  Ax   x  Qj_s
          EDL (specific 2,3,7,8  subst.-PCDD/PCDF)   •
                                                      Als  x  W   x   RRF(n)
          where
              EDL = estimated  detection  limit  for  homologous  2,3,7,8-substituted

                    PCDDs/PCDFs.
              Ax,  Ajs,  W,  RRF(n),  and Q^s  retain  the  same  meanings  as  defined

              in Section 14.1.
                                     D-67
                                           314

-------
14.5.2  Samples characterized by a response above the background level  with a

        S/N of at least 2.5 for at least one of the quantification ions

        (Tables 6 and 9).



14.5.2.1  When the response of a signal having the same retention time  as  a

          2,3,7,8-substituted congener has a S/N in excess of 2.5 and does not

          meet any of the other qualitative identification criteria listed in

          Section 13.4, calculate the "Estimated Maximum Possible Concentration'

          (EMPC) according to the expression shown in Section 14.1.



14.6  The relative percent difference (RPD) is calculated as follows:
                                     I  Si - S2 |
                           RPD  =  	  x  100
                                   ( Si + S2 ) /  2
      Si and 82 represent sample and duplicate sample results.



14.7  The 2,3,7,8-TCDD toxic equivalents (TE) of PCDDs and PCDFs present in the

      sample are calculated, only at the data user's request,  according to the

      method recommended by the Chlorinated Dioxins Workgroup  (CDWG) of the EPA

      and the Center for Disease Control (CDC).   This method assigns a 2,3,7,8-

      TCDD toxicity equivalency factor (TEF) to each of the fifteen 2,3,7,8-

      substituted PCDDs and PCDFs (Table 3) and the non-2,3,7,8-substituted

    .  compounds as shown in Table 11.  The 2,3,7,8-TCDD equivalent of the PCDDs

      and PCDFs present in the sample is calculated by summing the TEF times

      their concentration for each of the compounds or groups  of compounds


                                      D-68
                                         315

-------
      listed In Table 11.   The  exclusion of  other  homologous  series  such  as



      mono-, di-,  tri- and  octachlorinated dibenzodioxins  and dibenzofurans



      does not mean that  they are  non-toxic.   Their  toxicity, as known  at this



      time, is much less  than the  toxicity of  the  compounds listed in Table  11.



      The above procedure for calculating the  2,3,7,8-TCDD toxic equivalents  is




      not claimed  by the  CDUG to be  based on a thoroughly  established scientific



      foundation.   The procedure,  rather, represents a  "Consensus  recommendation



      on science policy".   Since the procedure may be changed in the future,



      reporting requirements  for PCDD and PCDF data  would  still  include the



      reporting of the analyte  concentrations  of the PCDD/PCDF congener as



      calculated in Sections  14.1  and 14.4.








14.7.1  Two-GC Column TEF Determination








        Isomer specificity  for  all 2,3,7,8-substituted  PCDDs/PCDFs cannot be



        achieved on the 60-m  DB-5  GC column  alone.  In  order  to  determine the



        proper concentrations of the individual 2,3,7,8-substituted  congeners,




        the sample extract  must be reanalyzed  on a 60-m SP-2330  (or  SP-2331)  GC



        column.








14.7.1.1  The concentrations  of 2,3,7,8-TCDD (see  note  below), 2,3,4,7,8-PeCDF,



          l,2,3,4,6,7,8HpCDD, 1,2,3,4,6,7,8-HpCDF, and  1,2,3,4,7,8,9-HpCDF are



          calculated from the analysis of the  sample extract  on  the  60-m  DB-5




          fused-silica column.  The  experimental conditions remain the  same  as the




          conditions described  previously in Section 13 (this exhibit), and  the




          calculations are  performed as outlined in  Section 14 (this exhibit).






                                      D-69





                                        316

-------
14.7.1.2  The concentrations of 2,3,7,8-TCDF, 1,2,3,7,8-PeCDD and -PeCDF,




          1,2,3,4,7,8-HxCDD and -HxCDF, 1,2,3,6,7,8-HxCDD and -HxCDF,




          1,2,3,7,8,9-HxCDD and -HxCDF, and 2,3,4,6,7,8-HxCDF are obtained from




          the analysis of the sample extract on the second fused-silica capil-




          lary column (confirmation GC column:  60 m SP-2330).  However, the




          GC/MS conditions must be altered so that:  (1) only the first three




          descriptors (i.e., tetra-, penta-, and hexachlorinated congeners)




          of Table 6 are used; and (2) the switching time between descriptor 2




          (pentachlorinated congeners) and descriptor 3 (hexachlorinated



          congeners) takes place following the elution of ^C  ~!»2>3,7,8-PeCDD.




          The concentration calculations are performed as outlined in Section




          14 (this exhibit).








          NOTE:  The confirmation and quantification of 2,3,7,8-TCDD (this




          exhibit, Section 14.7.1.1) may be accomplished on the SP-2330 GC




          column instead of the DB-5 column, provided the criteria listed in




          Section 8.1.2 (this exhibit) are met and the requirements described




          in Section 2.2 (Exhibit E) are followed.








14.7.1.3  For a gas chromatographic peak to be identified as a 2,3,7,8-




          substituted PCDD/PCDF congener, it must meet the ion abundance and




          signal-to-noise ratio criteria listed in Sections 13.4.2 and 13.4.3




          (this exhibit), respectively.  In addition, the retention time




          identification criterion described in Section 13.4.1.1 (this exhibit)




          applies here for congeners for which a carbon-labeled analogue is




          available in the sample extract.  However, the relative retention






                                      D-70
                                        317

-------
time (RRT) of the 2,3,7,8-substituted congeners for which no carbon-




labeled analogues are  available must fall within 0.006 units of the




carbon-labeled standard  RRT.   Experimentally, this is accomplished by




using the attributions described  in Table 12 and the results from the




routine calibration run  on  the SP-2330 column.
                           D-71
                                318

-------
APPENDICES
      318

-------
                         APPENDIX A
Procedure  for the Collection, Handling,  Analysis, and Reporting
  Requirements of Wipe Tests Performed within the Laboratory
                           D-72
                               320

-------
     This procedure  Is designed for the periodic evaluation of potential con-




tamination by 2,3,7,8-substituted PCDD/PCDF congeners of the working areas




inside the laboratory.








                              PERFORMING WIPE TEST








     Perform the wipe tests on surface areas of two inches by one foot with




laboratory wipers saturated with distilled-in-glass acetone using a pair of




clean stainless steel forceps.  Use one wiper for each of the designated areas.




Combine the wipers to one composite sample in an extraction jar containing 200




mL distilled-in-glass acetone.  Place an equal number of unused wipers in 200




mL acetone and use this as a control.








                          COMPOSITE SAMPLE PREPARATION








     Close the jar containing the wipers and 200 mL acetone and extract for 20




minutes using a wrist-action shaker.  Transfer the extract into a KD apparatus




fitted with a concentration tube and a three-ball Snyder column.  Add two




Teflon™ or Carborundum™ boiling chips and concentrate the extract to an apparent




volume of 1.0 mL on  a steam bath.  Rinse the Snyder column and the KD assembly




with two 1-mL portions of hexane into the concentrator tube.  Add 100 uL of the




sample fortification solution to the concentrator tube (Section 3.8, this




exhibit), and concentrate its contents to near dryness with a gentle stream of




nitrogen.  Add 1.0 mL hexane to the concentrator tube, and swirl the solvent on




the walls.
                                      D-73
                                           o f
                                            9 I
                                            >-* .L

-------
     Prepare a neutral alumina column as described in Section 12.7  (this




exhibit) and follow the steps outlined in Sections 12.8 thru 12.8.2 (this




exhibit).








     Add 10 uL of the recovery standard solution as described in Section




12.9.2 (this exhibit).








                                EXTRACT ANALYSIS








     Concentrate the contents of the vial to a final volume of 10 uL (either in




a minivial or in a capillary tube).   Inject two uL of each extract  (wipe  and




control) onto a capillary column and analyze for 2,3,7,8-substituted PCDDs/PCDFs




as specified in the analytical method Section 13 (this exhibit).  Perform




calculations according to Section 14 (this exhibit).








                                REPORTING FORMAT








     Report the presence of 2,3,7,8-substituted PCDDs and PCDFs as  a quantity




(pg or ng) per wipe test experiment  (WTE).  Under the conditions outlined in




this analytical protocol, a lower limit of calibration of 25 pg/WTE is expected




for 2,3,7,8-TCDD.   A positive response for the blank (control) is defined as a




signal in the TCDD retention time window at any of the masses monitored which




is equivalent to or above 8 pg of 2,3,7,8-TCDD per WTE.  For other congeners,




use the multiplication factors listed in Table 1, footnote (a) (e.g., for OCDD,




the lower MCL is 25 x 5 = 125 pg/WTE and the positive response for the blank
                                      D-74
                                          322

-------
would be 8 x 5 = 40 pg).  Also, report the recoveries of the internal  standards




during the simplified cleanup procedure.








                            FREQUENCY OF WIPE TESTS








     At a minimum, wipe tests should be performed when there is  evidence  of




contamination in the method blanks.








                               CORRECTIVE ACTION








     An upper limit of 25 pg per TCDD isomer and per wipe test experiment is




allowed.  (Use multiplication factors listed in footnote (a) from Table  1 for




other congeners.)  This value corresponds to the lower calibration limit  of the




analytical method.  Steps to correct the contamination must be taken whenever




these levels are exceeded.  To that effect,  first vacuum the working places




(hoods, benches, sink) using a vacuum cleaner equipped with a high-efficiency




particulate absorbent (HEPA) filter and then wash with a detergent.  A new set




of wipes should be analyzed before anyone is allowed to work in  the  dioxin area




of the laboratory.
                                      D-75
                                          323

-------
                              APPENDIX B
                   Standards  Traceabillty  Procedure
NOTE:  The content of this appendix is  based  on the  assumption  that EPA




will have within its repository a mixture  (named S2)  containing known




concentrations (e.g., 100 pg/uL) of the eight 13C-labeled  2,3,7,8-substi-




tuted PCDD/PCDF congeners marked with an asterisk in  Table 3  of this




exhibit, and a second solution (named SI,  with the same  concentration as




used for S2) containing the eight corresponding unlabeled  analogues.
                                 D-76
                                      ,24

-------
     All laboratories are expected to maintain traceablllty of their standard



solutions by verifying that all standard solutions used for direct quantifica-



tion of samples agree in chemical Identity and concentration with the EPA



primary standard solutions.  The specific procedures are described below:







     Each time a new laboratory working standard solution (W) is prepared, the



identities and concentrations of the components of this solution must be veri-



fied.  Verifications of the identities of the compounds are to be carried out



by HRGC/HRMS.  The EPA reference standard (S) and the laboratory working stan-



dard (W) are to be analyzed under the instrumental conditions described in this



exhibit, which are appropriate for the analysis of PCDDs and PCDFs.  Two



criteria must be satisfied to verify the identifications:






     o    Elution of the component(s) of the laboratory working standard must



          be at the same retention time(s) as those of the component(s) of the



          EPA reference standard solution.







     o    Concentration^ ) of the laboratory working standard component(s) must



          be equal to or less than 20 percent different from the EPA reference



          standard component(s).







Qualitative Characterization







     Due to the complexity brought by the large number of possible PCDD and



PCDF congeners, the requirement for qualitative verification by comparison of



the retention times applies only to the eight 2,3,7,8-substituted PCDD/PCDF





                                      D-77
                                            ^ (s
                                            G£

-------
congeners marked with an asterisk in Table 3 and for which a carbon-labeled




analogue is available.  Two situations need to be considered:








a)   The laboratory Is required to trace back Its unlabeled PCDD/PCDF standards




     to EPA standards.  This is accomplished by adding an appropriate aliquot




     of the EPA l^c-iabeied standard solution (S2) to an aliquot of the labora-




     tory working solution (VI) so that the concentrations are comparable; the




     new mixture is then analyzed by HRGC/HRMS.  The retention times of the




     eight unlabeled PCDDs/PCDFs discussed above must fall within -1 to +3




     seconds of the EPA 13c_iat,eie(j analogues.
b)   In addition to a), the laboratory is required to trace back its




     labeled standards to EPA standards.   Proceed as follows:  Add an aliquot




     of the laboratory working standard solution (W2) containing the carbon-




     labeled compounds to an aliquot of the EPA standard solution (SI) containing




     the eight unlabeled 2,3,7,8-substituted PCDD/PCDF congeners discussed




     above, and analyze by HRGC/HRMS.  The concentrations must be comparable.




     The retention times for the eight carbon-labeled compounds must fall




     within -3 to +1 seconds of the EPA unlabeled analogues.








Quantitative Characterization








     To establish that the concentration of the laboratory working standard is




correct with respect to the EPA reference standard, the relative response




factors (RRFs) for the eight 2,3,7,8-substituted PCDD/PCDF congeners (marked




with asterisks In Table 3) must be determined as described in this exhibit.
                                      D-78
                                             326

-------
The concentrations of the EPA reference and laboratory working  standards should

be approximately the same (e.g., 50 pg/uL/congener).   Proceed as  follows:



1)   Mix equal portions of the two EPA standard  solutions  (SI and S2) and

     analyze by HRGC/HRMS.  Calculate two RRFs for each of  the  eight analytes

     as shown below:



     Response factor of unlabeled congener (i) relative to  carbon-labeled

     analogue (j):
                                            AI  x
                             RRF (Sl,i)   =
                                            Qi  x AJ
     Response factor of carbon-labeled congener  (j)  relative  to unlabeled

     analogue (i):
                                            AJ  x
                             RRF (S2,j)
                                            Qj
     where A^ and AJ represent the integrated  ion  abundances  of,  respectively,

     the unlabeled congener and carbon-labeled congener,  and  Q^ and Q-j  the

     quantities of,  respectively,  the  unlabeled congener  and  carbon-labeled

     congener,  with  i = 1 to 8,  j = 1 to 8.
                                      D-79
                                        327

-------
2)   Add an appropriate aliquot  of  the  laboratory  working solution Wl  (or W2)

     to an aliquot of  the  EPA solution  S2  (or  SI).   Analyze the  mixture  by

     HRGC/HRMS and calculate the corresponding response factors  as indicated

     below:



                                            Ai x Q-j
                             RRF (Wl,i) =  	
                                            Qi x Aj

               or

                                            AJ x Qi
                             RRF (W2,j) =  	
                                            Qj x Ai



     A and Q have the  same meanings as  in  (1).



3)   When the percent  difference between each  congener relative  response factor

     — RRF (Sl,i) and RRF (Wl,i),  and  RRF (S2,j)  and RRF (W2,j) — does not

     exceed 20 percent, the concentration  of the laboratory working standard is

     correct.  (RPD =  relative percent  difference.)
                            |  RRF (Sl,i)  - RRF (Wl,i)  |
                    RPD  =  •                            x  100
                                    RRF (SI,  i)
               and
                            |  RRF (S2,j)  - RRF(W2,j) |
                    RPD  =                           •   x  100
                                    RRF (S2,j)
                                      D-80
                                            0
8

-------
Traceability Requirements








     If any or all of the above conditions for qualitative and quantitative




verifications for the laboratory working standard are not met, the standard is




not traceable to the EPA reference standard and can therefore not be used for




the analysis of samples.








     NOTE:  The procedure outlined above is required for laboratories which use




     different batches of analytical standard compounds in the preparation of




     the sample fortification and recovery standard solutions and in the prepara-




     tion of the HRCC solutions.  Laboratories which use the same batch of




     analytical standards during the preparation of the sample fortification




     and recovery standard solutions and the HRCC solutions are exempt from




     following the above procedure, provided proper traceability documentation




     is available.








     In addition, the records pertaining to the above qualitative and




quantitative requirements, records of all verifications, documentation of the




preparation, and all inventory must be kept for all contract laboratory pri-




mary, secondary, and working standards that are generated for the purpose of




analyzing samples for EPA.  These records should include the signed and dated




logbooks containing the information pertaining to the preparation of the




laboratory standards (weight of compound(s), volume and nature of the solvent,




laboratory code name, EPA reference standard lot number) and of any modification




made to the EPA reference standard.  All standards should be used on a first




in, first out basis.  The raw data, quantification reports and calculations




must be kept on file.




                                      D-81
                                          329

-------
            APPENDIX C
Signal-to-Nolse Ratio Determination
               D-82
                 330

-------
                      SIGNAL-TO-NOISE RATIO DETERMINATION








MANUAL DETERMINATION








     This method describes a manual determination of the signal-to-noise  ratio




(S/N) from a GC/MS signal, based on the measurement of its peak height  relative




to the baseline noise.  The procedure is composed of four steps as  outlined




below.  (Refer to Figure 7 for the following discussion.)








1.   Estimate the peak-to-peak noise (N) by tracing the two lines (El and E2)




     defining the noise envelope.  The lines should pass through the estimated




     statistical mean of the positive and the negative peak excursions  as shown




     on Figure 7.  In addition, the signal offset (0) should be set high  enough




     such that negative-going noise (except for spurious negative spikes) is




     recorded.








2.   Draw the line (C) corresponding to the mean noise between the  segments




     defining the noise envelope.








3.   Measure the height of the GC/MS signal (S) at the apex of the  peak relative




     to the mean noise C.  For noisy GC/MS signals, the average peak height




     should be measured from the estimated mean apex signal D between E3  and




     E4.
                                      D-83
                                          331

-------
4.   Compute the S/N.








     This method of  S/N measurement is a conventional, accepted method of noise




     measurement in  analytical chemistry.
                                     D-84





                                        09
33

-------
                                FIGURES CAPTIONS








1.   Method flow chart for sample extraction and cleanup as used  for the




     analysis of PCDDs and PCDFs In complex waste and biological  samples.








2.   General structures of dibenzodioxin and dibenzofuran.








3.   Peak profile displays demonstrating the effect of the  detector zero  on  the




     measured resolving power.  In this example, the true resolving power is




     5,600.








     A)  The zero was set too high; no effect is observed upon the measurement




     of the resolving power.  (Not aesthetic.)








     B)  The zero was adjusted properly.








     C)  The zero was set too low; this results in overestimating the actual




     resolving power because the peak-to-peak noise cannot  be measured




     accurately.








4.   Typical 12-hour analysis sequence of events.








5.   Selected ion current profile for m/z 322 (TCDDs) produced by MS analysis




     of the GC performance check solution on a 60-m DB-5 fused-silica capillary




     column under the conditions listed in Table 7.
                                      D-85




                                       333

-------
                              I  DRAFT
—     '   ~^Z^_-
              ){
6.    Peak profiles representing two "ffTC'TtifUluuLe -Iwnd at m/z 305 and 381.   The

     resolution of the high-mass signal is 95 ppm  at  5 percent of the peak

     height; this corresponds  to a resolving power M/AM of 10,500 (10 percent

     valley definition).



7.    Manual determination of S/N.

     The peak height (S)  is measured between the mean noise (lines C and D).

     These mean signal values  are obtained by tracing the line between the

     baseline average noise extremes, El and E2, and  between the apex average

     noise extremes, E3 and E4, at the apex of the signal.  Note, it is

     imperative that the instrument interface amplifier electronic zero offset

     be set high enough such that negative-going baseline noise is recorded.
                                    D-86
                                       33-1

-------
           Complex
            Waste
            Sample
  Soil/
Sediment
  oistur
Fish and
Adipose
Tissues
                       1) Internal
                         Standards
                       2)Extraction
                     Sample  Extract
                1) Acid-Base  Cleanup
                2)Chromatographic Cleanup
                3) Recovery Standards
                       HRGC/HRMS
                         Figure 1

                           D-87
                            335

-------
8
                0
          Dibenzodioxin
8
                0
          Diben zof ura n
            Figure 2




              D-88
             33G

-------
                         M/AM
                         5,600
B
                         5,600
       Moo
                         8,550
          Figure 3




           D-89
             .3'

-------
                                 Analytical Procedure
                8:00 AM
       I
       vo
      rO
CO
Oc
Mass Resolution
 Mass Accuracy
                                    Thaw Sample Extract
                                            1
                                    Concentrate to 10 uL
                        9:00 AM
 initial or
 Routine
Calibration
                           GC Column
                           Performance
              11:00 AM
                               Method
                               Blank
        8:00 PIVi

          Mass
        Resolution
'Routi
                           ne
                                      Figure A

-------
           o
CO
CO
                                                                                                          *ta±soAv/A~*~»*-W
                                                                Figure 5

-------
O
         o
         i
         VO
                                                Ref. mass 304.9824
                                                Span. 200 ppm
                                                System file name
                                                Data file name
                                                Resolution
                                                Group number
                                                lonization mode
                                                Switching
                                                Ref. masses
Peak top

 YVES150
A:85Z567
   10000
        1
      EI+
VOLTAGE
304.9824
380.9260
                                                     M/M A M~10.500
                                                Channel B 380.9260 Lock mass
                                                Span 200 ppm
                                          Figure 6

-------
Go
        o
        vo
100-n

 90-

 80-

 70-

 60-

 50-

 40-

 30-
               20-
               10-
                20:00
                                      117
N
                                          = 19.5
                                                                                    D
              22:00
       24:00
26:00
28:00
30:00
                                            Figure 7

-------
Table 1. Types of Mat
Method Calibi
Soil Fly
Sediment Ash
' DRAFT 1
B^^"-'Sgrr^ ^iT? anflf?, ^T^-Trnn-RaaoH
ration Limits" (ParL-s---p«t Trillion)
Fish
Tissue
Sludges Still- Paper
Water Fuel Oil Bottom Pulp
Lower MCL 2.5 2.5 0.025 12.5 25 2.5
Upper MCL 200 200
Weight (g) 10 10
2 1000 2000 200
1000 2 1 10
Human
Adipose
Tissue
2.
200
10
5


IS Spiking
Levels (ppt)    100       100

Final Extr.
Vol. (uL)        10        50
                                   10
                                            500       1000    100       100
50         50     10        10
(a>For other congeners multiply the values by 1  for TCDF/PeCDD/
PeCDF, by 2.5 for HxCDD/HxCDF/HpCDD/HpCDF, and by 5 for OCDD/OCDF.

NOTE:   Chemical reactor residues are treated as  still-bottoms if
their appearances suggest so.
                                      D-94
                                         342

-------
       Table 2.  Composition of the Sample Fortification
                 and Recovery Standard Solutions
Analyte
                      Sample Fortification
                            Solution
                         Concentration
                       (pg/uL; Solvent:
                          Isooctane)
Recovery Standard
    Solution
  Concentration
 (pg/uL; Solvent:
    Tridecane)
13C12-2,3,7,8-TCDD
J3C12-2,3,7,8-TCDF
1JC12-1,2,3,4-TCDD
                           10
                           10
                                                      50
1JC12-l,2,3,7,8-PeCDD
13C12-l,2,3,7,8-PeCDF
13C,2-l,2,3,6,7,8-HxCDD
}3C12-l,2,3,4,7,8-HxCDF
1JC12-l,2,3,7,8,9-HxCDD
13C12-l,2,3,4,6,7,8-HpCDD
13C12-1,2,3,4,6,7,8-HPCDF
13C12-OCDD
10
10
25
25
25
25
50
                                                      50
                              D-95
                                   o /? '1

-------
  Table 3.   The Fifteen 2,3 ,7 ,8-Substltuted PCDD and  PCDF  Congeners


             PCDD                                  PCDF


    2,3,7,8-TCDD(*>                        2,3,7,8-TCDF(*>

    l,2,3,7,8-PeCDD(*>                     1,2, 3,7,8-PeCDF<*)

    l,2,3,6,7,8-HxCDD(*)                   2, 3,4,7,8-PeCDF

    1,2,3,4,7,8 -HxCDD                     1 , 2 , 3 , 6 , 7 , 8 -HxCDF

    1,2,3,7,8, 9-HxCDD ( +)                   1,2,3,7,8, 9-HxCDF

    l,2,3,4,6,7,8-HpCDD(*>                 l,2,3,4,7,8-HxCDF(*>

                                          2,3,4,6,7,8-HxCDF

                                          l,2,3,4,6,7,8-HpCDF<*)

                                          1,2,3,4,7,8,9-HpCDF
(*)xhe l^c-iabeled analogue is  used  as  an  internal  standard.
       13c_].abej.e(j analogue is  used  as  a recovery standard.
                                 D-96
                                   344

-------
     Table 4.  Isomers of Chlorinated Dioxlns  and  Furans  as  a
               Function of the Number of Chlorine  Atoms
Number of
Chlorine
 Atoms
Number of
 Dioxin
Isomers
Number of
 2,3,7,8
 Isomers
Number of
 Furan
Isomers
Number of
 2,3,7,8
 Isomers
   1

   2

   3

   4

   5

   6

   7

   8
   2

  10

  14

  22

  14

  10

   2

   1
    1

    1

    3

    1

    1
   4

  16

  28

  38

  28

  16

   4

   1
    1

    2

    4

    2

    1
Total
  75
               135
                 10
                                D-97
                                   45

-------
         Table  5.  High-Resolution Concentration Calibration Solutions
Concentration (pg/uL)
Compound HRCC
Unlabeled Analytes
2,3,7,8-TCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDD 1 ,
OCDF ' 1,
Internal Standards
J3c12-2,3,7,8-TCDD
};?C12-2,3,7,8-TCDF
|3C12-l,2,3,7,8-PeCDD
|3c12-l,2,3,7,8-PeCDF
1JC12-l,2,3,6,7,8-HxCDD
|3Cl2-l,2,3,4,7,8-HxCDF
;3Cl2-l,2,3,4,6,7,8-HpCDD
;3Cl2-l,2,3,4,6,7,8-HpCDF
C12-OCDD
Recovery Standards
13C12-l,2,3,4-TCDD(a)
13C12-1,2.3,7,8,9-
HxCDD>
7

200
200
200
200
200
500
500
500
500
500
500
500
500
500
500
000
000

50
50
50
50
125
125
125
125
250

50
125
6

100
100
100
100
100
250
250
250
250
250
250
250
250
250
250
500
500

50
50
50
50
125
125
125
125
250

50
125
5

50
50
50
50
50
125
125
125
125
125
125
125
125
125
125
250
250

50
50
50
50
125
125
125
125
250

50
125
4

25
25
25
25
25
62.5
62.5
62.5
62.5
62.5
62.5
62.5
62.5
62.5
62.5
125
125

50
50
50
50
125
125
125
125
250

50
125
3

10
10
10
10
10
25
25
25
25
25
25
25
25
25
25
50
50

50
50
50
50
125
125
125
125
250

50
125
2

5
5
5
5
5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
25
25

50
50
50
50
125
125
125
125
250

50
125
1

2.5
2.5
2.5
2.5
2.5
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
12.5
12.5

50
50
50
50
125
125
125
125
250

50
125
(a)tJsed  for  recovery determinations of TCDD, TCDF, PeCDD and PeCDF
   internal  standards.
(b)llsed  for  recovery determinations of HxCDD, HxCDF, HpCDD, HpCDF,
   and OCDD  internal standards.
                                     D-98
                                       346

-------
Table 6.  Ions Monitored for HRGC/HRMS analysis of PCDD/PCDFs

          ( S = Internal/recovery standard)
Descriptor Accurate(a)
Mass
1 303.9016
305.8987
315.9419
317.9389
319.8965
321.8936
331.9368
333.9339
375.8364
[354.9792]
2 339.8597
341.8567
351.9000
353.8970
355.8546
357.8516
367.8949
369.8919
409.7974
[354.9792]
Ion
ID
M
M+2
M
M+2
M
M+2
M
M+2
M+2
LOCK
M+2
M+4
M+2
M+4
M+2
M+4
M+2
M+4
M+2
LOCK
Elemental
Composition
C12H435C140
C12H435C1337C10
13C12H435C140
13C12H435C1337C10
C12H435cl4°2
C12H435C1337C102
13C12H435C14°2
13C12H435C1337C102
C12H435C160
C9F13
C12H335C1437C10
C12H335C1337C120
13C12H335C1437C10
13r H 35r1 37ri n
o i ^rio v> J. o \s*.*t\J
C12H335C1437C102
C12H335C1337C1202
13C12H335C1437C102
13C12H335C1337C1202
C12H335cl7°
CqF13
Analyte
TCDF
TCDF
TCDF (S)
TCDF (S)
TCDD
TCDD
TCDD (S)
TCDD (S)
HxCDPE
PFK
PeCDF
PeCDF
PeCDF (S)
PeCDF (S)
PeCDD
PeCDD
PeCDD (S)
PeCDD (S)
HpCDPE
PFK
                                                       (Continued)
                             D-99
O
                                   A '-)
                                   'i i

-------
Table 6.  Continued
Descriptor Accurate
Mass
3 373.8208
375.8178
383.8642
385.8610
389.8156
391.8127
401.8559
403.8529
445.7555
[354.9792]
4 407.7818
409.7789
417.8253
419.8220
423.7766
425.7737
435.8169
437.8140
479.7165
[430.9728]
Ion
ID
M+2
M+4
M
M+2
M+2
M+4
M+2
M+4
M+4
LOCK
M+2
M+4
M
M+2
M+2
M+4
M+2
M+4
M+4
LOCK
Elemental
Composition
C12H235C1537C10
C12H235C1437C120
13C12H235C160
13C12H235C1537C10
C12H235C1537C102
C12H235C1437C1202
13C12H235C1537C102
13C H 35C1 37C1 0
C12H235C1637C120
C9F13
C12H35C1637C10
C12H35C1537C120
13C12H35C170
13C12H35C1637C10
C12H35C1637C102
C12H35C1537C1202
13C12H35C1637C102
13C12H35C1537C1202
C12H35C1737C120
C9F17
Analyte
HxCDF
HxCDF
HxCDF (S)
HxCDF (S)
HxCDD
HxCDD
HxCDD (S)
HxCDD (S)
OCDPE
PFK
HpCDF
HpCDF
HpCDF (S)
HpCDF (S)
HpCDD
HpCDD
HpCDD (S)
HpCDD (S)
NCDPE
PFK
                              (Continued)
      D-100
          348

-------
                         Table 6.  Continued
   Descriptor  Accurate     Ion
                Mass         ID
Elemental
Composition
Analyte
5 441.7428
443.7399
457.7377
459.7348
469.7779
471.7750
513.6775
[430.9728 ]
M+2
M+4
M+2
M+4
M+2
M+4
M+4
LOCK
c1235ci737cio
C1235ci637ci2o
C1235ci737cio2
C1235ci637ci2o2
13c1235ci737cio2
13c1235ci637ci2o2
C1235ci837ci2o
C9Fi7
OCDF
OCDF
OCDD
OCDD
OCDD (S)
OCDD (S)
DCDPE
PFK
(a)The following nuclidic masses were used:

       H = 1.007825            0 = 15'. 994915

       C - 12.000000        35Cl = 34.968853

     13C = 13.003355        37C1 = 36.965903
                               D-101
                                 349

-------
            Table  7.  Recommended GC Operating Conditions
Column coating              DB-5
Film thickness              0.25 urn
Column dimension            60 m x 0.32 mm
Injector temperature         270° C
Splitless valve time         45 s
Interface temperature        Function of the final temperature
                      Temperature program
Stage
1
2
3

In it. Temp. Inlt. Hold. Temp.
(° C) Time (min) Ramp
(° C/min)
200 2 5
5
5
Total
Fin. Temp
(° c)
220
235
330
time: 60
. Fin.
Hoi.
Time
16
7
5
min
                               D-102
                                 350

-------
        Table 8.  PCDD and PCDF Congeners  Present  In  the GC  Performance
                  Evaluation Solution and  Used  for Defining  the
                  Homologous GC Retention  Time  Windows  on a
                  60-m DB-5 Column
No. of
Chlorine
Atoms
4<*>
5
PCDD-Positional Isomer
Early Late
Eluter Eluter
1,3,6,8 1,2,8,9
1, 2,4,6,8/ 1,2,3,8,9
PCDF-Positional
Early
Eluter
1,3,6,8
1,3,4,6,8
Isomer
Late
Eluter
1,2,8
1,2,3,8


,9
,9
        6

        7

        8
    1,2,4,7,9

  1,2,3,4,6,8    1,2,3,4,6,7

1,2,3,4,6,7,8  1,2,3,4,6,7,9

             1,2,3,4,6,7,8,9
  1,2,3,4,6,8     1,2,3,4,8,9

1,2,3,4,6,7,8   1,2,3,4,6,7,9

              1,2,3,4,6,7,8,9
(a)in addition to these two PCDD isomers,  the  1,2,3,4-,  1,2,3,7-,
   1,2,3,8-,  2,3,7,8-,  13C12-2,3,7,8-,  and 1,2,3,9-TCDD  isomers
   must also  be present.
                                     D-103
                                         351

-------
         Table 9.   Theoretical Ion Abundance Ratios and Their
                   Control Limits  for PCDDs  and PCDFs
Number of
Chlorine
Atoms
4


5


6


6(a)


7Used only for 13C-HxCDF (IS).
(b>Used only for 13C-HpCDF (IS).
                                D-104

-------
Table 10.  Relative Response Factor [RRF (number)] Attributions
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

21
22

23
24
25
26
27
28
29
30

2,3,7
2,3,7
1,2,3
1,2,3
2,3,4
1,2,3
1,2,3
1,2,3
1,2,3
1,2,3
1,2,3
2,3,4
1,2,3
1,2,3
1,2,3
OCDD
OCDF
l^Cj-
J7C12
*-*r
W 1 rt
it 12
UP"
17 12
Up**
** 1 O
17 1 ^
l^r1
11 12
•Up
17 12
UP**
17 12
Up"
^11
Totl?
Total
Total
Total
Specific Congener Name
,8-TCDD (and total TCDDs)
,8-TCDF (and total TCDFs)
,7,8-PeCDD (and total PeCDDs)
,7,8-PeCDF
,7,8-PeCDF
,4,7,8-HxCDD
,6,7,8-HxCDD
,7,8,9-HxCDD
,4,7,8-HxCDF
,6,7,8-HxCDF
,7,8,9-HxCDF
,6,7,8-HxCDF
,4,6,7,8-HpCDD (and total HpCDDs)
,4,6,7,8-HpCDF
,4,7,8,9-HpCDF


-2,3,7,8-TCDD
-2,3,7,8-TCDF
-1,2,3,7,8-PeCDD

-1,2,3,7,8-PeCDF
-1,2,3,6,7,8-HxCDD

-1,2,3,4,7,8-HxCDF
-1,2,3,4,6,7,8-HpCDD
-1,2,3,4,6,7,8-HpCDF
-OCDD
PeCDFs
HxCDFs
HxCDDs
HpCDFs
                             D-105
                               i O O

-------
      TABLE 11.  2,3,7,8-TCDD Equivalent Factors (TEFs) for the
                 Polychlorinated Dlbenzodioxins and Dibenzofurans
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Compound (s)
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,6,7,8-HxCDD
1, 2,3,7,8,9-HxCDD
1, 2,3,4,7,8-HxCDD
1, 2,3,4,6,7,8-HpCDD
* Total - TCDD
* Total - PeCDD
. * Total - HxCDD
* Total - HpCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,7,8-HpCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
* Total - TCDF
* Total - PeCDF
* Total - HxCDF
* Total - HpCDF
TEF
1.00
0.50
0.04
0.04
0.04
0.001
0.01
0.005
0.0004
0.00001
0.10
0.10
0.10
0.01
0.01
0.01
0.01
0.001
0.001
0.001
0.001
0.0001
0.00001
*Excluding the 2,3,7,8-substituted congeners,
                                D-106
                                     Q cr ,1
                                     OU'i

-------
  Table 12. Toxiclty Equivalency Factor:   Analyte  Relative
            Retention Time Reference Attributions
      Analyte                 Analyte RRT Reference(a


 1,2,3,4,7,8-HxCDD            13C12-l,2,3,6,7,8-HxCDD

 1,2,3,6,7,8-HxCDF            13C12~1,2,3,4,7,8-HxCDF

 1,2,3,7,8,9-HxCDF            13C12-l,2,3,4,7,8-HxCDF

 2,3,4,6,7,8-HxCDF	13C12-l,2,3,4,7,8-HxCDF
The retention time of 2,3,4,7,8-PeCDF on the DB-5  column  is
measured relative to *3Cj2-l,3,7,8-PeCDF and the retention
time of 1,2,3,4,7,8,9-HpCDF relative to 13C12-1,2,3,4,6,7,8
HpCDF.
                           D-107
                               355

-------
     QUALITY ASSURANCE REQUIREMENTS
(Quality Assessment and Quality Control)
              (Exhibit E)
                  n r-11
                  0Gb

-------
                                      DRAFT
1.    SUMMARY  OF QA/QC ANALYSES            •**—»•  I
     o   Initial and periodic calibration and instrument performance checks.








     °   HRGC/HRMS method blank analysis.








     0   Field blank analyses (Section 2.4.2, this exhibit);  a minimum of one




        fortified field blank shall be analyzed with each sample batch; an




        additional fortified field blank must be analyzed when a new lot of




        absorbent or solvent is used.  A matrix spike may be used  in place of




        a  fortified field blank.








     °   Analysis of a batch of samples with accompanying QA/QC analyses:








        Sample Batch — < 24 samples,  including field blank  and rlnsate




        sample(s).








        Additional QA/QC analyses  per batch:








        Fortified field blank or matrix spike (MS)                  1




        Method blank (MB)                                          1




        Duplicate sample or matrix spike duplicate (MSD)            1




                                                          Total    3








     0   "Blind" QC samples (soil,  sediment, water) may be submitted to the




        laboratory as ordinary samples included in the sample batch.






                                     E-l

-------
        Blind samples include:







        Uncontaminated soil,  sediment,  or water samples



        Split samples,



        Unidentified duplicates,  and



        Performance evaluation  samples.







2.  QUALITY ASSESSMENT/QUALITY  CONTROL







2.1  Performance Evaluation Samples  --  Included among the samples  in all



     batches will be samples  (blind  or  double blind)  containing known amounts



     of unlabeled 2,3,7,8-substituted PCDDs/PCDFs  or  other PCDD/PCDF congeners.







2.2  Performance Check Solutions







2.2.1  At the beginning of each 12-hour period during which samples are to be



       analyzed, an aliquot of  the 1) GC column performance check  solution and



       2) high-resolution concentration calibration solution  No.  3 (HRCC-3)



       shall be analyzed to demonstrate adequate GC resolution and sensitivity,



       response factor reproducibility,  and mass range calibration, and to



       establish the PCDD/PCDF  retention time windows.  A mass resolution check



       shall also be performed  to demonstrate adequate mass resolution using an



       appropriate reference  compound (PFK is recommended).







       These procedures are described in Section 8 of Exhibit D.  If the



       required criteria are  not  met, remedial action must be taken before any





                                     E-2
                                       O KT O
                                       ouo

-------
       samples are analyzed.








2.2.2  To validate positive sample data, the routine or continuing calibration




       (HRCC-3) and the mass resolution check must be performed also at the end




       of each 12-hour period during which samples are analyzed.  Furthermore,




       an HRGC/HRMS method blank run must be recorded following a calibration




       run and the first sample run.








2.2.2.1  If the laboratory operates only during one period (shift) each day




         of 12 hours or less, the GC performance check solution must be




         analyzed only once (at the beginning of the period) to validate the




         data acquired during the period.  However, the mass resolution and




         continuing calibration checks must be performed at the beginning as




         well as at the end of the period.








2.2.2.2  If the laboratory operates during consecutive 12-hour periods (shifts),




         analysis of the GC performance check solution must be performed at the




         beginning of each 12-hour period.  The mass resolution and continuing




         calibration checks from the previous period can be used for the




         beginning of the next period.








2.2.3  Results of at least one analysis of the GC column performance check




       solution and of two mass resolution and continuing calibration checks




       must be reported with the sample data collected during a 12-hour period.








2.2.4  Deviations from criteria specified for the GC performance check or for






                                      E-3
                                         9 c r
                                         o u

-------
       the mass resolution check (Section 8,  Exhibit D) invalidate all positive




       sample data collected between analyses of the performance check solu-




       tion, and the extracts from those positive samples shall be reanalyzed




       (Exhibit C).








       If the routine calibration run fails at the beginning of a 12-hour shift,




       the instructions in Exhibit D, Section 9.4.4 must be followed.   If the




       continuing calibration check performed at the end of a 12-hour  period




       fails by no more than 25 percent RPD,  use the mean RRFs from the two




       daily routine calibration runs to compute the analyte concentrations,




       instead of the RRFs obtained from the initial calibration.  A new




       initial calibration (new RRFs) is required immediately (within  two hours)




       following the analysis of the samples, whenever the RPD from the end-




       of-shift routine calibration exceeds 25 percent.  Failure to perform a




       new initial calibration immediately following the analysis of the




       samples will automatically require reanalysis of all positive sample




       extracts analyzed before the failed end-of-shift continuing calibration




       check.








2.3  The GC column performance check mixture, high-resolution concentration




     calibration solutions,  and the sample fortification solutions may be




     obtained from the EMSL-LV.  However, if not available from the EMSL-LV,




     standards can be obtained from other sources, and solutions can be pre-




     pared in the laboratory.  Concentrations of all solutions containing




     2,3,7,8-substituted PCDDs/PCDFs, which are not obtained from the  EMSL-




     LV, must be verified by comparison with the EPA standard solutions that






                                      E-4
                                          ,60

-------
     are available from the EMSL-LV.   (Refer to Appendix B, Exhibit D,  for



     details on the recommended standards traceability procedure.)
2.4  Blanks
2.4.1  Method Blank
       One method blank is required per batch of samples.  To that effect,



       perform all steps detailed in the analytical procedure (Section 12,



       Exhibit D) using all reagents, standards, equipment, apparatus, glass



       ware and solvents that would be used for a sample analysis, but omit



       addition of the soil, aqueous or any other matrix sample portion.
2.4.1.1  The method blank must contain the same amount of x C



         internal standards that is added to samples before extraction.







2.4.1.2  An acceptable method blank exhibits no positive response as stated in



         Section 3.16, Exhibit D.  If the method blank, which was extracted



         along with a batch of samples, is contaminated, all positive samples



         must be rerun (Exhibit C).







2.4.1.2.1  If the above criterion is not met, check solvents, reagents, forti-



           fication solutions, apparatus and glassware to locate and eliminate



           the source of contamination before any further samples are extracted



           and analyzed.







2.4.1.2.2  If new batches of reagents or solvents contain interfering



                                      E-5
                                          o f> \
                                          ool

-------
           contaminants, purify or discard them.








2.4.2  Field Blanks








       Each batch of samples contains a field blank sample of uncontaminated




       soil, sediment or water that is to be fortified before analysis accord-




       ing to Section 2.4.2.1 (this exhibit).  In addition to this field blank,




       a batch of samples may include a rinsate,  which is a portion of the sol-




       vent (usually trichloroethylene) that was  used to rinse sampling equip-




       ment.  The rinsate is analyzed to assure that the samples were not




       contaminated by the sampling equipment.








2.4.2.1  Fortified Field Blank








2.4.2.1.1  Weigh a 10-g portion or use 1 L (for aqueous samples) of the speci-




           fied field blank sample and add 100 uL of the solution containing




           the nine internal standards (Table 2,  Exhibit D) diluted with 1.5 mL




           acetone (Section 12.1,  Exhibit D).








2.4.2.1.2  Extract by using the procedures beginning in Sections 12.2.5 or




           12.2.6 of Exhibit D, as applicable,  add 10 uL of the recovery stan-




           dard solution (Section 12.9.2, Exhibit D) and analyze a 2-uL aliquot




           of the concentrated extract.








2.4.2.1.3  Calculate the concentration (Section 14.1, Exhibit D) of 2,3,7,8-




           substituted PCDDs/PCDFs and the percent recovery of the internal




           standards "(Section 14.2, Exhibit D).  If the percent recovery at the




                                      E-6





                                           o f*'»

-------
           measured concentration of any 2,3,7,8-substltuted PCDD/PCDF congener




           Is <40 percent or >120 percent, report the results to SMO before




           proceeding with the samples.








2.4.2.1.4  Extract and analyze a new simulated fortified field blank whenever




           new lots of solvents or reagents are used for sample extraction or




           for column chromatographic procedures.








2.4.2.2  Rinsate Sample








2.4.2.2.1  The rinsate sample must be fortified like a regular sample.








2.4.2.2.2  Take a 100-mL (+ 0.5 mL) portion of the sampling equipment rinse




           solvent (rinsate sample), filter, if necessary, and add 100 uL of the




           solution containing the nine internal standards (Table 2, Exhibit D).








2.4.2.2.3  Using a Kuderna-Danish appparatus, concentrate to approximately




           5 mL.








2.4.2.2.4  Transfer the 5-mL concentrate from the K-D concentrator tube in 1-mL




           portions to a 1-mL minivial, reducing the volume in the minivial as




           necessary with a gentle stream of dry nitrogen.








2.4.2.2.5  Rinse the K-D concentrator tube with two 0.5-mL portions of hexane




           and transfer the rinses to the 1-mL minivial.  Blow down with dry




           nitrogen as necessary.






                                      E-7
                                        3i

-------
2.4.2.2.6  Just before analysis, add 10 uL tridecane recovery standard solution


           (Table 2,  Exhibit D), and reduce the volume to a final volume of 10


           uL,  or 50  uL,  as necessary (Section 12.9.2, Exhibit D).   No column


           chromatography is required.






2.4.2.2.7  Analyze an aliquot following the same procedures used to analyze


           samples (Section 13,  Exhibit D).






2.4.2.2.8  Report percent recovery of the internal standard and the presence


           of any PCDD/PCDF compounds on Form (to be determined) in pg/mL of


           rinsate solvent.






2.5  Duplicate Analyses






2.5.1  In each batch  of samples, locate the sample specified for duplicate


       analysis,  and  analyze a second 10-g soil or sediment sample portion or


       1-L water sample,  or an appropriate amount of the type of matrix under


       consideration.






2.5.1.1  The results  of the laboratory duplicates (percent recovery and concen-


         trations of  2,3,7,8-substituted PCDD/PCDF compounds) must agree within


         25 percent relative difference (difference expressed as percentage of


         the mean).  If the relative difference is >25 percent for any one of


         the fifteen  2,3,7,8-substituted PCDDs/PCDFs, the laboratory shall


         immediately  contact the Sample Management Office for resolution of the


         problem.  Report all results.




                                      E-8
                                        o n
                                        00

-------
2.5.1.2  Recommended actions to help locate problems:








2.5.1.2.1  Verify satisfactory Instrument performance  (Section 8,  Exhibit  D).








2.5.1.2.2  If possible, verify that no error was made  while weighing  the sample




           portions.








2.5.1.2.3  Review the analytical procedures with the performing laboratory




           personnel.








2.6  Matrix Spike and Matrix Spike Duplicate








2.6.1  Locate the sample for the MS and MSD analyses (the sample may  be  labeled




       "double volume").








2.6.2  Add on appropriate volume of the matrix spike fortification solution




       (Exhibit D, Section 3.24), adjusting the fortification level as specified




       in Exhibit D, Table 1, under IS Spiking Levels.








2.6.3  Analyze the MS and MSD samples as described in  Exhibit D, Section 12.








2.6.4  The results obtained from the MS and MSD samples (percent recovery  and




       concentrations of 2,3,7,8-substituted PCDDs/PCDFs) must agree  within  20




       percent relative difference.








2.7  Percent Recovery of the Internal Standards






                                      E-9
                                         381)

-------
     For each sample,  method blank and rinsate,  calculate  the percent  recovery



     (Section 14.2,  Exhibit  D).   It is recommended  that  the  percent  recovery  be



     >40 percent and <120 percent for  all 2,3,7,8-substituted internal standards.








     NOTE:   A low or high percent recovery for a blank does  not  require dis-



     carding the analytical  data  but it may indicate a potential problem with



     future analytical data.








2.8  Identification Criteria








2.8.1  If either one of the  identification criteria appearing in Sections



       13.4.1.1 through 13.4.1.4, Exhibit D,  is  not met  for  an homologous



       series, it is reported that the sample does  not contain unlabeled




       2,3,7,8-substituted PCDD/PCDF isomers  for that homologous series at



       the  calculated detection limit  (Section 14.5, Exhibit D).








2.8.2  If the first  initial  identification criteria (Sections 13.4.1.1 through



       13.4.1.4) are met, but the criteria appearing in  Sections 13.4.1.5  and



       13.4.2.1, Exhibit D,  are not met,  that sample is  presumed to  contain



       interfering contaminants.   This must be noted on  the  analytical report



       form, and the sample  must  be rerun or the extract reanalyzed.   Detailed



       sample rerun  and extract reanalysis requirements  are  presented  in




       Exhibit C.








2.9  Blind  QA/QC Samples








     Included among  soil, sediment and aqueous samples may be QA/QC  samples



                                      E-10






                                          3GG

-------
     that are not specified as such to the performing laboratory.  Types that




     may be included are:








2.9.1  Uncontaminated soil, sediment, or water.









2.9.1.1  If a false positive is reported for such a sample, the laboratory




         shall be required to rerun the entire associated batch of samples




         (Section to be determined, Exhibit C).








2.9.2  Split samples — composited sample portions sent to more than one




       laboratory.








2.9.3  Unlabeled field duplicates — two portions of a composited sample.








2.9.4  Performance evaluation samples — soil/sediment or water samples con-




       taining a known amount of unlabeled 2,3,7,8-substituted PCDDs/PCDFs




       and/or other PCDD/PCDF compounds.









2.9.4.1  If the performance evaluation sample result falls outside the accept-




         ance windows established by the EPA, the laboratory shall be required




         to rerun the entire associated batch of samples (Exhibit C).








         NOTE:  EPA acceptance windows are based on previously generated data.








2.10  Quality Control Charts








      The performance of the entire measurement system (i.e., from the extraction




                                      E-ll

-------
     of the sample to the mass  spectrometric  determination)  must  be  documented




     by using germane control charts.   The  selection  and  design of a specific




     measurement control chart  must  be  accomplished  in a  rational manner  so




     that the measurement process  can be  adequately  surveyed.  By using the




     standard deviations obtained  from  control  samples or control runs, the




     laboratory must delineate  control  limits,  i.e.,  statistically congruous




     extreme values, which should  warn  the  operator  of possible problems.   It




     is recommended to consider the  values  corresponding  to  two standard  devi-




     ations as warning limits and  the values  from three standard  deviations  as




     control limits (i.e., corrective actions are required).   For some par-




     ticular applications, however,  the control limits must  not exceed the




     limits set forth by the EPA (e.g., ion-abundance ratios).   [Specific and




     required QC charts, such as mass and GC  resolutions, ion  abundance ratios,




     RRF values, etc., will be  described  in the final version  of  this protocol.]








2.11  Standard Operating Procedures  (SOPs)








      As part of the quality assurance  program, the  laboratory must  use in-house




      SOPs describing how the basic  operations  executed within the laboratory




      are done.








2.12  Internal Audits








      Internal audits of records,  Instrumentation performances and calibration




      data are highly encouraged in  order to  identify defects  that could




      compromise the quality of the  results.






                                     E-12
                                         368

-------
2.13  Records








      At each laboratory, records must be maintained on site for six months




      after contract completion to document the quality of all data generated




      during the contract period.  Before any records are disposed, written




      concurrence from the Contracting Officer must be obtained.








2.14  Unused portions of samples and sample extracts must be preserved for six




      months after sample receipt; appropriate samples may be selected by EPA




      personnel for further analyses.








2.15  Reuse of glassware is to be minimized to avoid the risk of contamination.








3.  Laboratory Evaluation Procedures








3.1  On a quarterly basis, the EPA Project Officer or his/her designated repre-




     sentatives may conduct an evaluation of the laboratory to ascertain that




     the laboratory is meeting contract requirements.  This section outlines




     the procedures which may be used by the Project Officer or his/her author-




     ized representative in order to conduct a successful evaluation of




     laboratories conducting dioxin analyses according to this protocol.  The




     evaluation process consists of the following steps:  1) analysis of a




     performance evaluation (PE) sample, and 2) on-site evaluation of the




     laboratory to verify continuity of personnel, instrumentation, and quality




     assurance/quality control functions.  The following is a description of




     these two steps.






                                      E-13
                                          36i)

-------
3.2  Performance Evaluation (PE)  Sample Analysis








3.2.1  The PE sample set  will  be  sent  to  a participating  laboratory  to verify




       the laboratory's  continuing  ability to  produce  acceptable  analytical




       results.   The PE  sample will be representative  of  the  types of samples




       that will be analyzed under  this contract.








3.2.2  When the  PE sample results are  received, they are  scored using the PE




       Sample Score Sheet shown in  Figure (to  be  determined).  If a  false




       positive  (e.g., a  PE sample  not containing 2,3,7,8-TCDD or other  PCDD/




       PCDF but  reported  by the laboratory to  contain  it  or them) is reported,




       the laboratory has failed  the PE analysis  requirement.  The Project




       Officer will notify the laboratory immediately  if  such an  event occurs.








3.2.3  As a general rule, a laboratory should  achieve  75  percent  or  more of  the




       total possible points for  all three categories  listed  on the  PE Sample




       Score Sheet, and  75 percent  or  more of  the maximum possible points in




       each category, to  be considered acceptable for  this program.  However,




       the Government reserves the  right  to  accept scores of  less than 75




       percent.








3.2.4  If unanticipated  difficulties with the  PE  samples  are  encountered, the




       total points may be adjusted by the Government  evaluator in an impartial




       and equitable manner for all participating laboratories.








3.3  On-site Laboratory Evaluation






                                      E-14
                                             370

-------
3.3.1  An on-site laboratory evaluation is performed to verify that (1) the

       laboratory is maintaining the. necessary minimum level in instrumentation

       and levels of experience in personnel committed to the contract and (2)
                                         i
       that the necessary quality assurance activities are being carried out.

       It also serves as a mechanism for discussing laboratory weaknesses

       identified through routine data audits, PE sample analyses results, and

       prior on-site evaluations.  Photographs may be taken during the on-site

       laboratory evaluation tour.




3.3.2  The sequence of events for the on-site evaluations is shown in Figure

       (to be determined).  A Site Evaluation Sheet (SES) is used to document

       the results of the evaluation.
                                         E-15


                                            i-> r~t -I
                                            O ( 1

-------
      METHOD	:   MEASUREMENT OF 2.3.7,8-TETRACHLORINATED
      DIBENZO-P-DIOXIN (TCDD)  AND 2,3.7.8-TETRACHLORINATED
      DIBENZOFURAN  (TCDF)  IN PULP.  SLUDGES.  PROCESS  SAMPLES
            AND WASTEWATERS  FROM PULP AND PAPER MILLS
1.   SCOPE AND APPLICATION

     1.1  This method is  appropriate for  the determination of
          2,3,7,8-tetrachlorinated dibenzo-p-dioxins  (TCDD)   and
          2 , 3 , 7,8-dibenzofurans  (TCDF)  in paper mill  process
          samples,  including  paper pulp, sludge, ash,  mud,
          woodchips,  and treated and  untreated wastewaters.

                                Chemical Abstracts Service      STORET
	Analyte	Registry Number (CASRN)	Number

2,3,7,8-Tetrachlorodibenzofuran    •        51207-31-9
2,3,7,8-Tetrachlorodibenzo-p-dioxin          1746-01-6             3475

     1.2  The sensitivity of  this method  is  dependent upon  the
          level of interferences  within  a  given matrix.   Target
          quantification  levels for the analytes  are 1 ppt  in
          solid samples  and  10  ppq in  water  and  wastewater
          samples.

     1.3  Certain   *• 3 Ct 2 -labelled  and  3 7 Cl« -2 , 3 ,7 , 8-substituted
          congeners  are used  to  provide calibration and  method
          recovery information.   Appropriate capillary GC columns
          and reference  isomer  standards  are used  to  achieve
          isomer  specific data.   Other isotopically  labelled
          congeners  are  also used to  refine method recovery data.

     1.4  This method is  recommended for  use only  by analysts
          experienced with residue analysis and skilled  in  mass
          spectral analytical techniques.


2.   SUMMARY OF THE  METHOD

     2.1  This procedure uses  a  matrix-specific  extraction,
          analyte-specific cleanup,  and high-resolution capillary
          column  gas chromatography/ mass  spectrometry  (HRGC/MS)
          techniques.

     2.2  If  interferences  are  encountered, the method  provides
          selective cleanup procedures to aid the  analyst  in
          their elimination.   The analysis  flow chart is  shown in
          Figure  1.


3.   DEFINITIONS

     Internal standard:    a pure  compound  added  to  a sample  in
     known  amounts  prior to sample preparation  and  used  to
                                 O f-t
                                 O (

-------
      Paper/Pulp
        Process
      Sample(s)
                 (1)

                 (2)
Add Internal Standards:  *3 Ci2-TCDD
and laCitTCDF
Perform matrix-specific extraction.
      Sample
      Extract
                 (1)
                 (2)
                 (3)
                 (4)
                 (5)
                 (6)
                 (7)
                 (8)
                 (9)
                (10)
Wash with 40% KOH
Wash with doubly distilled water
Wash with cone. Ha SO«
Repeat wash with Cone. H2SO«
Wash with doubly distilled water
Dry extract
Solvent Cone.
Silica gel column
Solvent Cone.
Alumina column
      50% CH2 C12/hexane
          Fraction
                 (1)  Concentrate eluate
                 (2)  Repeat alumina column
      50% GHz C12/hexane
          Fraction
                 (1)  Concentrate eluate
                 (2)  Perform carbon column cleanup
                 (3)  Repeat alumina column
      50% CH2 C12/hexane
          Fraction
                 (1)  Concentrate eluate
                 (2)  Add Total Recovery  Standard(s)
      Analyze by GC/MS
 Figure 1.  Flow chart for sample extraction  and cleanup
as used for the analysis of  2,3,7,8-TCDD  and  2,3,7,8-TCDF
             in paper/pulp process samples.
                                6

-------
     calculate  concentration  of the analyte that  is   a  sample
     component.   The internal  standard  is used to  indicate
     recovery of the  analyte in the method.

     External standard:   a known amount of a pure  compound that
     is added to the sample  extract prior  to analysis.   From
     measured instrument  response  for a known  amount  of the
     external standard,  the recovery  of  the  internal standard is
     calculated.

     Surrogate  standard:   a compound that is not  expected  to be
     found in the sample,  is  added to  the  environmental sample
     prior to sample preparation to monitor performance,  and is
     measured with the same procedures used to  measure  sample
     components.   A final calculation ratio  to  the  internal
     standard  indicates  any interfering responses  in the region
     of the analyte.

     Method blank:   a quality assurance sample that is prepared
     through sample  preparation,  but without any  environmental
     sample.

     Native spike:   a small volume of  a solution containing the
     analytes that is added to a sample and/or blank solution and
     is analyzed with the procedure used for an environmental
     sample.   Results  of analyses  are  used  to determine
     statistically  the accuracy  and precision that  can be
     expected.

     Laboratory duplicate:  two  aliquots of  the same sample that
     are treated in  exactly the same manner throughout  laboratory
     analytical  procedures.    Analysis  of laboratory  duplicates
     indicate precision associated with laboratory procedures but
     not  with  sample collection, preservation or storage
     procedures.
4.    INTERFERENCES

     4.1   Residues  or contaminants  present  in solvents, or
          reagents,  or on glassware and  other  sample  processing
          equipment may yield  elevated  baselines  in the mass
          chroroatograms which may  cause  misinterpretation  of  GC-
          MS  data.   All of  the materials and  apparatus used in
          the analysis must be demonstrated  to be  free from
          interferences under  the  conditions  of  analysis by
          running laboratory method blanks.

     4.2   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.

     4.3   Interferences co-extracted  from the  sample may vary
          considerably  from  sample  to  sample, depending  upon  the
                 «


                          3     374

-------
          process from  which  the samples originated.   The
          2,3,7.8-TCDD  and 2,3,7,8-TCDF isomers are  often
          associated with other  interfering chlorinated  compounds
          which  may be present  in the sample  at  concentrations
          several orders  of magnitude  higher  than those of the
          analytes of interest.  Retention  times of target
          analytes must  be  verified  using  reference standards.
          These  values must correspond  to the retention time
          windows  established   as   described  in   Section
          9.   While  certain cleanup  techniques are provided  as
          part  of this  method,  unique samples may  require
          additional cleanup techniques in order  to achieve the
          stated method detection  limit.

     4.4  High  resolution capillary  gas  chromatography  (GO
          columns are used to  provide  optimum resolution of TCDD
          and TCDF isomers as possible.


5.   SAFETY AND  HANDLING  PROCEDURES FOR 2.3.7.8-TCDD AND 2.3.7.8-
     TCDF

     5.1  The 2,3,7,8-TCDD isomer  has been found to be acnegenic,
          carcinogenic,  and  teratogenic in the  course   of
          laboratory animal  studies.  The 2,3,7,8-TCDD is a  solid
          ar  room temperature,  and has a relatively  low vapor
          pressure.   The  solubility of  this compound in water  is
          only about 200  parts-per-trillion, but  the solubility
          in  various organic solvents  ranges  from about 0.001
          percent to 0.14 percent.   The physical  properties  of
          2,3,7,8-TCDF  have  not been well  established,  although
          it  is  presumed  that  the physical properties   of  these
          congeners are generally similar  to  those of the
          2,3,7,8-TCDD isomer.  On the basis  of  the  available
          toxicological and  physical property data for TCDD, this
          compound,  as well  as the 2,3,7,8-TCDF should be handled
          only by highly  trained  personnel who are thoroughly
          versed in  the  appropriate  procedures,   and  who
          understand the  associated risks.

     5.2  The 2,3,7,8-TCDD and  2,3,7,8-TCDF isomers, and samples
          containing these, are handled using essentially the
          same  techniques  as  those  employed  in   handling
          radioactive or  infectious materials.   Well-ventilated,
          controlled-access  laboratories  are  required,  and
          laboratory personnel entering these laboratories  should
          wear appropriate safety  clothing,  including disposable
          coveralls, shoe covers, gloves, and  face  and head
          masks.  During analytical operations which  may give
          rise  to aerosols or dusts, personnel should wear
          respirators equipped  with activated carbon  filters.
          Eye protection  equipment (preferably  full face shields)
          must  be worn  at  all  times  while workin.g in the
          analytical laboratory  with TCDD/TCDF.   Various types  of
          gloves  can be  used by personnel , depending  upon the
                              o t-i i,;
                              O ( O

-------
     analytical  operation  being  accomplished.  Latex gloves
     are generally  utilized, and  when  handling  samples
     thought  to  be particularly hazardous, an additional set
     of  gloves are also worn beneath the  latex gloves  (for
     example,  Playtex gloves supplied by American  Scientific
     Products,  Cat.  No.  67216).   Bench-tops and  other work
     surfaces in  the  laboratory  should  be covered  with
     plastic-backed  absorbent paper  during all  analytical
     processing.  When finely divided samples (dusts, soils,
     dry chemicals)  are processed,  removal of these from
     sample  containers,  as well  as  other  operations,
     including  weighing,  transferring,   and  mixing  with
     solvents, should all  be accomplished  within a glove
     box.  Glove boxes,  hoods and the  effluents  from
     mechanical  vacuum  pumps and gas chromatographs on the
     mass spectrometered should  be vented  to the  atmosphere
     preferably  only after passing  through HEPA particulate
     filters  and vapor-sorbing charcoal.

5.3  All laboratory  ware,  safety clothing,  and other  items
     potentially contaminated with 2,3,7,8-TCDD and 2,3,7,8-
     TCDF in  the  course  of analyses  must be  carefully
     secured and subjected to proper  disposal.   When
     feasible,  liquid  wastes are  concentrated,  and the
     residues are  placed in  approved steel  hazardous  waste
     drums fitted with  heavy gauge  polyethylene liners.
     Glass and  combustible  items  are compacted  using  a
     dedicated trash compactor used only for hazardous  waste
     materials and then  placed in  the same type of disposal
     drum.   Disposal of accumulated  wastes  is  periodically
     accomplished  by high  temperature  incineration at  EPA-
     approved facilities.

5.4  Surfaces of laboratory  benches, apparatus  and other
     appropriate areas  should be periodically subjected  to
     surface  wipe tests  using  solvent-wetted  filter  paper
     which  is  then  analyzed  to   check  for   TCDD/TCDF
     contamination  in the  laboratory.   Typically, if the
     detectable  level  of TCDD or TCDF  from such  a test  is
     greater  than 50 ng/m2 ,  this indicates the need for
     decontamination of  the laboratory.   In the  event of  a
     spill within the laboratory, absorbent paper  is used  to
     wipe up   the spilled  material  and  this is then placed
     into a hazardous waste drum.   The  contaminated surface
     is  subsequently cleaned thoroughly by washing with
     appropriate solvents  (methylene chloride  followed  by
     methanol)  and laboratory detergents.   This is repeated
     until  wipe tests  indicate  that  the  levels  of  surface
     contamination are below  the limits  cited.

5.5  In   the  unlikely  event that  analytical  personnel
     experience skin  contact with  TCDD/TCDF or samples
     containing these, the  contaminated  skin  area should
     immediately be  thoroughly scrubbed using mild  soap and
     water.   Personnel  involved  in  any  such accident  should
                            O P*-I / *
                            o ft)

-------
          subsequently be taken to the nearest medical facility,
          preferably a facility whose staff  is  knowledgeable  in
          the toxicology of  chlorinated hydrocarbons.  Again,
          disposal  of  contaminated  clothing is  accomplished  by
          placing it in hazardous  waste drums.

     5.6  It is desirable that personnel  working in laboratories
          where TCDD/TCDF are handled be  given periodic physical
          examinations  (at  least yearly).   Such  examinations
          should include specialized tests,  such  as those for
          urinary porphyrins and  for certain blood parameters
          which,  based upon published clinical observations, are
          appropriate  for  persons who  may  be  exposed  to
          TCDD/TCDF.   Periodic facial  photographs to document the
          onset of  dermatologic problems are also advisable.


6.   APPARATUS AND  EQUIPMENT   '

     6.1  GAS CHROMATOGRAPH/MASS SPECTROMETER DATA SYSTEM:

          6.1.1   Gas chromatograph:   An  analytical system which
                  incorporates  a  temperature-programmable gas
                  chromatograph  and  associated  accessories,
                  including  syringes, analytical   columns and
                  gases,  is required  for these analyses.

          6.1.2   Fused silica capillary GC  columns  are used.  As
                  shown  in  Tables   1  and  2,   two  different
                  capillary GC columns  are employed and the
                  performance  of  these  is  evaluated  on   a
                  continuing  basis by using  column performance
                  check mixtures containing all  22  TCDD isomers
                  and all 38  TCDF  isomers.

                  The columns used include the following:   (a)  a
                  60-m  DB-5  column operated  at an  initial
                  temperature of 180°C, held for 1 minute at that
                  temperature,  and then programmed  from   180°  to
                  240°C at a  rate  of  2°C/minute, and held at.that
                  temperature for  14  minutes;  (b) a hybrid column
                  consisting  of a  section of DB-5 (10-m x 0.25  mm
                  I.D.;  0.25  urn film thickness)  coupled  to  a
                  section of    DB-225  (30-m  x  0.25  mm I.D.;  0.25
                  urn film thickness),  which  is operated initially
                  at  180°C,   held at that  temperature  for  1
                  minute, then programmed from 180°  to 220°C at a
                  rate of 2°C/minute.   The  column  described  in
                  Section (a)  provides  complete   separation  of
                  2,3,7,8-TCDD from the other 21 TCDDs,  but does
                  not  completely  resolve  2,3,7,8-TCDF  from  all
                  other  TCDFs.   The column  described  in  (b)
                  provides complete  separation  of  2,3,7,8-TCDF
                  from the other 37 TCDFs,  but does not uniquely
                  resolve 2,3,7,8-TCDD.  Other  capillary columns


                           6        377

-------
             which provide separation  of  2,3,7,8-TCDD from
             all of  the other  21. TCDD  isomers  which  is
             equivalent to that  specified  in this protocol
             may also  be  used   for  2,3,7,8-TCDD analysis.
             This  separation must  be demonstrated  and
             documented periodically using the performance
             text  mixture  described in  Section 7.3.5.
             Moreover,  any capillary  column  which provides
             separation of 2,3,7,8-TCDF from all  the  other
             37 TCDF isomers  equivalent to  that specified in
             this protocol   may be used for 2,3,7,8-TCDF
             analysis.   This  separation  must  also  be
             demonstrated   and  documented  at  regular
             intervals  using  the performance  test  mixture
             described  in  Section 7.3.7.

     6.1.3   Mass spectrometer:   An instrument  capable  of
             mass spectral resolution of  at  least 1:6500 is
             desirable.  This instrument should utilize the
             electron impact  ionization mode (70 eV nominal
             electron energy).   The  system must  be capable
             of selected ion  monitoring  (SIM)  for at  least
             11 ions simultaneously,  with  a cycle time of 1
             sec or  less. Minimum integration  time  for SIM
             is 50 ms per  mass peak.  The use of systems not
             capable of  monitoring 13  ions simultaneously
             will  require the analyst to  make multiple
             injections.

     6.1.4   GC/MS interface:   Any  GC-to-MS interface that
             gives an  acceptable calibration  response  for
             each analyte of  interest  at   the  concentration
             required  and achieves  the  required tuning
             performance  criteria (see Section 9) may be
             used.    GC-to-MS  interfaces constructed  of all
             glass or glass-lined materials are required.

     6.1.5   Data  system:    A computer  system  must  be
             interfaced  to  the  mass  spectrometer.   The
             system must   allow   for    the   continuous
             acquisition  and storage  on machine-readable
             media  of  all data  obtained throughout  the
             duration of  the  chromatographic program.   The
             computer must have software that can search any
             GC/MS data file  for ions of a  specific mass and
             can plot  such ion abundances versus time or
             scan number.   This  type of plot is defined as
             an  Selected Ion Current Profile (SICP).
             Software must also be able to integrate  the
             abundance, in any SICP,  between specified time
             or scan number limits.

6.2  LABORATORY APPARATUS

     6.2.1   For Standard  Preparation
                            o (

-------
        6.2.1.1   Analytical balance

        6.2.1.2   Class A pipets,  various sizes

        6.2.1.3   Glove box

6.2.2   For Sample Homogenization

        6.2.2.1   Desiccator, to dry solid samples

        6.2.2.2   Blender, glass  jar,  for mixing dried
                  pulp and sludge sample

        6.2.2.3   Laboratory mill,   for  grinding  wood
                  chips

6.2.3   For Per Cent- Moisture Determination

        6.2.3.1   Desiccator

        6.2.3.2   Drying oven

6.2.4   For Sample Preparation

        6.2.4.1   Filtering funnel

        6.2.4.2   Round bottom flask, 5 L

        6.2.4.3   Teflon magnetic stir bag, egg-shaped,
                  (51 mm x 19 mm)

        6.2.4.4   Magnetic stir plate

        6.2.4.5   Soxhlet  extractor,   45  mm  I.D.,  with
                  500 mL erlenmeyer flask

        6.2.4.6   Snyder column, three ball macro

        6.2.4.7   Extraction heater

        6.2.4.8   Nitrogen   blowdown  apparatus,  water
                  bath   capable   of   maintaining
                  temperation ±5°

        6.2.4.9   Wrist action shaker

        6.2.4.10  Tube furnace, for silica activation

        6.2.4.11  Muffler    furnace,   capable  of
                  maintaining  600° ±10° ,  for alumina
                  activation

-------
7.   REAGENTS  AND CONSUMABLE MATERIALS

     7.1   CONSUMABLE MATERIALS

          7.1.1   Filter paper, 15  cm, 2.5  micron  retention
                 rating,   glass  fiber   filters   are  also
                 recommended

          7.1.2   Aluminum weighing boats

          7.1.3   Glass bottles,  various  sizes,  2L,  1L,  250  mL,
                 and 125 mL, equipped with Teflon-lined caps

          7.1.4   Dish, petri, 150 x 15 mm

          7.1.5   Silica  gel (Bio-Sil  A 100/200 mesh):   Bio-Rad
                 Rockville Centre, NY.   the  silica  gel  is
                 conditioned prior to use by initially placing a
                 200 g portion  of the silica  in  a  30  mm  x 30 cm
                 long  glass tube  (the silica gel  is held  in
                 place by glass wool plugs) which is placed in a
                 tube furnace.  The glass tube is connected to a •
                 pre-purified nitrogen cylinder through a series
                 of oxygen scrubber traps. The first step  in
                 conditioning the silica gel entails heating  the
                 glass  tube containing the  200 g  aliquot  of
                 silica  for 30 minutes  at 180° C  while  purging
                 with  nitrogen (flow  rate 50-100  mL/minute),
                 subsequently the  tube  is  removed from  the
                 furnace   and  allowed   to  cool   to  room
                 temperature.   Methanol  (175  mL) is then passed
                 through the tube, followed by 175 mL methylene
                 chloride.   The tube  containing the  silica  is
                 then  returned to  the furnace,  the nitrogen
                 purge is again established (50 to 100 mL/minute
                 flow), the tube  is  heated  at  50°C  for  10
                 minutes,  then the  temperature  is  gradually
                 increased  to 180°C  over  a period  of  25 minutes
                 and  maintained  at  180°C  for  90  minutes.
                 Heating is then  discontinued but  the nitrogen
                 purge  is  maintained until the tube cools  to
                 room temperature.  Finally, the  silica  is
                 transferred to a  clean,  dry, glass  bottle  and
                 capped with  a Teflon-lined  screw cap  for
                 storage in a desiccator.

          7.1.6   Silica Gel  Impregnated with  Sulfuric Acid (30%
                 w/w) :   Concentrated sulfuric  acid  (4.4  g)  is
                 combined with  10.0 g silica gel (conditioned as
                 described  in  Section  7.1.5)  in a  screw capped
                 bottle  and   agitated  to  mix   thoroughly.
                 Aggregates  are dispersed  with  a  stirring  rod
                 until  a  uniform  mixture is  obtained.    The
                 Hz SO*-silica  gel  is  stored  in a  screw-capped
                 bottle  (equipped with a Teflon-lined cap).
                                 o c
SO

-------
     7.1.7   Silica Gel  Impregnated  with Sodium  Hydroxide:
             IN Sodium hydroxide (30  g)  is combined with 100
             g Bio-Sil  A (conditioned as  described in
             Section 7.1.5)  in a screw  capped bottle  and
             agitated  to mix thoroughly.  Aggregates  are
             dispersed with  a  stirring  rod until  a  uniform
             mixture is  obtained.    The  NaOH-silica gel is
             stored in  a screw-capped bottle  (Teflon-lined
             cap) .

     7.1.8   Carbon/Celite:    A 10.7 g   aliquot  of PX-21
             carbon (Anderson Development Co.,  Adrian,
             Michigan)  is combined with 125 g of  Celite 545
             in a 250 mL glass bottle,  fitted with a Teflon-
             lined cap, and the mixture is shaken  to obtain
             a uniform  mixture.   The Carbon/Celite  mixture
             is stored in' the screw-capped bottle.

     7.1.9   Micro-reaction vial,  3 mL

     7.1.10  Glass boiling beads,  solvent washed

     7.1.11  Soxhlet thimble (90 mm x 35  mm),  glass

     7.1.12  Glass wool, silanized

     7.1.13  Nitrogen gas, prepurified

     7.1.14  Hydrogen,  ultra high purity

     7.1.15  Syringes,  various sizes

     7.1.16  Capillary gas chromatography column,  DB-5,  60  M
             x 0.25 mm I.D.; 0.25 um fil  thickness (J  and  W
             Scientific Co.)

     7.1.17  Capillary gas chromatography column,  DB-225, 30
             M x 0.25 mm I.D.; 0.25 um film thickness (J and
             W Scientific Co.)
7.2  REAGENTS
     7.2.1   Potassium Hydroxide (ACS):   40 percent (w/v)  in
             distilled water

     7.2.2   Sulfuric acid (ACS),  concentrated

     7.2.3   Methylene chloride, methanol,  acetone, benzene,
             hexane, ethyl acetate,  cyclohexane,  tridecane.
             Distilled in glass or highest  available purity

     7.2.4   Anhydrous sodium sulfate (ACS), Dry  in an  oven
             at 400°C for 3 hours,  cool  in  a desiccator  over
             self-indicating silica gel,  transfer to Teflon-
                      10
                              381

-------
             lined  screw  cap bottles and seal.  Store in  a
             desiccator until use.

7.3  CALIBRATION AND SPIKING STANDARDS

     7.3.1    Stock  standard solutions of the  appropriate
             TCDD  and TCDF  isomers,  and mixtures  thereof,
             are  prepared in  a  glovebox,  using  weighed
             quantities  of the authentic isomers.  These
             stock  solutions are  contained  in  appropriate
             amber  bottles  and  are stored tightly  stoppered
             in a  refrigerator.    Aliquots  of the stock
             standards are removed  for direct  use or for
             subsequent serial  dilutions  to  prepare working
             standards.   These standards must  be checked
             regularly (by comparing instrument  response
             factors  for them  over  a  period of  time)  to
             ensure that solvent evaporation or other losses
             have  not occurred  which  would alter the
             standard  concentration.  The standard solutions
             which  may   be  required  to   perform  the
             quantitative analyses of  2,3,7,8-TCDD and
             2,3,7,8-TCDF are listed  below.

     7.3.2    Internal  Standard  (IS)  Solution.    An  aliquot
             of this  solution  is  added to samples  which  are
             to be analyzed  for   2,3,7,8-TCDD and  2,3,7,8-
             TCDF.  Prepare a  stock solution containing  the
             following isotopically-labelled TCDD  and TCDF
             compounds  in isooctane  at  the indicated
             concentrations:  0.05 ng/uL l3Ciz-2,3,7,8-TCDD,
             0.02 ng/uL  37C14-2,3,7,8-TCDD,  0.05 ng/uL
             13Ci2-2,3,7,8-TCDF,  and 0.02  ng/uL 37Cl«-
             2,3,7,8-TCDF. Typically a twenty  microliter
             aliquot  of  this standard solution  is  added  to
             each  sample  aliquot  prior to  preparation  and
             the  13Ci2-labelled materials serve  as  internal
             standards for  use in  quantitation.  Recovery of
             these standards  is  also used to gauge  the
             overall  efficacy of the  analytical procedure.

     7.3.3    External Standard  (ES)  Solution.   Prepare  a
             stock solution containing 0.05  ng of  *3Ci2-
             1,2,3,4-TCDD  and  0.10  ng of  3 7 Cl« -1, 2 , 7 . 8-
             TCDF/uL  tridecane.  A 10 microliter  aliquot of
             this  standard is  added to the final extract
             obtained  for each sample  just  prior  to GC-MS
             analysis.   When  the  DB-5 capillary column  is
             employed, the  r3Ci2-1,2,3,4-TCDD  is used as  an
             external  standard  in the quantitation   of  the
             i3Ci2-2,3,7,8-TCDD  and  the  *3Ci2-2,3,7,8-TCDF
             internal standards present  in  the final
             extract,  and the  percent  recovery  of  each  of
             these *3Ci2-labelled internal  standards   is
             calculated  on  the  basis of this quantitative
                      11       9

-------
        analysis.   The  37Cl«-1,2,7,8-TCDF external
        standard is  employed in  quantitating  the
        concentration  of  13Ci2-2,3,7,8-TCDF when  the
        hybrid  DB-5/DB-225  capillary  column  is
        implemented  in  quantitating  2,3,7,8-TCDF.
        These   latter   results   are   subsequently
        implemented  in calculating  the percent recovery
        of the  l3Ci2-2,3,7,8-TCDF  internal standard
        achieved during  the  analysis  performed  using
        the hybrid column.

7.3.4   Calibration  Standard  Solutions  (CS1, CS2, CS3,
        CS4  and CSS).     Prepare   five  separate
        calibration  standards  as  follows:    (a)
        Calibration  Standard  (CS1),  0.2 ng/uL 2,3,7,8-
        TCDD,  0.2 ng/uL  2,3,7,8-TCDF, 0.05 ng/uL »3Ci2 -
        2,3,7,8-TCDD,  0.05  ng/uL   »3Ci2-2,3,7,8-TCDF,
        0.02 ng/uL-37C14-2,3,7,8-TCDD,  0.02  ng/uL
        37C1«-2,3,7,8-TCDF,   0.05 ng/uL  »3Ci2-1,2,3,4-
        TCDD and 0.10 ng/uL 37Cl«-1,2,7,8-TCDF;  (b)
        Calibration  Standard  (CS2),  0.05  ng/yL 2,3,7,8-
        TCDD,  0.05  ng/uL 2,3,7,8-TCDF plus the same -
        concentration   of   isotopically-labelled
        standards included  in  CS1;  (c) Calibration
        Standard (CS3),   0.01  ng/uL 2,3,7,8-TCDD,  0.01
        ng/uL 2,3,7,8-TCDF plus the  same concentration
        of isotopically-labelled standards  included=s_jjiv
        CS1; (d) Calibration Standard  (CS4) , ^6.0025^-
        ng/uL  2,3,7,8-TCDD,  0.0025  ng/uL 2 , 3,7Y8-TCDF'
        plus the  same  concentration  of isotopically-
        labelled standards  included  in  CS1;  (e)
        Calibration  Standard  (CSS), 2.0 ng/uL  2,3,7,8-
        TCDD,  2.0  ng  2,3,7,8-TCDF  plus  the  same
        concentrations  of  isotopically-labelled
        standards included in  CS1.   Aliquots  of these
        standards are  injected to obtain data which is
        implemented in  constructing the calibration
        curve  used  in quantitating  2,3,7,8-TCDD and
        2,3,7,8-TCDF.

7.3.5   TCDD Gas Chromatographic  Resolution   Standard
        Mixture (RM1).    Prepare  an isooctane  solution
        containing  0.05  ng/uL concentrations of each of
        the  following  TCDD  isomers:   1,3,6,8-TCDD;
        1,2,3,7-TCDD;   1,2,3,9-TCDD;   2,3,7,8-TCDD;
        1,2,3,4-TCDD  and 1, 2 , 8 , 9-TCDD .   Two of  the
        isomers in  this  mixture are used to define  the
        gas Chromatographic  retention  time  window  for
        TCDDs  (1,3,6,8-TCDD  is  the first eluting TCDD
        isomer  and  1,2,8,9-TCDD  is  the last  eluting
        TCDD  isomer  on the DB-5  GC column).   The
        remaining isomers serve to demonstrate  that  the
        2,3,7,8-TCDD isomer is resolved  from the other
        nearest  eluting TCDD  isomers,  and that  the
        column  therefore yields  quantitative  data  for

                        383
                 12

-------
                  the  2,3,7,8-TCDD isomer alone.

          7.3.6   TCDF Native Standard Mixture  (NS1).    Prepare
                  a  solution  containing 0.250 ng/uL 2,3,7,8-TCDF
                  in tridecane.  This standard  is implemented
                  when it  is  desired to add only native 2,3,7,8-
                  TCDF to  a sample.

          7.3.7   TCDF Gas  Chromatographic  Resolution  Standard
                  Mixture  (RM2).   Prepare  a solution containing
                  approximately 0.250  ng/uL of  each  of the  37
                  TCDF isomers  (exclusive  of  2,3,7,8-TCDF)  in
                  isooctane.   This  standard  is used when  it  is
                  desired  to  add all of the  TCDF isomers except
                  2,3,7,8-TCDF to a  sample.   This standard  is
                  also implemented  to determine  the relative
                  retention  times  of the TCDF  isomers  and,  when
                  this standard is co-injected with an aliquot of
                  standard NS1,  the  efficacy  of a particular gas
                  Chromatographic column  for  separating 2,3,7,8-
                  TCDF from each of  the 37  other TCDF isomers can
                  be ascertained.

          7.3.8   TCDD/TCDF  Native  Standard Mixtures  NS2  and
                  NS3.  For  Mixture NS2, prepare a solution
                  containing  0.025  ng 2,3,7,8-TCDD  and  0.025  ng
                  2,3,7,8-TCDF  per microliter of tridecane.   For
                  mixture  NS3,  prepare a  solution containing
                  0.005 ng 2,3,7,8-TCDD  and 0.005 ng 2,3,7,8-
                  TCDF.  These  standards  are  employed  when  it  is
                  desired to simultaneously add  both native
                  2,3,7,8-TCDD and native  2.3,7,8-TCDF  to a
                  sample.
8.   SAMPLE COLLECTION,  PRESERVATION AND HANDLING

     8.1  Grab and composite samples must be  collected  in glass
          containers.    Conventional  sampling  practices must  be
          followed.  The bottle must not be prewashed with sample
          before  collection.   Composite samples  should  be
          collected  in glass containers.  Sampling equipment must
          be free of tygon, rubber  tubing and  other potential
          sources  of contamination which  may  absorb  the  target
          analytes.

     8.2  All samples  are refrigerated at  4°C,  and sample
          preparation must  commence within 30 days.


9.   CALIBRATION AND STANDARDIZATION:

     9.1  Calibrating the  MS  Mass Scale:   Perfluorokerbsene,
          decafluorotriphenyl  phosphine,  or  any  other  accepted
          mass marker compound must be introduced into the MS,  in



                           13

-------
     order to calibrate the mass scale through at least  m/z
     350.    The procedures  specified by the manufacturer  for
     the particular MS  instrument  used are to  be  employed
     for this  purpose.   The mass  calibration should  be
     rechecked  at least at  8  hr. operating intervals.

9.2  Gas Chromatograph Temperature Program:   Table 1  shows
     the GC  temperature  program typically used  to  resolve
     2,3,7,8-TCDD from each of the  21 other TCDD isomers  and
     indicates  the ion-masses  monitored and  the time
     analytical  sequence  implemented  for  isomer specific
     quantitation of  2,3,7,8-TCDD  and non-isomer  specific
     quantitation of 2,3,7,8-TCDF.  This temperature  program
     and ion monitoring  time cycle must be  established  by
     each  analyst for  the  particular instrumentation used by
     injecting  aliquots of Standard Mixture RM1,  as well  as
     the calibration mixtures (CS1, CS2,  CS3  and  CS4)  into
     the  GC-MS .    It may be  necessary  to  adjust  the
     temperature program and ion monitoring cycles slightly
     based  on  the  observations  from analysis of these
     mixtures.

9.3  Checking  GC Column Resolution for 2,3,7,8-TCDD  and
     2,3,7,8-TCDF:  Utilize Standard  Mixture  109071-1  to
     check the  DB-5 column resolution  for 2,3,7,8-TCDD,  and
     utilize a combination  of  Standards  NS1  and  RM2  to
     verify that 2,3,7,8-TCDF is separated  from  all  of  the
     other TCDF  isomers on the  hybrid  DB-5/ DB-225  column.
     A 25% valley or less must be obtained between the mass
     chromatographic  peak observed for 2,3,7,8-TCDD  and
     adjacent  peaks arising from  other TCDD isomers  and
     similar  separation  of  2,3,7,8-TCDF  from  other
     neighboring  TCDFs is required.   Analyze the column
     performance standards  using the instrumental parameters
     specified  above  and  in  Table 1  and 2.   The column
     performance evaluation must be accomplished each time a
     new column  is  installed  in the gas  chromatograph,  and
     at the beginning  and  conclusion  o.f  each  8 hour
     operating  period.   If  the column resolution is found to
     be  insufficient  to resolve 2,3,7,8-TCDD  and  2,3,7,8-
     TCDF from their  neighboring  TCDD and  TCDF isomers,
     respectively,  (as measured  on  the  two different columns
     used   for resolving   these two isomers) ,   then   a  new
     DB-5  and/or DB-5/DB-225 hybrid GC column must  be
     installed.

9.4  Calibration of the  GC-MS-DS  System:  To accomplish
     quantitative analysis of 2,3,7,8-TCDD and 2,3,7,8-TCDF
     contained  in the  sample extract,  the  GC-MS  system  is
     calibrated  by  analyzing a series of  at least  three
     working calibration standards.  Each of these standards
     is  prepared to contain  the same  concentration of each
     of   the »3Ci 2-2,3,7,8-TCDD   and  >3Ci 2-2,3,7,8-TCDF
     internal   standards  used   here  but  a  different
     concentration of   the  native  2,3,7,8-TCDD and 2,3,7,8-
                      14
                               °> O ^
                               ooO

-------
          TCDF.  Typically, mixtures will be prepared so  that  the
          ratio  of the native  2,3,7,8-TCDD and 2,3,7,8-TCDF  to
          the  isotopically-labelled TCDD/TCDF ranges between 0.05
          and  4.0 in  the four  working calibration  mixtures.
          Prior  to injecting  aliquots  of  actual  sample extracts,
          an aliquot of a standard containing typically 0.2 ng of
          1 3Ci 2-1,2,3,4-TCDD and 0.4  ng  of a 7 CI* -1 , 2 , 7 , 8-TCDF
          (External  Standard)  is used  to  dilute the  extract  in
          the  sample vials and  is  therefore co-injected along
          with the  sample extract,  in  order  to  obtain  data
          permitting calculation of the percent recovery of  the
          1 3Ci 2-2,3,7,8-TCDD  and *3Ci 2-2,3,7,8-TCDF internal
          standards.  When  the analysis  of  the extract   is
          performed  using the DB-5 capillary column,  the x 3 Ci 2 -
          1,2,3,4-TCDD  standard is  implemented  as  the  external
          standard in quantitating  both l3Ci2-2,3,7,8-TCDD  and
          13Ci2-2,3,7,8-TCDF.   However, when the hybrid  DB-5/DB-
          225  column is employed in analyzing  the  2,3,7,8-TCDF,
          the  3TCl«-1,2,7,8-TCDF is implemented in quanitiating
          the  J 3Ci 2-labelled  TCDF internal  standard.  Equations
          for  calculating  relative response  factors from  the
          calibration  data  derived  from the calibration  standard
          analyses,  and  for  calculating the  recovery  of   the
          13Ci2-2,3,7,8-TCDD  and »3Ci 2-2,3,7,8-TCDF,  as  well  as
          the  concentration of  native  2,3,7,8-TCDD  and  2,3,7,8-
          TCDF in the sample  (from the extract analysis) ,  are
          summarized below.

     9.5  Daily Checks of the Instrument Response:   These will be
          accomplished using  Standard CS3.  This standard will be
          injected at the  beginning of each work-day  (or  the
          beginning  of each  8-hour shift)  and RRF  values  for
          2,3,7,8-TCDD  and  2,3,7,8-TCDF will be calculated.    If
          either of  these  RRF  values  deviate from  the values
          contained  in the calibration curve by more  than +20%,
          then a second injection will be made and RRF values  for
          the  two  compounds will be again  calculated.  If either
          of  these  RRF  values  also  fail to  agree with  the
          calibration  curve by  more than  ±20%,  then  the entire
          series  of  calibration standards will  be  analyzed,  new
          RRF  values will be  calculated,  and a new  calibration
          curve  will be constructed and applied in subsequent
          analyses.


10.   QUALITY ASSURANCE/QUALITY CONTROL PROCEDURES

     10.1  The  Quality Assurance  and Quality  Control  procedures
          itemized below  will  be  implemented  throughout   the
          course of  the 2,3,7,8-TCDD and 2,3,7,8-TCDF analyses:

          10.1.1   Each sample  analyzed is  spiked with stable
                  isotopically-labelled internal standards, prior
                  to extraction and analysis.   Recoveries
                  obtained for  each of these standards should
                          15
                                   386

-------
             typically be in the range from 40-120%.   Since
             these  compounds  are used as true  internal
             standards however,  lower  recoveries  do not
             necessarily  invalidate the  analytical  results
             for native  2,3,7,8-TCDD  and  2,3,7,8-TCDF, but
             may result in higher detection limits than are
             desired.

     10.1.2  Processing and  analysis  of  at least one method
             blank sample is generally accomplished for each
             set of samples.

     10.1.3  It is desirable to  analyze at least one  sample
             spiked with  representative native TCDD/TCDF for
             each  set  of samples.   The  results of this
             analysis  provides an indication of the efficacy
             of the entire analytical  procedure.  The
             results of  this  analysis will  be considered
             acceptable if the detected concentration of the
            . native 2,3,7,8-TCDD and  2,3,7,8-TCDF  added  to
             the  sample is  within +_50% of  the  known
             concentration.

     10.1.4  At least one of  the samples  analyzed  out  of
             each set  is  usually analyzed in duplicate and
             the  results of  the duplicate  analysis are
             .included  in  the report of data.

10.2 A report  describing  the  results  of the  analyses
     discussed above will, at a  minimum,  include  copies  of
     original mass chromatograms  obtained during analyses  of
     the  sample  extracts and associated calibration
     standards, a description  of the analytical methodology
     employed, and tabulations of  calculated  results.
     Calculations and  manipulation  of  data  are most
     efficaciously accomplished using computerized data
     reduction  techniques.   The tabulations  of calculated
     results  provided in the report will  include  tables
     showing  the  concentrations of  2,3,7,8-TCDD  and  of
     2,3,7,8-TCDF which were measured in  each sample.  Also
     typically  shown  in  these tables are the quantity  of
     each  sample  analyzed;  the  detection  limits  for those
     samples which were found  to contain no 2,3,7,8-TCDD  or
     2,3,7,8-TCDF; the GC-MS  instrument  implemented  in  the
     analysis; the date and time of the analysis;  the ratio
     of  the intensities of m/z  320 vs.  m/z  322  and  m/z  332
     vs.  m/z 334  for  TCDD and the ratio of the intensities
     of  m/z  304 vs. m/z  306  and m/z 316  vs. m/z  318  for
     TCDF; the  percent recovery of the  internal standard
     (i3Ci 2-2,3,7,8-TCDD  or  »3Ci2-2,3,7,8-TCDF); and the ion
     intensities  for  the  following  m/z's:  320,  322,  257,
     332  and  334  for  TCDD,  and 304,  306,  241,  316  and  318
     for  TCDF.  Other  tabular data which should be provided
     in  the report include a table showing a summary of  the
     calibration data  obtained for  2,3,7,8-TCDD and 2,3,7,8-
                      16    387

-------
          TCDF,  which indicates  the  date of the calibration;  the
          GC-MS  instrument  implemented;  the  WSU  identification
          number of  the  calibration solution; the calculated
          response  factors  and the mean response factors obtained
          for native  2,3,7,8-TCDD,  native  2,3,7,8-TCDF,  *3Ci2 -
          2,3,7,8-TCDD  and »•Ci2-2,3.7 , 8 ;  and  the % valley
          observed  for the  GC separation  of  the 2,3,7,8-TCDD from
          adjacent-eluting  TCDD  isomers  and of the 2,3,7,8-TCDF
          from adjacent-eluting TCDF  isomers.   Additional tables
          which  present calibration data  and results obtained for
          each individual sample in  a more  detailed manner  than
          that given  in  the summary tables mentioned earlier
          should also be  included in  the  data package.


11.   PROCEDURE

     11.1 SAMPLE PREPARATION  -

          11.1.1 Sludge  Samples

                 11.1.1.1  Open the  sample  container and using a
                           spatula,  break  the sludge into small
                           particles (about  2 cm  diameter  or
                           less) and stir  the sample vigorously
                           to   make it  as  homogeneous   as
                           possible.   Remove  an aliquot of  this
                           sample  (approximately  5  g) for  an
                            "oven-dried  solids  as-received"
                           determination,  using  the  procedures
                           described below  (Section 11.1.1.4).
                           Remove the  remaining sample from the
                           container and distribute it uniformly
                           on a  stainless  steel screen which is
                           supported at  a distance of about 1 cm
                           above a  sheet of aluminum foil,  both
                           the   foil . and  the  screen  being
                           contained   within  a   desiccator
                           containing  an appropriate  water
                           sorbent.  To minimize  the possiblity
                           of    contamination    or   cross-
                           contamination of the sample, only one
                           sample at a  time is dried in a given
                           desiccator.  Allow the  sample  to
                           remain in the desiccator until it is
                           essentially  dry, as indicated  by the
                           sample  color, consistency,  and  ease
                           of mixing.    For each group of  five
                           sludge samples  which are desiccated,
                           prepare  a   laboratory  blank  as
                           follows.  Place  a  15 cm paper filter
                           on a stainless steel screen supported
                           at a  distance of 1  cm  above a sheet
                           of  aluminum  foil  contained in  a
                           desiccator  and  allow  the filter  to
                           17    388

-------
          remain in the desiccator for the same
          period  as that which was used  for
          that  sample  of  the   five   which
          required the longest drying  time.
          Subsequently remove  the filter from
          the  desiccator and continue with the
          homogenization,  drying,  and  other
          sample  preparative steps  described
          below.

11.1.1.2  When the sample  has  been  dried
          sufficiently,  remove it from  the
          desiccator  and  transfer  it to  a
          laboratory  blender which  is  housed
          within  a  glove  box   or  similar
          enclosure.   Following homogenization
          in  the blender,  remove an  aliquot
          (approximately  5 g)  of  the  blended
          solids,  accurately weigh this  sample
          aliquot,  and  subject  it to an oven-
          dried  solids   determination,   as
          described in Section 11.1.1.3 of this
          protocol.

11.1.1.3  Place  the  remaining  desiccated,
          blended  sample  into  a  clean  sample
          bottle  fitted with  a  Teflon-lined
          screw cap, and store  the bottle in a
          refrigerator  (5°C).  An aliquot of
          this desiccated, blended  sample is
          subsequently  analyzed  for  2,3,7,8-
          TCDD and 2,3,7,8-TCDF by applying the
          extraction  and  analysis procedures
          which are described in  Sections 11.3
          and 11.4,  respectively,  of  this
          protocol.

11.1.1.4  Determination of oven-dried-solids on
          a  sample aliquot is  accomplished by
          placing  the  weighed  aliquot  into a
          tared  aluminum boat  which is then
          placed  in  an  oven  maintained  at
          105°C.  After heating for a period of
          twenty-four hours, the  aluminum boat
          containing the sample is removed from
          the oven, allowed to  cool for  30
          minutes  in a desiccator,  and then
          weighed.   The boat  and sample  are
          then returned  to the  oven for  an
          additional  four-hour period,  after
          which, the boat is again removed from
          the  oven, allowed to cool and weighed
          again.   The latter  procedure  is
          repeated until  the  weight  of  the
          sample as indicated by two  successive
                 380

-------
                 weighings is observed to be constant.
                 From  the observed weight loss upon
                 drying,  the percentage of oven-dried-
                 solids  in  the  original  sample  can  be
                 determined.      This  result,   as
                 determined for  an  aliquot of the
                 sample   as  received,  is  reported  as
                 the  "initial  oven-dried  solids  as
                 received."  The  oven-dried weight
                 loss, as determined for an aliquot  of
                 the  previously desiccated  sample  (a
                 separate   aliquot   of   which   is
                 subsequently analyzed  for TCDD/TCDF)
                 is used only  to determine the  actual
                 weight  of the  sample aliquot which  is
                 analyzed on the  oven-dried solids
                 basis.

11.1.2  Wood Chip Samples:   Samples  of wood  chips  in
        which the chips are relatively large  (typically
        1-1.5 inches  in  length)  are  initially reduced
        to smaller particle size  (2  cm  diameter  or
        less)  using a  laboratory mill.   This mill  is
        cleaned  thoroughly before  each  sample  is
        introduced.    The  pulverized  wood sample
        resulting from  this operation is subsampled and
        dried  using  exactly  the same  procedures
        described  for  sludge samples  in  the foregoing
        Section 11.1.1.1.

11.1.3  Ash Samples:    Ash Samples  are  prepared  using
        the same procedures as  described above for
        sludge (Section  11.1.1.1),  with  the exception
        that these samples  cannot be  supported on a
        screen to dry,  and are  therefore placed  into a
        shallow,  flat  dish in order to dry them  in the
        desiccator.  The  ash" is  spread as a thin layer
        over  the bottom  of  the dish and  is gently
        stirred periodically during the drying period.

11.1.4  Paper Pulp Samples:   Remove  the pulp sample
        from the  container,  and then  express as  much
        water  from   the  sample  as   possible   by
        compressing  it with  a   spatula  after wrapping
        the sample  in aluminum foil. Using the spatula,
        separate  the sample  mass into  pieces which are
        about 2 cm  or  less in diameter,  and distribute
        these pieces uniformly  on a  stainless  steel
        screen supported about  1 cm  above  a  sheet  of
        aluminum  foil, both  the screen  and  the  foil
        being placed in a desiccator.  Allow the sample
        to  remain in the  desiccator  until  it  is
        essentially  dry,  as   gauged  by  color and
        consistency.    For  each group  of  five  pulp
        samples,  prepare a  laboratory blank using  the
                 19      390

-------
        procedures described in 11.1.1.1.   Proceed with
        the subsaropling and other drying procedures, as
        described  for  sludge  samples,  beginning with
        Section 11.1.1.2.

11.1.5  Slurry-Type Samples  (Secondary  Sludge,  etc.):
        Shake the  sample  bottle  vigorously  so as to
        obtain a uniform  suspension  of  the sample and
        when the  sample  is homogeneous  throughout, as
        judged by visual  inspection, remove an  aliquot
        of  the  sample and subject  it  to  a  Total
        Suspended Solids Determination, as  described in
        Standard Methods For the  Examination of  Water
        and Wastewater,   17th   Edition.  APHA,  AWWA,
        WPCF,  1986, Method  209C.   Allow the  remainder
        of the sample to  stand  under refrigeration and
        when the solids appear  to have totally  settled
        to  the  bottom of  the   container,  filter the
        supernatant using  a previously  desiccated and
        XXX tared Gelman Type A/E filter contained in  a
        glass filtering funnel. Remove the solids from
        the sample container using a clean spatula and
        utilize three 100 mL aliquots of HPLC water to
        accomplish three  successive  rinses  of the
        sample container,  and  to  effect a  quantitative
        transfer  of  the  solids  from   the  sample
        container to the  filter.  Following separation
        of the water from the  solid,  remove the  solid
        along with the  filter paper  from the funnel and
        distribute the solids and the filter paper  on
        a stainless steel screen supported about 1 cm
        above a sheet of aluminum foil, both the screen
        and the  foil being placed  in  a  desiccator.
        Allow the  sample  to dry  until  it  is  friable.
        For each group  of  five  samples,  prepare   a
        laboratory blank using  the procedures  described
        in 11.1.1.1.  Proceed  with the subsampling and
        other drying  procedures  described  for  sludge
        samples,  beginning with Section 11.1.1.2.  Note
        that the  tare  weights   of all  filters used in
        the separation of the  liquid and solid  phases
        of  these  samples  must  be subtracted  from the
        combined solids-filter  weight to determine the
        actual weights of the  solids samples  prepared,
        since the  filter  and solids cannot be  readily
        separated.

11.1.6  Water and Wastewater Samples

        11.1.6.1  Clean   and prepare four  new  2   L
                  bottles  fitted with  Teflon-lined
                  caps.   Mark the  1 gallon  bottle
                  containing the aqueous  sample,  as
                  received,  to  show the original  level
                  of the  liquid in  the  bottle.   Shake
                 20

-------
          the  bottle  vigorously  until  all
          solids  in the bottle (which may have
          settled to the  bottom  of  the bottle
          if the  sample was  undisturbed  for
          some time prior to analysis)  are
          suspended,  as visually estimated.
          Pour  approximately equal portions of
          the resuspended  aqueous  sample  from
          the 1 gallon bottle into each of the
          four  2  L bottles using a funnel.   To
          accomplish this transfer,  pour small
          portions  from the  1  gallon  bottle
          into each  of the  2 L  bottles  in
          succession,  repeating this  cycle as
          many  times  as  necessary to dispense
          all of  the  contents  of  the 1 gallon
          container.  Following each  pouring
          step,  recap the  1 gallon bottle and
          shake it  vigorously to ensure  that
          any particulate in the liquid  remains
          suspended.

11.1.6.2  After all of  the contents of the  1
          gallon  bottle  have  been  transferred
          to the  four 2 L bottles, rinse the  1
          gallon  bottle  successively  with two
          50 mL  portions  of HPLC grade water,
          accumulating these rinses  in a 250 mL
          graduated beaker.  Transfer  one-
          fourth  of each of these  accumulated
          water rinses to each of the four 2  L
          bottles.   Rinse  the  250 mL  beaker
          successively with two 40 mL portions
          of HPLC water,  transferring  one-
          fourth  of these rinses to  each of the
          four  2  L bottles.   Recap all four 2  L
          bottles  and retain  for  subsequent
          extraction and analysis.

11.1.6.3  Rinse  the original  1  gallon empty
          sample  bottle  successively  with two
          50 mL portions of methylene chloride,
          accumulating these  in the  250 roL
          beaker  used earlier.  Transfer these
          methylene chloride rinses to  a clean
          250 mL bottle  fitted with a  Teflon-
          lined cap.  Rinse the  250 mL beaker
          successively with two 50 mL portions
          of methylene chloride, and  pool these
          rinses with the  other accumulated
          methylene chloride rinses in  the 250
          mL bottle.    Reserve   the   pooled
          methylene chloride  rinses  for later
          splitting and  combination with the
          methylene chloride  rinses collected
         21

-------
          as described  in  Section 11.2.2 below.

11.1.6.4  Select  one  of the  2  L  bottles
          containing  the split water/wastewater
          sample,  and  add to  this  bottle  a
          solution of  the »3Ci2-labelled  TCDD
          and TCDF internal standards, prepared
          by combining the 20  iiL  of Internal
          Standard with 1.0 mL  of  acetone  in a
          glass test  tube.  Rinse the test tube
          with 0.5 mL  acetone,  followed  by a
          second 0.5  mL portion of acetone, and
          transfer these  rinses  to the  aqueous
          sample.

11.1.6.5  Place a  Teflon-coated,  magnetic
          stirring bar  in  the sample container,
          and stir the aqueous  sample  using a
          magnetic stirplate  for  15 minutes to
          disperse   the  spiking  solution.
          Position   the  stem  of  a  glass
          filtering  funnel to  discharge into a
          pre-cleaned  5 L round bottom  flask
          and place  a  paper  filter retention
          rating,  2.5 u, filter,  into  the
          funnel.

11.1.6.6  Decant  and/or  pour  the  internal
          standard-spiked  water sample from the
          2  L  bottle  into  the  filter  and
          collect the filtrate  in  the  5 L
          flask.

11.1.6.7  Rinse the  empty 2 L sample container
          sequentially  with  three  100  mL
          aliquots of HPLC grade water,  pouring
          each rinse through  the  filter,  and
          collecting the  filtrate in  the  5 L
          vessel.  Check  to  ensure that  all
          residual  particulates  and sediments
          are removed from the original sample
          container by  the  aqueous rinsing
          procedure.   Retain  this  filtrate for
          subsequent  extraction using  the
          procedures  described in Section 11.2.

11.1.6.8  Transfer  the  combined  filter  and
          particulate to a clean Petri dish and
          place them into a desiccator.  Allow
          these solids to dry  completely (as
          indicated  by constant  weight  upon
          successive  weighings).   Retain these
          solids for subsequent  extraction as
          described  in  Section 11.3.2.
         22       393

-------
            11.1.6.9   Rinse  the  original  2  L  sample
                      container  sequentially  with  three 50
                      mL  aliquots of methylene chloride.
                      Pour  the rinsates  through  the  empty
                      funnel and collect  in a clean 1000 mL
                      glass bottle fitted  with a Teflon-
                      lined   lid.     Retain   this   for
                      subsequent combination with  the
                      methylene chloride extract of  the
                      aqueous   filtrate,    obtained   as
                      described  in Section 11.2.

     11.1.7  Exceptional Samples:   Some of  the samples
             received may  be too  wet to  dry efficiently in a
             desiccator,   but   may  still   not  contain
             sufficient  liquid  to  permit  separation  of  the
             phases by filtering or other such means.   Such
             samples will  be  distributed on  sheets  of
             aluminum foil and  allowed to  air-dry at ambient
             temperature  on  a bench top.  For such samples,
             the surface  area  will be  recorded, so  that a
             correction  can  be  made,  if this  is necessary,
             for contamination  or cross-contamination of the
             samples.  The  extent  of  contamination of such
             samples  will be estimated  by placing  a filter
             paper blank in  the  same area where  these
             samples are air-dried  and this  blank will
             subsequently  be analyzed for 2,3,7,8-TCDD  and
             2,3,7,8-TCDF.   If the blank is  found  to be
             positive for  these  compounds, the corresponding
             levels  of these compounds  in the samples will
             be corrected for  the levels detected in  the
             blank.

11.2 PROCEDURES FOR EXTRACTING  2,3,7,8-TCDD AND 2,3,7,8-TCDF
     FROM AQUEOUS FILTRATE

     The  internal-standard-spiked,   aqueous  filtrate
     resulting from application of the  procedures described
     in Section 11.1.6.  is extracted utilizing the following
     procedures.

     11.2.1  Add 400  mL  of methylene chloride  to the aqueous
             filtrate contained in the  5  L  flask  (from the
             step described  in  Section 11.1.6.7).    Place a
             magnetic stirring  bar  into  the 5  L flask, place
             the flask on a  stir-plate,  and  stir the liquid
             in the flask  for 16  hours.

     11.2.2  Discontinue  stirring of the contents of the 5 L
             flask, allow the aqueous  and organic phases to
             separate, then  remove  the organic layer using a
             pipette, and place  it in  the  1000 mL bottle
             containing  the accumulated methylene  chloride
             rinsates collected as described  in  Section
                      23
0 f\ A
o d '1

-------
             11.1.6.9.    At  this point,  also  add to  the
             contents of this  same 1000  mL bottle,  one-
             fourth  of  the  methylene chloride rinsates
             collected  as described  in  Section  11.1.6.3.
             Retain  the  rest of the latter  rinsate  for
             subsequent  splitting among the  extracts of the
             other three splits  of the  original  aqueous
             sample,  if  these are subsequently  analyzed.

     11.2.3  Sequentially,  repeat  the extraction of  the
             aqueous  filtrate two additional times using a
             100  mL portion of methylene  chloride  each time,
             and  combine each  extract with the  original
             extract  in  the   1000 mL  bottle.   Reserve this
             pooled extract  for later combination with the
             Soxhlet  extract  of  the particulate and filter,
             as described in  Section 11.3.

11.3 PROCEDURES  FOR  SOXHLET-EXTRACTING  2,3,7,8-TCDD  AND
     2,3,7,8-TCDF FROM DRIED  BULK SOLIDS AND FILTERED
     WASTEWATER SOLIDS

     11.3.1  Dried Sludges,   Ash Samples,  Wood   Chips  and
             Paper Pulp:  Solid samples  of these  types,
             prepared as  described  in  Section  11.1,  are
             extracted using  the following procedures.

             11.3.1.1 Prepare  a glass Soxhlet extraction
                      thimble  (90 mm  by  35  mm)  for use by
                      rinsing   it   sequentially   with
                      methanol,  acetone  and  methylene
                      chloride.  Add silica to  form a 3-6
                      mm layer on the surface of the glass
                      frit at  the  bottom of the  thimble,
                      and place a 10 mm  layer  of  glass wool
                      over the layer of  silica.

             11.3.1.2 Prepare   a    Soxhlet   extraction
                      apparatus, consisting of  a  Soxhlet
                      extraction tube, a 250 mL  Erlenmeyer
                      flask  and a water-cooled  condenser,
                      for use  by rinsing it sequentially
                      with methanol,  acetone,  and roethylene
                      chloride, and allowing it to  air-dry.
                      Place  175 mL of a  solution  consisting
                      of 50% benzene and 50% acetone (by
                      volume) ,  along with  about  10  pre-
                      cleaned  2 mm glass beads,  into the
                      Erlenmeyer flask.   Place the Soxhlet
                      thimble  (prepared  as described  in
                      Step   11.2.1.1)  into  the  Soxhlet
                      extraction   tube,   assemble   the
                      Soxhlet-extraction apparatus ,  heat
                      the contents of the Erlenmeyer flask
                      to  reflux temperature,  and  continue
                     24
                           o
                           O

-------
          the  Soxhlet extraction procedure  for
          a period  of  3 hours.

11.3.1.3  Remove  the heat  source  from  the
          Soxhlet   apparatus,   allow   the
          apparatus to cool,  and then  decant
          the  benzene/acetone solution  into  a
          clean,  250  mL  flint glass  bottle  and
          seal  the bottle  with  a Teflon-lined
          screw cap.   This  solution is retained
          in case additional  analyses  are
          required  to check the  cleanliness of
          the  Soxhlet  apparatus,  as  a  QC
          measure.

11.3.1.4  Place a  fresh 175 mL aliquot of 50:50
          (volume/volume)  benzene/acetone  into
          the  Erlenmeyer  flask  of  the  Soxhlet
          extraction  apparatus  and  re-connect
          the  Soxhlet extraction tube  to  the
          Erlenmeyer  flask. Remove the layer of
          glass wool from  the glass thimble.
          Transfer  an   accurately  weighed
          aliquot  (approximately 7-10  grams,
          depending upon the sample type)  of
          the  previously desiccated solid
          sample,  prepared  as  described in
          Section  11.1,  from  the sample bottle
          containing the dried  sample  to  the
          Soxhlet  extraction  thimble.

11.3.1.5  Using  a  microsyringe,  add   the
          appropriate   internal   standard
          solution  described  in  Section 9.3.2
          to the solid sample in the  Soxhlet
          extraction  thimble.   Place  the
          previously  .removed  glass  wool (Step
          11.3.1.4) on top  of the sample in the
          glass thimble. Place the condenser on
          the  Soxhlet extraction tube and heat
          the  solvent reservoir  so  that  the
          extraction  solvent  refluxes.  Soxhlet
          extract  the sample  for  a period of 16
          hours,  then discontinue  heating  the
          apparatus  and  allow  it  to cool to
          ambient  temperature.

11.3.1.6  Remove the  Soxhlet  extractor from the
          Erlenmeyer  flask reservoir   and
          replace   the extractor  with  a 3-ball
          Snyder column.   Resume heating  the
          reservoir  and   concentrate   the
          benzene/acetone  extract to  a volume
          of about 15 mL.    Rinse the  Snyder
          column twice with small quantities of
                o r\ o
25     39

-------
                 hexane,  then  continue heating  and
                 concentrating the  solution  in  the
                 reservoir  with the column  in place
                 until  a  final volume of  10  mL is
                 attained.

        11.3.1.7  Using  a 10  mL disposable  pipette,
                 transfer the concentrated  solution
                 obtained in  Step  11.3.1.6  to a pre-
                 rinsed,  125 mL flint glass  bottle
                 fitted with a Teflon-lined screw cap.
                 Rinse the Erlenmeyer flask four times
                 using  10  mL  aliquots  of  hexane,
                 transferring  each  rinse  solution to
                 the 125 mL  bottle,  to  effect  a
                 quantitative  transfer   of   the
                 concentrate from the Erlenmeyer flask-
                 to. the bottle.

        11.3.1.8  Proceed with  the  remainder  of  the
                 clean-up and analytical procedures
                 described in Section 11.4.

11.3.2  Soxhlet  Extraction of Filtered Wastewater
        Solids

        11.3.2.1  Remove  the  desiccated filter  and
                 associated  solids  resulting from
                 filtration  of a  water/wastewater
                 sample  containing particulates, as
                 described  in  Section 11.1  from its
                 sample  container,  and  immediately
                 place  the  filter  and  solids into a
                 Soxhlet  extraction thimble  which has
                 been pre-cleaned,   as  described above
                 in Step  11.3.1.1.

        11.3.2.2  Pre-clean   a   Soxhlet  extraction
                 apparatus  as  described   in  Steps
                 11.3.1.2 through 11.3.1.4.

        11.3.2.3  Concentrate  the methylene  chloride
                 extract  resulting  from extraction of
                 the aqueous  filtrate which has been
                 pooled  with  other  methylene chloride
                 rinsates (obtained as described in
                 Section  11.2.2) by transferring about
                 150 mL  of  the methylene  chloride
                 extract  to a 250 mL Erlenmeyer flask,
                 attaching  a  3-ball Snyder   column to
                 the flask  and heating the  flask to
                 concentrate  the methylene  chloride.
                 Continue to  transfer  150 mL aliquots
                 of the  methylene  chloride extract to
                 the Erlenmeyer  flask  as  each  portion
                 26
                        O f\'-,
                        od <

-------
                      is    reduced    in    volume    by
                      concentration,  until the  volume of
                      the  extract is reduced  to about 25
                      mL.    Then  add  150  mL  of  50:50
                      volume:volume benzene-acetone to  the
                      Erlenmeyer flask containing  the
                      residue  from the  methylene  chloride
                      concentration and reconnect the flask
                      to the Soxhlet extractor.   Note  that
                      it  is not necessary  to spike  this
                      sample with  internal  standards since
                      the  wastewater  sample was  previously
                      spiked with  internal  standards prior
                      to filtering.

             11.3.2.4  Heat the  Soxhlet  apparatus  and
                      extract  the  filter and  solids for  a
                      period of  16 hours,  then discontinue
                      heating  and allow the  apparatus to
                      cool.   Remove  and concentrate  the
                      extract  as described  in  Section
                      11.2.1.6.    Transfer the concentrate
                      to a new sample bottle,  as described
                      in Section 11.3.1.7.

             11.3.2.5  Proceed  with the  remainder  of  the
                      clean-up and analytical procedures
                      .described in Section 11.4.

11.4 PROCEDURES  FOR ISOLATING AND QUANTITATING 2,3,7,8-TCDD
     AND 2,3,7,8-TCDF  PRESENT IN ORGANIC EXTRACTS   OF PAPER
     MILL PROCESS AND  EFFLUENT SAMPLES

     11.4.1   Preliminary Separation  of  2,3,7,8-TCDD  and
             2,3,7,8-TCDF  From  Other  Chemical  Residues in
             the Extracts  Obtained As  Described in Sections
             11.2 and 11.3.   Organic extracts obtained
             utilizing the  procedures  described in Sections
             III. and  IV. are subjected to the fractionation
             procedures which follow.

             11.4.1.1  Add   30  mL  of  aqueous  potassium
                      hydroxide  (20%  w/v)  to the bottle
                      containing  the  sample extract,   seal
                      the  bottle and agitate  it for  a
                      period  of  10  minutes  on  a wrist
                      action shaker.   Aspirate and discard
                      the  aqueous phase,  retaining the
                      organic phase.

             11.4.1.2  If   the  aqueous   layer  from  the
                      previous  step  appears  to  be  colored
                      following  the   base  extraction
                      procedure, then repeat this operation
                       (Step 11.4.1.1).
                      27      •}<
398

-------
        11.4.1.3  Add 30 mL  of double-distilled water
                  to  the  organic  phase  from  Step
                  11.4.1.1,  seal the bottle, and agitate
                  the mixture for  a  period of 1 minute.
                  Again,  aspirate  and discard  the
                  aqueous phase, retaining the organic
                  phase.

        11.4.1.4  Add 30 mL  of  concentrated sulfuric
                  acid to  the  residual  hexane  extract
                  from the previous step, seal  the
                  bottle,  and agitate it  for  a period
                  of 10 seconds.   If  emulsions  form,
                  centrifuge  the bottle  to achieve
                  separation  of the organic and acidic
                  aqueous phases.    Remove  and  discard
                  the aqueous  acidic  layer, retaining
                  the organic layer.

        11.4.1.5  Repeat  the  concentrated sulfuric acid
                  wash (Section 11.4.1.4)  this  time .
                  adding  30 mL of  sulfuric acid to the
                  sample extract,  and  agitating  the
                  acidified sample for  10 minutes.
                  Again,  aspirate  and discard  the
                  aqueous  layer.   Repeat  this  step
                  until  the  acid  layer  is  visibly
                  colorless.

        11.4.1.6  Repeat  Step 11.4.1.3.

        11.4.1.7  Add 5 g  of anhydrous  sodium sulfate
                  to the  organic extract and allow the
                  mixture  to stand  for at least  15
                  minutes.

        11.4.1.8  Quantitatively transfer  the  organic
                  extract,  using  hexane to rinse  the
                  sample  bottle, to a clean test tube,
                  and    reduce   the    volume    to
                  approximately  5 mL  by  passing  a
                  stream  of pre-purified nitrogen over
                  the extract,  while  maintaining  the
                  test tube  at 55° C in  a water  bath
                  (nitrogen slowdown apparatus).

        11.4.1.9  Proceed  with  the  liquid   column
                  chromatographic  procedures described
                  in Section  11.4.2.

11.4.2  Liquid  Column Chromatographic  Procedures  for
        Isolating  2,3,7,8-TCDD and  2,3,7,8-TCDF  From
        Extracts Previously Washed With Acids and Bases
                 28

-------
11.4.2.1  Fabricate  a   silica   gel   glass
          chromatography-coluron.  Pack  the
          column,  in succession, with a plug of
          glass wool (silanized),  1.0 g silica,
          2.0 g silica containing  28%  (w/w) 1 M
          NaOH, 1.0 g silica,  4.0  g  silica
          containing 30%  (w/w)  sulfuric acid,
          and 2.0  g silica.

11.4.2.2  Quantitatively    transfer    the
          concentrated extract obtained in Step
          11.4.1.8,  along with two rinsings of
          the  sample container, using 1  mL
          portions of hexane each time,  to the
          column and elute  the  column with 90
          mL of hexane.   Collect the  entire
          eluate and concentrate to a volume of
          1-2 mL in a centrifuge tube.

11.4.2.3  If any layer of the  silica gel  column
          implemented in Step 11.4.2.2 becomes
          visibly colored as  the column  is
          eluted, repeat Steps  11.4.2.1  and
          11.4.2.2.

11.4.2.4  Prepare  a  liquid  chromatography
          column by packing  the  constricted end
          with a  plug of silanized glass wool
          and then adding three grams of Woe 1m
          basic alumina.

11.4.2.5  Aspirate  the   concentrated  extract
          obtained  in   Step  11.4.2.2  and
          transfer it onto  the  alumina  column
          prepared in Step 11.4.2.4.  Rinse the
          test  tube  which  contained  the
          concentrate successively with  two 1
          mL portions of hexane,  each time
          transferring the  rinse solution to
          the alumina column.

11.4.2.6  Elute the alumina column as  follows:
          (a)  Elute the  alumina column with 10
          mL of 3% (v/v)  methylene chloride-in-
          hexane, taking care not to  let the
          column  become  completely  dry  during
          the elution, and  discard the  entire
          eluate.  (b)  Elute the column with 15
          mL of  20% (v/v) methylene  chloride-
          in-hexane and discard the  entire
          eluate.  (c)  Elute the column with 15
          mL of  50% (v/v) methylene  chloride-
          in-hexane, retain  this entire eluate,
          and reduce the  volume  to about  1.0 mL
          by passing a  stream of  pre-purified
         29
                400

-------
          nitrogen  over the  solution  while
          heating the  solution in a  55° C water
          bath.

11.4.2.7.  Prepare  a  second alumina  column as
          described  in  Step  11.4.2.4,  transfer
          the  concentrated  eluate  obtained in
          Step 11.4.2.6 to the column,  and
          elute the column as described in  Step
          11.4.2.6.  Collect  the  column  eluate
          and  concentrate it  to a  volume of
          about 1 mL.

11.4.2.8   Prepare  a  liquid  chromatography
          column  by  cutting  off  a 9-inch
          disposable Pasteur pipette  1.25 cm
          above the  tip constriction leaving a
          straight   glass   tube   with   an
          indentation approximately  2.5 cm  inch
          from  the  top. Insert a filter paper
          disk  in the  tube  and position  the
          disk  2.5  cm below the indentation.
          Add  a  sufficient quantity of  PX-21
          Carbon/Celite  545   (prepared  as
          described  in Section  7.1.8) to the
          tube  to form a  2  cm length of  the
          Carbon/Celite.   Insert a  glass  wool
          plug  on top  of  the Carbon/Celite.
          Pre-elute  the  column sequentially
          with 2 mL  of  a  50% benzene/50% ethyl
          acetate solution  (v/v) , 2 mL of 50%
          methylene  chloride/50% cyclohexane,
          and 2 mL of hexane, and discard these
          eluates.   Transfer  the residual
          sample  extract (in 1  roL  of  hexane)
          resulting  from  the  alumina column
          cleanup  (Step 11.4.2.7) onto the top
          of  the  Carbon/Celite  column,  along
          with  1  mL of  a  hexane rinse of the
          original  sample vessel.   Elute the
          column  with  2 mL  of  50%  methylene
          chloride/50%  cyclohexane solution and
          2 mL of 50% benzene/50% ethyl acetate
          and  discard these eluates.   Invert
          the column  and  elute   it  in  the
          reverse  direction with  4  mL of
          toluene,  retaining  this  eluate.
          Concentrate the  collected column
          effluent  to  a volume  of  about 1 mL
          using  a   stream  of  pre-purified
          nitrogen.

11.4.2.9   Prepare  a third  alumina  column as
          described  in  Step  11.4.2.4,  transfer
          the concentrated  eluate from the
         30         401

-------
                      second  alumina column sequence to
                      this  column,  elute the column, and
                      collect  the   eluate  as   already
                      described in Step 11.4.2.6,  in  a test
                      tube.  Concentrate  the collected
                      eluate  to a volume  of about  1 mL,
                      then quantitatively transfer the
                      concentrate to  a  3 mL micro-reaction
                      vessel,  using  two 1 mL portions of
                      methylene chloride to rinse  the  test
                      tube, and also  transferring  these to
                      the     micro-reaction     vessel.
                      Concentrate  the solution in the
                      latter vessel just to  dryness,  using
                      a  stream of  dry N2 ,  as  described
                      previously.  Rinse  the walls of the
                      micro-reaction vessel using 0.5  mL of
                      methylene   chloride,  and  again
                      concentrate just  to    dryness.   Seal
                      the vessel and store  it  in a freezer
                      (-15°C).    Just prior  to  GC-MS
                      analysis, remove  the  vessel  from the
                      freezer, allow  it to  warm  to ambient
                      temperature,  and reconstitute the
                      residue in the vessel by adding  10 uL
                      of Standard External  Standard to the
                      vial.

11.5 GAS   CHROMATOGRAPHIC - MASS  SPECTROMETRIC   (GC-MS)
     PROCEDURES  FOR    QUANTITATING 2,3,7,8-TCDD AND
     2,3,7,8-TCDF PRESENT IN SAMPLE EXTRACTS

     Sample  extracts prepared by the procedures described in
     the  foregoing  are  analyzed  by  GC-MS utilizing the
     instrumentation and  operating parameters  listed  below.
     Typically, 1  to  5  uL portions  of the extract are
     injected  into  the GC.   Sample extracts  are  initially
     analyzed  using  the  DB-5 capillary GC column at a mass
     spectral  resolution of 1:600 to obtain data on the
     concentration  of  2,3,7,8-TCDD  and to ascertain if
     2,3,7,8-TCDF  or other isomers which coelute with
     2,3,7,8-TCDF are  present.   If the  latter are  detected
     in this analysis, then another aliquot  of the sample is
     analyzed  in a separate run,  using a"^ newly  developed
     hybrid  column which  consists of  a 10 meter length of  a
     0.25  mm I.D.  fused  silica  open  tubular  DB-5   capillary
     column  coupled  with a 30 meter  section of  a  0.25 mm
     I.D.  DB-225 column. Again,  the  mass spectrometer is
     operated  at low resolution  (1:600).  The  hybrid  column
     uniquely  separates  2,3,7,8-TCDF from the  other 37 TCDF
     isomers and therefore  yields  definitive data on the
     concentration of  2,3,7,8-TCDF  in the extract  which is
     analyzed. However,  in some instances compounds are
     present in the sample  extract which  give rise to ion
     masses  which, at low (1:600) mass resolution,  interfere
                     31
402

-------
with  the quantitation of 2,3,7,8-TCDF.   In  these
instances the  analysis of the sample extract  can be
repeated, using the DB-5/DB-225 hybrid column, but this
time  at  a mass  spectral  resolution  of  1:6,500.   The
instrumentation  and  operating parameters utilized  in
these analyses are as follows.

11.5.1  Gas   Chromatograph:     Carlo  Erba Mega 5000  or
        Varian 3740

        11.5.1.1  Injector:    Configured  for capillary
                  column,  splitless/split   injection
                  (split flow on  60  seconds following
                  injection):  injector temperature,
                  280°C.

        11.5.1.2  Carrier gas:

                  For DB-5 column:   Hydrogen,  30 Ib.
                  head pressure (MS-25 jet  separator);
                  18 Ib. head pressure  (MS-30  direct
                  coupled)

                  For DB-5/DB-225  column:  Hydrogen, 30
                  Ib.   head  pressure   (MS-25  jet
                  separator);  18  Ib.  head pressure  (MS-
                  30 direct coupled)

        11.5.1.4  Capillary Column 1:  For quantitation
                  of 2,3,7,8-TCDD  (isomer specific) and
                  2,3,7,8-TCDF  (non-isomer   specific),
                  60 M x 0.25 mm  ID fused silica  coated
                  with  a  0.25 micron film of  DB-5,
                  temperature programmed, see  Table  1
                  for  temperature program.   Capillary
                  Column 2:    For  quantitation  of
                  2,3,7,8-TCDF (isomer specific), 10  M
                  x  0.25  mm  ID  fused  silica  column
                  coated with a 0.25  micron  film  of DB-
                  5 coupled with a 30 M x 0.25 mm I.D.
                  fused  silica  column  coated with  a
                  0.25  micron  film of  DB-225.   This
                  column is  temperature  programmed as
                  indicated in Table  2.

        11.5.1.4  Interface Temperature:   250°C

11.5.2  Mass Spectrometer:   Kratos MS-30  or  Kratos MS-25
                            or     other     suitable
                            instruments

        11.5.2.1  lonization Mode: Electron impact  (70
                  eV)
                 32
                          403

-------
                  11.5.2.2   Static  Resolution:  1:3000  to 1:10,000
                            (10%   valley)    depending    upon
                            instrumentation.

                  11.5.2.3   Source  Temperature:   250°C

                  11.5.2.4   Accelerating Voltage:  2KV or 4KV,
                            depending  upon  instrument.

                  11.5.2.5   Ions  Monitored:  Computer controlled
                            Selected Ion  Monitoring, See Tables  1
                            and  2  for  list  of  ion   masses
                            monitored  and  time intervals   during
                            which ions characteristic  of 2,3,7,8-
                            TCDD  and 2,3,7,8-TCDF are monitored.
                            Note  that  in  the  case of quantitation
                            of    the    2,3,7,8-TCDF,     the
                            hexachlorinated     diphenylether
                            molecular  ion,  which could give rise
                            to an  interference  at  m/z 304 and
                            306,  is also monitored  as indicated
                            in Table 2.
12.  CALCULATIONS
     12.1 EQUATIONS USED FOR CALCULATING ANALYTICAL RESULTS  FROM
          THE GC-MS DATA

          12.1.1   Equation  1:   Calculation of  Relative  Response
                                 Factor for  native 2,3,7,8-TCDD
                                 (RRF2)  using »3Ci2-2,3,7,8-TCDD
                                 as  an internal  standard.

                             RRF2  =     (AsCia /AisCs )

                   where:     Aa    =  SIM  response for  2,3,7,8-
                                    TCDD ion at  m/z 320 + 322

                             Ai 3   =  SIM  response  for   * 3 Ci 2 -
                                    2,3,7,8-TCDD     internal
                                    standard ion at m/z 332 + 334

                             Cia   =  Concentration of the  internal
                                    standard (pg./uL.)

                             C8    =  Concentration of the  2,3,7,8-
                                    TCDD (pg./uL.)
                                    404

-------
12.1.2   Equation  2:  Calculation of Relative Response
                       Factor  for   l3Ci2-1,2,3,4-TCDD
                       (RRFi, )

         RRFb  = (Ai s Ces /Acs Ci s )

   where:  At8 =  SIM response for 13Ci2-2,3,7,8-TCDD
                  internal standard ion at m/z 332 + 334

           Acs =  SIM  response for *3Ci2-1,2,3,4-TCDD
                  external standard at m/z 332 + 334

           Cia =  Concentration  of the  *3Ci2-2,3,7,8-
                  TCDD internal standard  (pg./pL.)

           Ces =  Concentration of  the  *3Ci 2-1,2,3,4-
                  TCDD standard (pg./uL.)
12.1.3   Equation 3:   Calculation  of concentration of
                       native    2,3,7,8-TCDD     using
                       13Ci2-2,3,7,8-TCDD  as  internal .
                       standard

        Concentration, pg./g. =  (As) (Is)/(At9)(RRF2)(W)

   where:    As  = SIM response  for   2,3,7,8-TCDD ion
                   at m/z 320 +  322

             Ai s = SIM response  for the  *• 3 Ci 2 -2, 3 ,7 , 8-
                   TCDD internal standard  ion at  m/z
                   332 + 334

             Is  = Amount of internal standard added to
                   each sample  (pg.)

             W   = Weight of sample in grams

          RRF2   = Relative   response   factor    from
                   Equation 1
                 34

                         405

-------
12.1.4  Equation  4:   Calculation  of % recovery  of
                       13Ci2-2,3,7,8-TCDD    internal
                       standard

        % Recovery = 100 (At s ) (Es ) / (A« » ) (It ) (RRFb )

            Ais   = SIM response  for »3Ci2-2,3,7,8-TCDD
                   internal standard  ion   at  m/z 332
                   + 334

            Aes   = SIM response for l3Ci2-1,2,3,4-TCDD
                   external standard ion at  m/z 332 +
                   334

            EB    = Amount of 13Ci2-1,2,3,4-TCDD external
                   standard  co-injected   with   sample
                   extract

            It    = Theoretical amount of l 3 Ci 2 -2, 3 , 7 , 8--
                   TCDD internal standard in injection

         RRFb    = Relative   response   factor    from •
                   Equation 2
12.1.5   Equation  5:  Calculation of Relative Response
                       Factor for  native  2,3,7,8-TCDF
                       (RRFc)  using *3Ci2-2,3,7,8-TCDF
                       as an internal standard.

            RRFc = (As Ci s /Ai s Cs )

  where:    As    = SIM  response  for 2,3,7,8-TCDF  ion
                   at m/z 304 + 306

            Ais   = SIM  response for *• 3 Ci 2 -2 , 3 , 7 , 8-TCDF
                   internal  -standard  ion at m/z 316 +
                   318

            Cis   = Concentration   of   the     internal
                   standard  (pg./uL.)

            Cs    = Concentration   of  the   2,3,7,8-TCDF
                   (pg./uL.)
                 35
                         4GG

-------
    12.1.6  Equation 6:  Calculation  of  Relative  Response
                         Factor  for l3Ci2-1,2,3,4-TCDF
                         (RRFd)  (When analysis is performed
                         using DB-5 Column)
      RRFd

where:  Ai a
=  (Al aCes /Aes Ct s )

=  SIM response for 13Ci2-2,3,7,8-TCDF
   internal standard ion at m/z 316 + 318
        Acs =  SIM  response for »3Ci2-1,2,3,4-TCDD
               external standard at m/z 332 + 334

        Cj8  = Concentration of  the 13Ci2-2,3,7,8-TCDF
               internal standard  (pg./uL.)

        Ces  = Concentration of  the *3Ci2-1,2,3,4-TCDD
               external standard  (pg./uL.)
    12.1.7  Equation  7:
     where
 RRFc

  Ais




  Acs



  Cis
               -e s  =
    Calculation of Relative Response
    Factor   for   37C«-1,2,7,8-TCDF
    (RRFe)     (When    analysis   is
    performed   using  .  DB-5/DB-225
    Hybrid Column)

(Al 3  Ces /Aes Cls )

SIM response  for l3d2-2,3,7,8-TCDF
internal standard ion  at  m/z 316 +
318

SIM  response for 37Cl«-1,2,7,8-TCDF
external standard at m/z 312

Concentration  of the l3Ci2-2,3,7,8-
TCDF internal standard (pg/uD

Concentration  of the 37Cl«-1,2,7,8-
TCDF external standard (pg/uD
                     36
                              407

-------
12.1.8  Equation 8:   Calculation  of concentration  of
                      native 2,3,7,8-TCDF  using l3Ci2 -
                      2,3,7,8-TCDF as internal standard

 Concentration, pg./g. = (As) (Is)/(Aja)(RRFC)(W)

   where:    A8  = SIM response for 2,3,7,8-TCDF ion at
                   m/z 304 + 306

             Aia = SIM response for  the J3Ci2-2,3,7,8-
                   TCDF internal standard ion at m/z 316
                   + 318

             Is  = Amount of internal standard added to
                   each sample  (pg.)

             W   = Weight of sample in grams

          RRFc   = Relative   response   factor    from
                         Equation 5

12.1.9  Equation  9:  Calculation  of % recovery  of
                      13Ci2-2,3,7,8-TCDF  internal
                      standard  (When analysis is
                      performed using DB-5 Column)

      % Recovery = 100(Ai.) (E. )/(A..) (Ii) (RRF*)

            Ais  = SIM response  for 13Ci2-2,3,7,8-TCDF
                   internal  standard  ion at m/z 316 +
                   318

            Acs  = SIM response  for *• 3 Ci 2 -1, 2 , 3 , 4-TCDD
                   external standard  ion  at m/z 332 +
                   334

            E8   = Amount of -1 3 Ci 2-1, 2 , 3 , 4-TCDD  external
                   standard   co-injected  with   sample
                   extract

            Ii   = Theoretical amount of * 3 Ci 2-2 , 3 , 7 , 8-
                   TCDF internal standard in injection

         RRFd     = Relative   response   factor    from
                   Equation 6
                 37         408

-------
          12.1.10  Equation 10:  Calculation   of   %  recovery  of
                                 13Ci2-2,3,7,8-TCDF      internal
                                 standard   (when   analysis   is
                                 performed    using  hybrid DB-5/
                                 DB-225 column

             % Recovery   =  100 (Ai s )  (E8)/(Aes)  (Ii)  (RRFe )

                  Ai 3     = SIM response  for  13Ci2-TCDF  internal
                             standard ion at m/z 316  +  318

                  Aes      =  SIM  response for 37Cl4-1,2,7,8-TCDF
                             external standard ion at m/z 312

                  Es       =  Amount of 37Cl4-1,2,7,8-TCDF external
                             standard in injection

                  RRFe    =  Relative    response   factor   from
                             Equation 7

     12.2 REPORTING RESULTS

          The  2,3,7,8-TCDD  and 2,3,7,8-TCDF data  for solid'
          samples are reported on a dry weight basis  in picograms
          of analyte  per gram  of solid material,  that  is, parts-
          per-trillion.   The  2,3,7,8-TCDD  and 2,3,7,8-TCDF data
          for  liquid samples are reported in  femtograms of
          analyte per gram  of liquid  sample  (wastewater or
          water), that is, parts-per-quadrillion.
13.  INTERPRETATION OF RESULTS

     13.1 Spectral  responses must be  observed  at both the
          molecular and fragment ion masses  corresponding to the
          ions indicative of  TCDD  and  TCDF  (see Tables 1 and  2)
          and intensities of these ions must maximize essentially
          simultaneously  (within + 1 second).   In  addition, the
          chromatographic retention  times observed  for  2,3,7,8^
          TCDD and  2,3,7,8-TCDF  must be correct relative to the
          appropriate  stable-isotopically labelled  internal
          standard.

     13.2 The ratio of the intensity of  the  response  for the
          molecular ion,  [M]*, to the  response for the [M+2]*
          ion must  be within  ±15% of the  theoretically  expected
          ratio for both  the  native  TCDD  and native TCDF  signals
          (for  example,  0.77 in  the  case of  TCDD  and TCDF;
          therefore,  the  acceptable range  for  this  ratio  is  0.65
          to 0.89).

     13.3 The intensities of  the ion signals for either  2,3,7,8-
          TCDD or 2,3,7,8-TCDF are considered to be detectable  if
          each exceeds the baseline noise by a  factor of  at  least
          2.5:1.
                           38
                                   409

-------
     13.4  For reliable detection  and quantitation of 2,3,7,8-
          TCDF,  it  is  also necessary  to monitor the molecular ion
          of  hexachlorinated diphenyl  ether which,  if  present,
          could  give rise to  fragment ions yielding  ion  masses
          identical to those monitored as indicators of the TCDF.
          Accordingly,  in Tables  1  and 2,  the  appropriate  ion-
          mass  for  hexachlorinated  diphenyl ether  is  specified
          and this  ion-mass must be monitored simultaneously with
          the 2,3,7,8-TCDF ion-masses.   Only when  the  response
          for the diphenyl  ether ion-mass  is not detected  at the
          same  time as the 2,3,7,8-TCDF ion mass  can the  signal
          obtained  for 2,3,7,8-TCDF be considered unique.
14.   PRECISION AND  ACCURACY

     14.1 This analytical protocol was  found  to  be satisfactory
          for isomer-specific  determinations  of 2,3,7,8-TCDD and
          2,3,7,8-TCDF  in a variety  of selected pulp  and paper
          mill  sample  matrices.   Intra-laboratory  method
          validation  experiments for pulp, sludge, and wastewater
          samples indicate that the performance of the analytical
          method with respect  to  precision and spike recovery is
          demonstrably uniform.    The method performance does not
          appear to  be  sensitive to  any specific  matrix  or
          chemical  effects which  might be associated with the
          manufacturing  processes  at a  given mill.

     14.2 Laboratory precision  for the method  expressed as
          relative  percent difference  between duplicate analyses
          for thirty-five 2,3,7,8-TCDD determinations was 15
          percent  mean  (range  1-138  percent);  and for thirty-
          three 2,3,7,8-TCDF  determinations, 16  percent  mean
          (range 0-62 percent).

     14.3 Field precision for  eight  2,3,7,8-TCDD determinations
          was 14 percent mean  (range  4-19 percent) ; and for nine
          2,3,7,8-TCDF  determinations,  22 percent mean (range 0-
          99  percent).

     14.4 For thirty-five 2,3,7,8-TCDD determinations,  accuracy
          expressed as  percent spike  recovery was 103 percent
          mean  (range  66-168 percent); and for  thirty-five
          2,3,7,8-TCDF  determinations,   102  percent  mean  (range
          58-153 percent).

     14.5 Including results  from  method validation experiments,
          97  percent  of  the analyses for a previous study met the
          quality  assurance  objectives  for  laboratory precision
          and   accuracy.      Ninety-five   percent  of   133
          determinations for  2,3,7,8-TCDD  and  for 2,3,7,8-TCDF
          resulted in  analytical data suitable  for project
          objectives.
                           39

                                  410

-------
     14.6 Target analytical detection levels of 1 ppt  for  solid
          samples  were achieved  for all  but  one sample  for
          2,3,7,8-TCDD and all  but one sample  for  2,3,7,8-TCDF
          (different  samples)  in a  prior  study.    Target
          analytical  detection levels  of  0.01 ppt  for liquid
          samples were achieved for all but three  samples  for
          2,3,7,8-TCDD and all  but  two  samples  for  2,3,7,8-TCDF
          (different samples).


15.   REFERENCES

     1.    "Total Suspended  Solids Determination," Method  209C,
          Standard  Methods for  the Examination  of Water  and
          Wastewater,  17th Edition.  APHA, AWWA,  WPCF,  1986.

     2.    Tiernan,  T.  O.,  Garrett,  J. H., Solch, J. G.,  Wagel, D.
          J.,  VanNess,  G. F-.  and Taylor, M.  L., "Improved
          Separation Procedures for Isolating TCDD and  TCDF from
          Chemically-Complex  Aqueous and  Solid Sample  Matrices
          and for Definitive Quantitation of These Isomers at PPQ
          to PPT Concentrations,"  Chemosphere, 1988 (In  Press).
                           40       411

-------
                                                                TULS 1

                              SIQUUCl Of OPKHTIOK II 6C-K-JS QUUTITHIOI Of 2,3,1,8-TCD? 119 2,3,7,8-TCDD
                                             I! EITI10IUIT1L SlIPLES OSIIG 1 (0 I DB-5 COIUII
Elapsed
Tiie dial

 0.00
1.00


31.00

31.00



45.00
      Erent

Injection, Splitless

Stirt letra Progra:
steep =  350 ppi;
tiie/iiss = 0.08 sec.
CC Coloin      Teiperatnre
Teiperatare    Projra lite
    CO         CC/iis)

   180
Tin 01 split Tilre;
Bejin Teip Projrai to 240'C

Stop Tetra Projrai

Increase Colou Teip
lallisticallf Projrai
to 300'

Couence Coolinj GC Colou
Oien leiperatore to 180*
   180
   180

   240
   240
                                           300
Ipproiinate
Ions Ionitored
by lass
Spectroieter d/z)
240.5378
256.9328
303. 90H
305.8996
311.8857
315.5418
317.5388
315.85(5
321.8535
327.8845
331.5357

333.5337

Theoritical
Identity of
fragient Ion
[I-COC1]'
[1-COC1]'
[1]'
[1*2]'
[1]'
[1]'
[1+2]'
[I]'
[1+2]'
[1]'
[1]'

[1+2]'

Coiposnds latio
Ionitored of [I]
TCDF
TCDD
TCDF
TCDf
"Cl4-2,3,7,8-TCDF
"dHJ^S-TCDr
''Cu-J.J.T.S-TCDJ
TCDD
TCDD
ITCl<.-2,3,7,8-TCDD
"Cu^^.S-TCDD,
llCu-l,2,3,4-TCDD
"Cu^.T^-TCDD,
l»Ctj-l,2,3,4-TCDD
of
!':[!+.



0.77


0.77

0.77




0.77
                                                                        373.8393
                                                                                              lezaehlorodiphenyl  Ether
                                                             412

-------
                                                                 TUIE 2
                                       SEQJEItt OF OPEHTIOIS II CC-IS-DS QDUTIT1TIOI Of 2,3,1,1-TCDF
                                         ii iniioimm SUPLES OSIK i IHIID  01-5/01-225 coimu
Elapsed
Tiie din)

 0.00
      Event

Injection, Splitless

Start  Tetra Prograi:
steep  = 350 ppi;
tiie/iass - 0.08 sec.
6C Cohu
Teiperatnre
    CO

   180
Teiperatore
Projrai late
  CC/iin)
Ions loaitored
by lass
Spectroieter d/i)
                                                                         240.911!
                                                                        303.9016
                                                                        305.8936
                                                                        ni.S»97
                                                                        315.H18
                                                                        311.935!
                                                                        331.9361
                                                                        333.9337
                                                                        373.1393
Identity of
Fragient Ion
                                                     ll-COCl]'
                                                   111*
                                                   U+2]'
                                                   II]'
                                                   HI*
                                                   [1+2]'
                                                   til*
                                                   [1+2]'
                                                   [I]'
Coiponnds
lonitored
Ipproxiiate
Tbeoritical
latio of
of [I]':[1*2]'
                                                       TCDF
                                                     TCDF
                                                     TCDF                0.71
                                                     »Cl4-l,2,7,HCDF
                                                     "Cu^.l.t-TCDF
                                                     »»Cn-2,3,7,l-TCIIF   0.77
                                                     l'Cn-l,2,3,HCDD
                                                     l>Ctt-l,2,3r(-!CDD
                                                     lezacilorodipienrl Ether
 1.00


41.00

11.00
Tin 01 split talfe;              ItO
Begin Teip Projra to  220'C;      110

Stop Tetra Prograi               220

Couence Cooling Cohu           220
to 1JO«
                                                            413

-------
ncasl
                      ««.
leal bufetin
NATIONAL COUNCIL Of TH« PAPER MOUSTRY FOR AM AND STREAM «4P*WVEi«W. •«, MO MAOaONAVOJUt NEW YORK. N.Y. 1001 e


                         eC£r/^D
                         HAY19
            NCASI PROCEDURES FOR THE PREPARATION AND
        ISCMER SPECIFIC ANALYSIS OF PULP AND PAPER INDUSTRY S/m£S
             FOR 2,3,7,8-TCDD AND 2,37,8-TCDF
                 TECHNICAL BUUHIN NO, 551
                     WY1989
                414

-------
NATIONAL COUNCIL OF THE PAPER INDUSTRY FOR AIR AND STREAM IMPROVEMENT, INC.
260 MADISON AVE. NEW YORK, N.Y. 10016 (212) 532-9000
                                                           Dr. Isaiah (tollman
                                         May 1,  1989         President
                                                            (212)5329000
 TECHNICAL BULLETIN NO. 551

             NCASI  PROCEDURES FOR THE PREPARATION AND
        ISOMER-SPECIFIC ANALYSIS OF PULP AND PAPER  INDUSTRY
        	SAMPLES FOR 2.3.7.8-TCDD AND 2.3.7.8-TCDF

      On discovering the presence of chlorinated  dioxins  and
 furans in pulp mill wastewater treatment plant sludges,  the paper
 industry embarked on a wide array of investigations  into the
 source and significance of these unintentional trace materials.
 The pursuit of these studies required the development  of
 reliable, sensitive and accurate analytical procedures.   To
 provide the industry with the necessary procedures,  NCASI
 undertook a method development program with the  objective of
 making considerable analytical procedures capable  of the isomer-
 specific determination of 2,3,7,8-TCDD and 2,3,7,8-TCDF  with
 detection limits in the low parts per trillion for solid sample
 matrices and low parts per quadrillion for aqueous matrices.

      This technical bulletin describes the sample  preparation and
 analytical protocols that resulted from over two years of
 intensive testing and evaluation.  The sample preparation
 protocol insures the stability and homogeneity of  the  samples.
 The analytical procedure builds on techniques that were  developed
 as part of the USEPA/Paper Industry Cooperative  Dioxin Screening
 Study by incorporating (a) matrix-specific extraction  procedures,
 (b)  an innovative quality control step to monitor  recoveries in
 the cleanup procedure, (c) an alternative method for the isomer-
 specific determination of 2,3,7,8-TCDF, and (d)  quantitation by
 high resolution GC/MS.  The procedure has been (a) demonstrated
 to be applicable to a variety of pulp and paper  industry matrices
 and (b)  shown to give comparable data to other procedures through
 interlaboratory comparison studies.

      Development of the analytical protocol was  a  collaborative
 effort between NCASI and Enseco-CAL.  The studies  were conducted
 under the direction of Lawrence E. LaFleur, Organic  Analytical
 Program Manager, assisted by Kenneth Ramage, Theresa Bousquet and
 Robert Brunck of the NCASI West Coast Regional staff and Dr.
 Michael J.  Miille,  Enseco-CAL Division Director, who was assisted
 by William J.  Luksemberg, Steve Valmores and Bob Peterson.
                             415
                 «N«BoMi Council ol 9» P»p«r Indutty tar Air and Sawn knpTMrant he. 1969

-------
                              - 2 -
     Your comments and suggestions on this technical bulletin are
solicited and should be directed to Lawrence LaFleur,  Organic
Analytical Program Manager at the West Coast Regional  Center,
P.O. Box 458 Corvallis, OR 97339 (503-752-8801),  or to this
office (212-532-9000).
                                        Very truly yours,
                                        Isaiah Gellman
                                416

-------
              NCASI PROCEDURES FOR THE PREPARATION AND
       ISOMER SPECIFIC ANALYSIS OF PULP AND PAPER INDUSTRY
       	SAMPLES FOR 2.3.7.8-TCDD AND 2.3.7.8-TCDF	

                   TECHNICAL BULLETIN NO.  551
                            MAY  1989

ABSTRACT: This technical bulletin describes the sample
          preparation and isomer specific analysis procedures
          NCASI has developed in support of the industry programs
          investigating the source and significance of 2,3,7,8-
          TCDD and 2,3,7,8-TCDF in pulp and paper industry
          matrices.  Part I describes the sample preparation
          procedures which insure the stability and homogeneity
          of the samples.  Part II describes the matrix-specific
          extraction,  cleanup and isomer-specific GC/MS analysis
          procedures.   These analytical procedures have been
          carefully optimized and thoroughly tested to insure the
          reliability of the results.  The procedures have been
          shown to be capable of achieving detection limits in
          the low parts per trillion (generally around one ppt)
          for solid matrices and in the low parts per quadrillion
          (generally less than ten ppq)  for liquid matrices.
          Through extensive interlaboratory comparison studies
          the NCASI procedure has been shown to give comparable
          results to other procedures.   In particular, USEPA has
          recognized that the NCASI procedures are equivalent to
          the procedures used in the USEPA/Paper Industry
          Cooperative Dioxin Screening Study (the five Mill
          Study).

KEYWORDS: 2,3,7,8-TCDD, 2,3,7,8-TCDF, analytical procedure,
          isomer specific, extraction procedure, clean-up
          procedure.

RELATED NCASI PUBLICATIONS:

     (1)   U.S. Environmental Protection Agency/Paper Industry
          Cooperative Dioxin Screening Study. NCASI Technical
          Bulletin No. 545, May,  1988
                                417

-------
                      TABLE OF CONTENTS
                                                          Page

 I INTRODUCTION                                             1

   PART I - NCASI PROCEDURES FOR THE PREPARATION
            AND ISOMER-SPECIFIC ANALYSIS  OF PULP
            AND PAPER INDUSTRY SAMPLES FOR
            2,3,7,8-TCDD AND 2,3,7,8-TCDF                   3

   PART II - NCASI METHOD TCDD/F - 88.01                    12

II LITERATURE REFERENCES                                   57
                           418

-------
            NCASI  PROCEDURES  FOR THE  PREPARATION AND
       ISOMER SPECIFIC ANALYSIS OF PULP AND PAPER INDUSTRY
       	SAMPLES FOR 2.3.7.8-TCDD AND 2.3.7.8-TCDF
                         I    INTRODUCTION

     Following the announcement that 2,3,7,8-TCDD had been
detected in pulp and paper mill wastewater treatment plant
sludges, NCASI undertook an evaluation of dioxin analysis
contract laboratories to determine which,  if any, might be able
to provide the analytical support required by the industry as
investigations of these findings were undertaken.  The evaluation
was based on experience, demonstrated performance and available
resources.  Once a suitable laboratory was selected, a program
was begun to test the applicability of existing analytical
procedures for the analysis of trace levels of 2,3,7,8-TCDD and
related compounds in pulp and paper industry samples.

     Quite early, it was recognized that commonly employed
cleanup and low resolution MS analysis procedures were inadequate
for the determination of 2,3,7,8-TCDD and related compounds in
pulp and paper industry matrices at the low parts per trillion
and low parts per quadrillion detection limits which were
required.  An extensive analytical methods development program
was therefore undertaken jointly by NCASI and Enseco-Cal Labs.

     The methods development work undertaken by NCASI/Enseco Cal
was conducted concurrently with the methods development work that
culminated in the analytical protocol used in the USEPA/Paper
Industry Cooperative Dioxin Screening Study (1).   Thus, the
NCASI/Enseco Cal analytical protocol presented in Part II,
represents a combination of many features of the USEPA/Paper
Industry Dioxin Screening Study plus features which resulted
directly from the continued research conducted by NCASI and
Enseco Cal.  Although the primary objective of the methods
development program was an analytical protocol capable of
providing quality analytical data which meet all the low
detection limit and QA/QC criteria, it was also realized that the
techniques and procedures used in the NCASI procedure were more
along the lines used for other analytical protocols commonly used
by contract laboratories.  Thus, it is believed that other
laboratories experienced in high resolution gas chromatographic/
high resolution mass spectrometric methods for analyses for
2,3,7,8-TCDD should be readily able to perform this procedure.

     The NCASI procedure is similar to the USEPA/Paper Industry
Cooperative Dioxin Screening Study procedure in that it
incorporates the techniques of (a) spiking the internal standards
into the aqueous samples using a water miscible solvent, and  (b)
filtration and soxhlet extraction of the filtered solids as part
                              419

-------
                              - 2 -

of the wastewater extraction procedures.  Studies conducted
parallel to, and following, the USEPA/Paper Industry Cooperative
Dioxin Screening Study developed matrix-specific extraction
procedures for solid samples to further optimize extraction
efficiencies (2).  Enseco Cal was able to demonstrate an
alternative isomer-specific procedure 2,3,7,8-TCDF which utilized
a 30 M DB-225 column.  The point of addition of 37Cl4-2,3,7,8-TCDD
was changed to be immediately prior to any cleanup procedures.
This .innovative use of the  CI^-2,3,7,8-TCDD provides additional
insight into the extraction efficiency and is a useful QC tool
for troubleshooting analytical problems.  The GC/MS quantitation
procedures were intentionally selected to follow along the lines
of generally accepted techniques for high resolution GC/MS dioxin
analysis procedures.  This facilitates the use of the procedure
by other contract laboratories and assures the acceptance of the
procedure by regulatory agencies.

     The objectives of this technical bulletin are to describe
(a) the NCASI procedures for the preprocessing and homogenization
of samples intended for PCDD/PCDF analysis, and (b) the
NCASI/Enseco Cal analytical procedures themselves.  Thus, the
technical bulletin is organized into two parts, each presenting
the details relevant to these objectives.  By making these
procedures available to other laboratories, it is hoped that this
will facilitate expansion of the number of laboratories capable
of analyzing pulp and paper industry samples.  This will in turn,
enhance the ability of the pulp and paper industry to conduct its
many programs investigating the source, potential control and
significance of the trace levels of chlorinated dioxins and
furans associated with the production of bleached pulp.

     A summary of the performance characteristics of the
procedure is in preparation.  Included will be discussions of (a)
accuracy as demonstrated by matrix spike recovery data and
interlaboratory comparisons, and  (b)" precision as measured by
routine laboratory duplicates and intra- and inter- batch
precision experiments.  In the interest of making a procedure,
which has been carefully optimized, validated and thoroughly
tested, immediately available for use by the industry, it was
decided that it would be best to publish the analytical
procedures as soon as possible, to be followed by distribution of
a second report once the performance characteristics have been
summarized.
                            42U

-------
                              - 3 -

   PART I - NCASI DIOXIN PROGRAM SAMPLE PREPARATION/PROCESSING

                      I   SAFETY GUIDELINES

     The analyst should be familiar with the location and proper
use of all safety equipment throughout the building (e.g. fire
extinguishers, respirators, spill kits, etc.).   Recommendations
on the use of specific safety precautions are referenced, where
appropriate, in the following procedures.  These recommendations
include the use of fume hoods for solvents or the processing of
samples with nuisance odors and dust masks and/or glove boxes to
prevent the inhalation of particulate matter.


                   II   GENERAL LAB PROCEDURES

     Under no circumstances should a sample be touched, stored or
in any way come in contact with any materials other than those
prescribed below and then only after they have been properly pre-
pared.  Aluminum foil or unpowdered latex gloves require no pre-
treatment but fresh foil or a new pair of gloves should be used
for each situation.


                    Ill   CLEANING PROCEDURES

A.   Solvent Cleaning

     All materials (except aluminum foil and latex gloves) which
come in contact with the sample (restricted to glass, stainless
steel and Teflon) shall be solvent cleaned.  Only Teflon squeeze
bottles are to be used.

     The following general cleaning procedure will be followed:

1.   Soap and tap water wash all items using Pierce RBS-35 soap
     (20 mL RBS-35 per liter of tap water).  Rinse with tap water
     followed by deionized water.

2.   Methanol (Burdick and Jackson) rinse.

3.   Acetone (Burdick and Jackson)  rinse.

4    Methylene chloride (Burdick,and Jackson) rinse.

5.   Air dry.

     Used solvents should be stored in separate bottles marked
"Used Methanol," "Used Acetone," and "Used DCM."  Conduct solvent
rinsing in a hood.

                                               Revision 4, 6/1/88
                               42.1

-------
                              - 4 -


B.   Glove Box Cleaning Procedure

     The following cleaning procedure should be used prior to and
between each sample when using the glove box for sample grinding
or sample splitting:

1.   Vacuum all interior surfaces of the glovebox.

2.   Spray all interior surfaces with tap water and wipe down all
     inside surfaces with a wet sponge or squeegee.

3.   Use a sponge to remove excess H2O from floor of  glove  box.
     If necessary an electric blow dryer can be used to speed up
     the drying process.

4.   The neoprene glove box sleeves will be vacuumed, wet wiped
     with a sponge and air dried.  A clean pair of latex gloves
     will be placed over them prior to processing any sample.

C.   Cleaning of Cabinets and Fume Hoods Used For Drying

     Drying cabinets and fume hoods should be cleaned between
usage by vacuuming, wiping all interior surfaces with a wet
sponge or sponge mop, and then should be left to air dry.  The
vent in the drying cabinet will be wiped clean with a wet sponge
monthly.  More frequent cleaning of the vent is required if the
analyst observes accumulated dust or particulate between clean-
ings .

D.   Cleaning of Blender Motor

     The blender motor should be dismantled and cleaned monthly
or any time the blender is dismantled for maintenance.

E.   Wiley Mill Cleaning Procedure

     The following cleaning procedure should be used prior to and
between each sample when using the Wiley Mill for sample
grinding.

1.   Vacuum all outside surfaces of the mill.  Remove the glass
     plate, stationary blades and rotor blade.

2.   Vacuum the chamber, hopper inlet, and blade slots.  Rinse
     all areas with deionized water and scrub with a sponge or
     sponge swab.  Rinse using Methanol, Acetone and Hexane
     catching waste in a 25 mL beaker.

3.   Dry using a forced air heater.

                                               Revision 4, 6/1/88
                         42


-------
                              -  5-


4.   Place small metal disc and screw in a stainless steel screen
     basket for cleaning.  Use solvent cleaning procedure (Section
     III) .

5.   Reassemble the mill.  Check alignment of the blades by plac-
     ing a strip of aluminum foil over each stationary blade and
     turning rotor shaft clockwise.  Adjust the stationary blades
     so they touch the foil but don't cut it.

F.   Laboratory Cleaning

     The analyst should observe daily the general cleanliness of
the laboratory and look for accumulations of dust or particulate
in the room.  If necessary wipe surfaces with a wet sponge and
maintain an uncluttered work area.  Every two weeks a more tho-
rough inspection and cleaning of the laboratory should be con-
ducted and all possible work surfaces should be cleaned with a
wet sponge.


                       IV   RECORD KEEPING

     All processing of any dioxin samples should be recorded in
an appropriate laboratory notebook in indelible ink.  The proces-
sor must date and initial each entry corresponding to a
processing step.


                       V   SAMPLE HANDLING

     A fresh pair of unpowdered latex gloves should be used for
each sample and should be discarded after use.  Reasonable
efforts should be taken to protect samples from direct light.
Thus the lights should be turned off in cabinets and fume hoods
used for drying samples except when required for handling and
inspection.  A dust mask should be worn during any processing
where the inhalation of particulate matter is possible (e.g.
during grinding of samples).  Samples producing a nuisance odor
should be handled with proper ventilation.  Samples with an
obvious chlorine odor should be prepared and dried in a fume
hood.


   VI   SOLID/SEMI-SOLID SAMPLE PREPARATION (DRYING TECHNIQUES)

     The following is a general procedure for processing samples
that require drying.  All air drying of samples must be done in a
drying cabinet or fume hood.  When air drying samples the cabinet
or hood should be turned on and the doors closed.  During the

                                               Revision 4, 6/1/88
                               4 - M
                               i f«*/ »_*

-------
                              - 6 -

evenings when the janitors are scheduled to come in or when acti-
vity in the room may increase particulate levels in the air, turn
the cabinets or hoods off with the doors closed.  Samples are
placed in the cabinets or hoods beginning with the top shelf
until all shelves are full.  If samples dry at varying rates no
additional samples will be added until the last sample is dry and
the cabinet or hood is cleaned.  The samples on each shelf are
segregated by a physical barrier.

A.   Blanks for Drying

     An 8" x 10" Gelman glass fiber filter type A should be
placed in the center of the cabinets or hoods for drying prior to
placing samples in the cabinet.  The filter sheet should not be
pre-treated.  Place the filter sheet on a piece of aluminum foil,
edges folded up and label the foil with the date and time exposed
in the laboratory.  Barriers should separate the blank from sam-
ples on the same shelf in a manner analogous to the way samples
are segregated.  At the conclusion of the drying of all the sam-
ples in the cabinet or hood, the blank should be folded so as to
cover the exposed upper surface and should be wrapped in aluminum
foil until it is blended.  Just prior to blending, the blank fil-
ter should be torn into small pieces and placed in the blender.
Blend as described in Section VIII.  Wrap the entire blended
blank filter (i.e. do no split the blended filter) in aluminum
foil and place the foil packet into an I-Chem bottle.  Do not
label the blank until it has been put into the I-Chem bottle.
Record the blank preparation, cabinet number, glove box number,
the dates exposed, blended and bottled, and sample identification
number assigned.

B.   Sample Typing and Preparation

     Sample types fall into the following categories as
identified on the analysis request form which should accompany
each sample set.  Samples received without an analysis request
form should be referred to the supervisor for further
clarification.

(1)  Total Sample - Refers to the entire sample.  The liquid and
     solid phases are thoroughly mixed to form a homogeneous mix-
     ture.  Proceed with Section VI, Part C.

(2)  Solids Only - Refers to the solid phase only.  The analyst
     squeezes, or decants, and centrifuges the sample to remove
     and discard the liquid phase.

     Friable samples such as pulp mats should be removed from the
sample jar and hand squeezed to remove as much water as possible,
discarding the water.

                                               Revision 4, 6/1/88


                       ^v  424.'    •

-------
                              - 7 -


     For slurry type samples, decant any liquid phase observed
after settling.  Use a clean stainless steel spatula to transfer
the remaining semi-solid to precleaned centrifuge tubes.  Centri-
fuge the sample for three to five minutes at full speed, decant
and discard the supernatant.  Transfer the solid material as per
Section VI, Part C.

C.   Drvina Samples

     After typing the sample and dewatering if necessary, the
analyst should visually inspect the samples moisture content,
texture, etc., and thereby determine if the sample would be amen-
able to drying on screens, pie plates or sheets of foil as des-
cribed below.  If the examination reveals obvious inert materials
(e.g. rocks) the entire sample should be weighed and those mater-
ials greater than 1/4 inch diameter should be removed, record the
weight and calculate the percent inert material.  Discard the
inert material.

(1)  Screens (moderate moisture and large pieces or corrosive
     samples) - Break the sample into small pieces (about dime
     size) and lay out on a stainless steel screen supported
     about 1 cm above a sheet of aluminum foil and transfer to a
     drying cabinet.  The size of the foil should at least equal
     the area of the screen to catch any fines that may fall
     through.  Wooden dowels wrapped in fresh aluminum foil can
     be used to support the screen over the foil.  Save the
     sample bottle, leaving the cap off until the inside moisture
     evaporates, for NCASI sample archives.

     Label the foil with the sample identification number and the
     time and date the sample was laid out in the cabinet to dry.
     This information and the drying cabinet number should also
     be recorded.

(2)  Pyrex Pie Plates (Hicrh Moisture Slurry Type Samples and/or
     Fine Particles) - The sample is transferred to a properly
     cleaned pyrex pie plate(s) and placed in the drying cabinet.
     Label the plate(s)  with the sample identification number and
     the time and date.   This information and the drying cabinet
     number should also be recorded.

(3)  Aluminum Foil (Low Moisture and/or Fine Particles) - Spread
     the sample on a sheet of aluminum foil, edges folded up,
     break up large pieces to at least dime size and place in the
     drying cabinet.  Label the foil with the sample
     identification number and the time and date.  This
     information and the drying cabinet number should also be
     recorded.

                                               Revision 4, 6/1/88

-------
                              - 8 -


     On a daily basis, check to see if the sample is completely
dry and if not turn the material and further break it up into
smaller pieces to facilitate drying.

     When the sample is completely dry, fold aluminum foil over
the sample to cover it while waiting to grind.  When screens are
used transfer the sample to the aluminum foil base and wrap for
grinding.  Record the date and time the sample was wrapped up.
If the dried sample is not ground immediately store the covered
sample in the dry sample storage cabinet.

     Grind, quarter and archive the dried sample following
General Procedures VIII, Part A or B, and IX.


                 VII   PAPER PRODUCT PREPARATION

     The following is a general procedure for sub-sampling and
compositing paper products.  The working surfaces should be cov-
ered with aluminum foil and equipment (e.g. scissors) coming in
contact with the paper product should be properly cleaned.  (See
Cleaning Procedures, Section III).

A.   Paper Product Blanks

     Tear an 8" x 10" Gelman Glass Fiber Filter Type A into small
pieces and grind as described in Section VIII, Part A or B.

     After the filter is completely blended, empty the entire
contents of the blender onto a fresh piece of aluminum foil.

     Carefully fold the foil over the blended filter material to
form a packet which will effectively enclose all the material so
none will be lost.

     Place the foil packet in an I-Chem bottle, label the bottle
and seal with a signed and dated custody seal.

B.   sub—Sampling of Paper Products

(1)  Paper specimens (e.g. sheets or rolls) - cut out and weigh a
     measured aliquot (e.g. 2" x 2" area) of paper specimen
     representative of each paper product source.  Calculate and
     record the weight per area.  Use this data to determine the
     number of equal aliquots from all sources required to
     prepare a composite of approximately 15 grams.  If a QA/QC
     duplicate and spike are required prepare an additional 30
     gram composite.


                                               Revision 4, 6/1/88



                            42G

-------
                              - 9 -

     Prepare an equal number  (as determined above) of measured
     aliquots from each of the different sources of paper
     specimens which will make up the composite.  If the amount
     of paper specimens provided will exceed the 15 gram aliquot
     required cut and discard the outer 1/2 inch of the roll or
     sheet to minimize the potential for contamination which may
     have occurred from the original cutting or shipment of the
     sheet or roll.  The measured aliquots (areas) should be cut
     randomly from various different paper specimens so as to
     provide a representative sub-sample.  Note that the
     composite will be comprised of equal measured aliquots
     (areas) not equal weights.

 (2)  Paper Products (e.g. cups, plates) - Weigh one unit (l cup)
     of paper product representative of each paper product
     source.  Record the weight per unit.  Use this data to
     determine the number of equal units from all sources re-
     quired to prepare a composite of approximately 15 grams.  If
     a QA/QC duplicate and spike are required prepare an
     additional 30 gram composite.

     Take an equal number (as determined above) of units from
     each of the different sources of paper products which make
     up the composite.  The units should be randomly selected
     from the various paper products so as to provide a
     representative sub-sample of the material.  Note that the
     composite will be comprised of equal units, not equal
     weights.

C.   Compositing of Paper Products

     Combine the aliquots or units of paper product sub-sampled
as described above (Section VII, Parts B., 1 and 2) and cut each
to a size appropriate for grinding.  Approximately l" x 1" for
the Waring Blender procedure  (Section VIII, Part A) or approxi-
mately l" strips for the Wiley Mill procedure (Section VIII, Part
B).

     Grind the composited paper product sub-samples following
General Procedures Section VIII, Part A or B.


                      VIII    SAMPLE  GRINDING

     The following are general procedures for grinding Paper
Specimens/Products dried pulps, sludges and compost sample.
Since each matrix presents its own grinding difficulties it is up
to the discretion of the analyst to determine the appropriate
method.  Typically the Waring Blender procedure (Part A) is
amenable for grinding most dried pulp and sludge samples.

                                               Revision 4,  6/1/88

-------
                              - 10 -


     Some sludge samples when ground in the Waring Blender will
generate both fines and a fluff type product creating
difficulties in achieving a homogeneous mixture.  Sludge samples
of this nature should be ground in the Wiley Mill.

     For paper specimens/products the Wiley mill procedure is
generally more suitable for grinding.

A.   Waring Blender

     The grinding  (or blending) of a sample should be conducted
in the glove box.  The working surface of the glove box should be
covered or lined with aluminum foil.  The door to the glove box
room should be closed and traffic through the room minimized.
The processes of air drying and blending will be separated by as
many physical barriers as possible (i.e. separated on different
floors).

     Blend the entire sample in a properly cleaned blender (see
Section III) .  Be  sure not to add too much sample into the
blender at one time otherwise blending will not be uniform and
the blender motor  may overheat causing fragments of the blender
to mix into the sample.  To check for overheating press the
bottom of the blender assembly with gloved hands.  If the
assembly feels warm discontinue grinding until cool.  Place the
blended sample on  a sheet of aluminum foil in the glove box.
Proceed with Section IX if an archived sample is required.

B.   Wilev Mill

     The grinding  of a paper product sample should be conducted
in the glovebox room.  The grinding of sludge samples should be
conducted in a fume hood or glovebox.  The door to the glovebox
room should be closed and traffic through the room minimized.
The processes of air drying and grinding will be separated by as
many physical barriers as possible (i.e. separated on different
floors).

     Blend the entire sample in a properly cleaned Wiley Mill
(see Section III).  If an archived sample is required place the
blended sample on  a sheet of aluminum foil in the glove box and
proceed with Section IX.


                   IX   QUARTERING AND ARCHIVING

     Thoroughly mix the blended sample,  using gloved hands or a
stainless steel spoon, by turning the entire sample at least
three times, then  form into a conical pile.  Carefully flatten

                                               Revision 4, 6/1/88
                          428

-------
                              -li-

the conical pile to a uniform thickness and diameter (as wide as
spatially possible) by pressing down the apex.  Divide the
flattened mass into four equal quarters.  Refer to ASTM "Standard
Methods for Reducing Field Samples of Aggregate to Testing Size."

     An oven dried solids determination (103-105°C) is required
for solid/semi-solid samples.  Subsample each quarter and place
on a small piece of foil to be transferred to a pre-tared
crucible.  A minimum of .5 grams is required.  Refer to Standard
Methods 209F pg. 99-100, of 16th (1985) edition.

     Repeat cycle of drying, cooling, desiccating, and weighing
until a constant weight is obtained or until weight loss reflects
0.1 percent or less difference in the final percent oven dried
solids calculation.  Combine the opposing wedges into separate
I-Chem jars (i.e., two opposite wedges per jar).  If more than
two containers are required successively mix and quarter the
opposing wedges until the sample aliquot is reduced to the size
needed.

     Label the jars with the sample code.


                    X   DRY SAMPLE PROCESSING

     Ash samples (if dry) require no processing except to split
the sample for archives.  The samples will be split in the glove-
box using the quartering technique (see ASTM Standard Methods for
Reducing Field Samples of Aggregate to Testing Size).  Combine
the opposing wedges into separate I-Chem jars (i.e., two opposite
wedges per jar) and label with the sample code.  The sample por-
tion for archives.
                  XI   LIQUID SAMPLE PROCESSING

     For liquid samples which do not require filtering no further
preparation is required.  For samples which require analysis of
solids and/or liquids only, follow the specified protocol for
waste water filtration.
                       XII    SAMPLE  STORAGE

     All samples other than ash, paper products, bleached pulps
and processed blanks are refrigerated (4°C).
                                               Revision 4, 6/1/88

-------
                              -12-

              PART II - NCASI METHOD TCDD/F - 88.01


ANALYTICAL PROCEDURES FOR THE ANALYSIS OF PULP AND PAPER INDUSTRY
WASTEWATERS, SOLID WASTES, ASHES AND BLEACHED PULPS FOR 2,3,7,8-
 TETRACHLORODIBENZO-p-DIOXIN AND 2,3,7,8-TETRACHLORODIBENZOFURAN


1.  Scope and Application

1.1  This method provides procedures for the detection and
     quantitative measurement of 2,3,7,8-tetrachlorodibenzo-p-
     dioxin (2,3,7,8-TCDD) and 2,3,7,8-tetrachloro-dibenzofuran
     (2,3,7,8-TCDF) in pulp and paper industry wastewaters, solid
     wastes, ashes and bleached or partially bleached pulps.
     Detection limits for the solid sample matrices are in the
     low parts per trillion (ppt) and in the low parts per
     quadrillion (ppq) for the wastewater matrices.  The
     analytical method calls for the use of high-resolution gas
     chromatography and high-resolution mass spectrometry
     (HRGC/HRMS) on purified sample extracts.  Table 1 lists the
     sample types covered by this protocol, Method Calibration
     Limits (MCLs) for 2,3,7,8-TCDD and 2,3,7,8-TCDF and other
     germane information.

1.2  The sensitivity of this method is dependent upon a number of
     factors, including (but not limited to) the level of
     interferents within a given matrix, the recovery of the
     internal standard, the instrument sensitivity, etc.  If an
     analyte is not detected, estimated limits of detection will
     be reported for each individual analysis.

1.3  This method is designed for use by analysts who are
     experienced with residue analysis and skilled in high-
     resolution gas chromatography/high resolution mass
     spectrometry (HRGC/HRMS).

1.4  Because of the extreme toxicity of these compounds, the
     analyst must take the necessary precautions to prevent
     exposure to himself or herself, or to others, of materials
     known or believed to contain 2,3,7,8-TCDD or 2,3,7,8-TCDF.
     It is the laboratory's responsibility to ensure that safe
     handling procedures are employed.

2.  Summary of the Method

2.1  This procedure uses a matrix-specific extraction, analyte-
     specific cleanup, and isomer specific high-resolution
     capillary column gas chromatography/high-resolution mass
     spectrometry (HRGC/HRMS) techniques.

                                      NCASI Method TCDD/F  - 88.01
                           430

-------
                             -13-
2.2  If interferents are encountered,  the method provides
     selected cleanup procedures to aid the analysts in their
     elimination.

2.3  An aliquot of a blended solid waste, ash or pulp is spiked
     with a solution containing specified amounts of each of the
     two isotopically (13C^2> labeled TCDD/TCDFs.  The sample is
     then extracted according to a matrix-specific extraction
     procedure.  The extraction procedures are:   (a) 95% ethanol
     Soxhlet extraction for pulps, (b) 68:32 ethanol/toluene
     Soxhlet extraction for waste treatment plant sludges, and
     (c) benzene or toluene Soxhlet extraction for ashes.
     Wastewaters are first spiked with a solution containing
     specified amounts of each of the two isotopically (13C12)
     labeled TCDD/TCDFs and are then stirred or shaken for thirty
     minutes to allow equilibration of the spike.  The sample is
     then filtered through a glass fiber filter.  The filtered
     solids and filter are soxhlet extracted for 16 hours using
     68:32 ethanol/toluene.  The filtered wastewater is liquid-
     liquid extracted first with methylene chloride then with
     toluene.  The extracts from the filtered solids and the
     aqueous portions are then combined prior to cleanup and
     analysis.  Immediately following extraction and prior to any
     extract cleanup procedures, the extracts are spiked with a
     cleanup recovery standard (37Cl4-2,3,7,8-TCDD) to monitor
     losses through the cleanup.

2.4  The extracts are then submitted to an acid-base washing
     treatment and dried.  Following a solvent exchange step, the
     residues are cleaned up with two or more column
     chromatographic procedures.  Just prior to HRGC/HRMS
     analysis, a recovery standard (13Ci2~l/2,3,4-TCDD) is added
     to all final extracts.

2.5  Two uL of the concentrated extract are injected into an
     HRGC/HRMS system capable of performing selected ion
     monitoring at resolving powers of at least 10,000 (10
     percent valley definition).

2.6  The identification of 2,3,7,8-TCDD and 2,3,7,8-TCDF is based
     on their elution at their exact retention time (-1 to +3
     seconds from the respective internal standard signal), the
     simultaneous detection of the M~*" and M+2* ions, on a
     comparison of the ratio of the integrated ion abundance of
     these ions to their theoretical abundance ratio and peak
     signal to noise ratio of >2.5:1 for both ions.
                                      NCASI Method TCDD/F - 88.01
                             431

-------
                              -14-


3.  Definitions

3.1  Internal Standards:  13c12-2,3,7,8-TCDD and 13C12-2,3,7,8-
     TCDF labeled internal standards are used in this method for
     the quantitation of the corresponding native analytes.   The
     internal standards are added to all samples including method
     blanks and quality control samples prior to extraction.

3.2  Cleanup Recovery standard (Table 2);  One standard (37cl4-
     2,3,7,8-TCDD) is used to determine the recovery throughout
     the cleanup procedure.  It is added to the sample extract
     immediately following extraction and prior to any cleanup
     procedures.

3.3  Recovery Standard:  One recovery standard (^Ci 2-1,2,3,4-
     TCDO) is used to determine the percent recoveries for the
     13C12-2,3,7,8-TCDD and 13C12-2,3,.7,8-TCDF internal standards
     and the 37Cl4-2,3,7,8-TCDD cleanup recovery standard.  It is
     added to the final sample extract immediately prior to
     HRGC/HRMS analysis.

3.4  Calibration Standard Solutions (CS1 to CSS, Table 4):  A set
     of a minimum of five solutions are prepared for instrument
     calibration.  Each solution containing known amounts of
     2.3,7,8-TCDD and 2,3,7,8-TCDF, the internal standards
     (^3Ci 2~labele<* TCDD and TCDF), the cleanup recovery standard
     (37Cl4-2,3,7,8-TCDD), and the recovery standard (13C12~
     1,2,3,4-TCDD).  The calibration solutions are used to
     determine the instrument response for the native analytes
     relative to the internal standards and for the internal
     standards and cleanup recovery standard relative to the
     recovery standard.

3.5  Standard Sample Fortification Solution (IS, Table 2);
     Solution containing the two internal standards (i;iC12-
     2,3,7,8-TCDD and 13C12-2,3,7,8-TCDF), which is used to spike
     all samples before extraction and cleanup.

3.6  Cleanup Recovery Standard Spiking Solution (CR, Table 2):
     Solution containing the cleanup recovery standard (^'Cl^
     2,3,7,8-TCDD), which is added to the sample extract
     immediately following liquid-liquid extraction or soxhlet
     extraction prior to any cleanup steps.

3.7  Recovery Standard Spiking Solution (Table 2);  Solution
     containing the recovery standard (^3Ci2-l,2,3,4-TCDD)'  which
     is added to the final sample extract before HRGC/HRMS
     analysis.


                                      NCASI Method TCDD/F - 88.01



                           432

-------
                              -15 -


3.8  GC Column Performance Check Mixture(s) (Table 3):   A
     solution (or solutions) containing a mixture of selected
     TCDD/TCDF standards/ which is used to demonstrate  continued
     acceptable performance of the capillary column (i.e., <25
     percent valley separation of 2,3,7,8-TCDD from all other 2.1
     TCDD isomers).

3.9  Relative Response Factor:  Response of the mass spectrometer
     to a known amount of an analyte relative to a known amount
     of an internal standard.

3.10 Laboratory Method Blank:  This blank is prepared in exactly
     the same manner as a sample, performing all analytical
     procedures except the addition of a sample aliquot to the
     extraction vessel.

3.11 Estimated Level of Method Blank Contamination:  The response
     from a signal occurring at the retention times for either
     2,3,7,8-TCDD or 2,3,7,8-TCDF and at any of the masses
     monitored is used, as described in Section 12, to  calculate
     the level of contamination in the method blank. The results
     from such calculations must be reported along with the data
     obtained on the samples belonging to the batch associated
     with the method blank.  Reporting a method blank
     contamination level for 2,3,7,8-TCDD or 2,3,7,8-TCDF that
     either equals or exceeds the level present in any  one sample
     from the batch or, which exceed the target detection limits
     would invalidate the results and require sample reruns for
     all positive samples found in that batch of samples.  A
     positive sample is defined as a sample found to contain
     either 2,3,7,8-TCDD or 2,3,7,8-TCDF.

3.12 Sample Rerun:  Extraction of another aliquot of the sample
     followed by extract cleanup and extract analysis.

3.13 Extract Reanalysis:  Analysis by HRGC/HRMS of another
     aliquot of the same final extract.

3.14 Mass Resolution Check:  Standard method used to demonstrate
     a minimum static resolving power (10 percent valley
     definition) of 10,000.

3.15 Method Calibration Limits (MCLs):  For a given sample size,
     the final extract volume, and the lowest and highest
     concentration calibration solutions, the lower and upper
     MCLs delineate the region of quantitation for which the
     HRGC/HRMS system was calibrated with standard solutions.
                                      NCASI Method TCDD/F - 88.01
                           M fl •)
                          4 o o

-------
                             - 16 -


3.16 HRGC/HRMS Solvent Blank:  This additional QC check analysis
     corresponds to a 2 uL injection of pure tetradecane into the
     GC column and a complete HRGC/HRMS analysis.  Such QC check
     can be used following the analysis of a method blank which
     has shown detectable levels of the target analytes in order
     to verify that the contamination was in the extract and was
     not the result of syringe, septum and/or solvent
     contamination.  Acceptable method blanks or HRGC/HRMS
     solvent blanks (Section 9.1 for guidelines) must be obtained
     before pursuing the analysis of subsequent sample extracts.

4.  Interferents

4.1  Solvents, reagents, glassware, and other sample processing
     hardware may yield discrete artifacts and/or elevated
     baselines which may cause misinterpretation of
     chromatographic data.  All of these materials must be
     demonstrated to be free from interferents under the
     conditions of analysis by running laboratory method blanks.

4.2  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.

4.3  Interferents co-extracted from the sample may vary
     considerably from sample to sample.  2,3,7,3-TCDD and
     2,3,7,8-TCDF are associated with other interfering
     chlorinated compounds which may be found at concentrations
     several orders of magnitude higher than that of the analytes
     of interest.  Retention times of the target analyte must be
     verified using reference standards.  These values must
     correspond to the retention time criteria established in
     Section 11.3.

4.4  A 60 m DB-5 fused silica capillary column is used to resolve
     2,3,7,8-TCDD from all other TCDD isomers.  A 30 m DB-225
     fused silica capillary column is used to resolve 2,3,7,8-
     TCDF from all other TCDF isomers.

5.  Apparatus and Materials

5.1  Gas chromatograph/mass spectrometer data system.

5.1.1     Gas chromatograph:  An analytical system with a
          temperature-programmable gas chromatograph and all
          required accessories including syringes, analytical
          columns, and gases.  A Grob type splitless injector is
          recommended for use with high resolution columns.


                                      NCASI Method TCDD/F  -  88.01
                        434

-------
                             - 17 -


5.1.2     A60mx0.25mm DB-5 fused silica capillary column is
          required for the analysis of 2,3,7,8-TCDD and a 30  m x
          0.25 mm DB-225 fused silica capillary column is
          required for the analysis of 2,3,7,8-TCDF unless the
          laboratory can achieve all isomer specificity and
          detection limit requirements with 0.32 mm DB-5 or DB-
          225 columns.

5.1.3     Mass spectrometer:   A high resolution instrument
          capable of performing selected ion monitoring at
          resolving powers of at least 10,000 (10 percent valley
          definition) and utilizing electron impact ionization is
          specified.  The system must be capable of selected  ion
          monitoring (SIM) for at least eleven ions
          simultaneously, with a total cycle time of 1 second or
          less.  At a minimum, the ions listed in Table 5 for
          each descriptor must be monitored.

5.1.4     GC/MS interface:  Any GC-to-MS interface that gives an
          acceptable calibration response for each analyte of
          interest at the concentration required and achieves the
          required tuning performance criteria (Section 7.2)  may
          be used.  Inserting a fused silica column directly  into
          the MS source is recommended, however, care must be
          taken not to expose the end of the column to the
          electron beam.  If other GC-to-MS interfaces are
          utilized, they must be constructed of all glass or
          glass-lined materials.  The glass can be deactivated by
          silanizing with dichlorodimethylsilane.

5.1.5     Data system:  A computer system must be interfaced  to
          the mass spectrometer.  The system must allow for the
          continuous acquisition and storage on machine-readable
          media of all data obtained throughout the duration  of
          the chromatographic program.  The computer must have
          software that can search any GC/MS data file for ions
          of a specific mass and can plot such ion abundances
          versus time or scan number.  This type of plot is de-
          fined as a Selected Ion Current Profile (SICP).  Soft-
          ware must also be able to integrate the abundance,  in
          any SICP, between specified time or scan number limits.

5.2  Pipets-Disposable

5.2.1     Pipets-Disposable,  Pasteur, 150 mm long x 5 mm ID
          (Fisher Scientific Company, No. 13-678-6A, or
          equivalent).
                                      NCASI Method TCDD/F - 88.01
                      435

-------
                             - 18 -


5.2.2     Pipets, disposable, serological 5 mL (6 nun i.d.)  for
          preparation of the carbon column specified in Section
          10.2.3.

5.3  Conical Reacti-vial, 2 mL (Pierce Chemical Company),   it is
     recommended that these vials be silanized with
     dichlorodimethlysilane.

5.4  Soxhlet Extractors, 40 mm i.d. with 250 mL boiling flask and
     appropriate thimbles.

5.5  Chromatography columns, 15 mm i.d. x 20 cm length, with 250
     mL reservoir, and 11 mm i.d. x 16 cm length with 50 mL
     reservoir.

5.6  125 mL and 2 L Separatory Funnels (Fisher Scientific Company,
     No. 10-437-5b, or equivalent).

5.7  Nitrogen blowdown apparatus (N-Evap Analytical Evaporator
     Model 111, Organomation Associates, Inc., Northborough,
     Massachusetts or equivalent).

5.8  Glass fiber filter paper (Whatman No. GF/D, or equivalent).

5.9  Top loading balance capable of weighing 200 g to the nearest
     0.1 g.

5.10 Roto-evapbrator, (R-110. Buchi/Brinkman, American Scientific
     No. E5045-10 or equivalent).

5.11 Pressure Filtration Apparatus, (Millipore No. YT30 142 HW or
     equivalent), or alternatively, 1 L glass Microanalysis (90
     mm) filter holder for vacuum filtration (Fisher No. 09-753-2
     or equivalent) with 1 L vacuum flask.

5.12 Glass wool, extracted with methylene chloride and stored in
     a clean glass jar.

5.13 Glass funnels, sized to filter 150 to 250 mL of solvent
     through a sodium sulfate bed.

5.14 Glass vials, 20 to 40 mL capacity with Teflon lined screw
     caps

6.  Reagents

6.1  Sodium hydroxide-(ACS), 20 percent (w/v) in distilled water.

6.2  Sulfuric acid-(ACS), concentrated.

                                      NCASI Method TCDD/F - 88.01
                             43G

-------
                             - 19 -


6.3  Methylene chloride,  n-hexane,  benzene,  toluene,  acetone,
     isooctane, methanol, tetradecane,  cyclohexane.   Distilled in
     glass or highest available purity.  95  percent ethanol should
     be prepared by mixing the appropriate quantities of absolute
     ethanol (100 percent, undenatured, highest purity available)
     and organic free water.

6.4  Prepare stock standards in a glovebox or hood from
     concentrates.  The stock solutions (50  ppm) are stored at
     room temperature in the dark,  and checked frequently for
     signs of degradation or evaporation, especially just prior
     to the preparation of working standards.

6.5  Prepurified nitrogen gas.

6.6  Anhydrous sodium sulfate (reagent grade).  Pre-extract with
     methylene chloride and dry at 130° C.

6.7  Silica gel - Kieselgel 60, activate for >12 hours at 130°C
     before use.  Store at 130°C in covered  flask.

6.8  Acid alumina - Bio-Rad Ag-4, activate for 12 hours at 130°C
     before use.  Store at 130°C in covered  flask.

6.9  Basic alumina - Bio Rad Ag-10, kiln at  600° C for >24 hours
     before use.  Store at 130° C in covered flask.   DO NOT USE
     IF OLDER THAN FIVE DAYS!

6.10 Carbopack/silica gel - Mix 3.6 g Carbopack C, 60/80 mesh
     (Supelco 1-0257) and 16.4 g activated silica gel.  Activate
     mix for >12 hours at 130°C before use.   Store at 130°C in
     covered flask.

6.11 AX-21 carbon/silica - AX-21 carbon is available through
     Anderson Development Company 1415 E. Michigan St., Adrian,
     MI 4922.  Wash 100 g of AX-21 carbon powder (as received) by
     suspending in 300 mL methanol and subsequently vacuum filter
     through a glass fiber filter (Gelman A/E or equivalent)
     fitted in a 350 mL Buchner funnel.  Follow with two 100 mL
     methanol rinses and draw the excess methanol from the filter
     cake by continued vacuum.   Dry the washed AX-21 carbon at
     130°C for a minimum of 72 hours.  Combine 5 g of prepared AX-
     21 carbon with 95 g of prepared silica gel (6.7) in a wide
     mouth bottle with a teflon lined screw cap.  Blend by shaking
     until a uniform color is achieved.  Activate the blended AX-
     21 carbon/silica at 130°C for a minimum of 24 hours and store
     in a desiccator at room temperature.
                                      NCASI Method TCDD/F - 88.01
                           437

-------
                             - 20 -


6.12 44% I^SCM/silica gel - Mix 24 mL cone. H2SC>4 and 56 g
     activated silica gel.  Stir and shake until free flowing.
     Store in sealed containers at room temperature.

6.13 33% NaOH/silica gel - Mix 34 mL 1 N NaOH and 67 g activated
     silica gel.  Stir and shake until free flowing.  Store in
     sealed containers at room temperature.

6.14 Calibration Solutions.

6.14.1    Calibration Standard Solutions (CS1 to CSS, Table 4)  -
          Five solutions containing native TCDD and TCDF,"
          labeled TCDD and TCDF and 37C14 TCDD at known
          concentrations used to calibrate the instrument.

6.14.2    High level stock solutions (50±5 ug/mL) obtained from
          NCASI will be used to prepare the appropriate dilutions
          required to make up the calibration standards described
          in Table 4. Suitable 13C^2 and 37Cl4 standards must be
          obtained from commercial sources.  It is the
          responsibility of the laboratory to ascertain that the
          calibration solutions prepared are indeed at the
          appropriate concentrations before they are used to
          analyze samples.  Store the high level stock solutions
          protected from light at room temperature.

6.14.3    Store the working concentration calibration solutions
          in 1 mL minivials at 4°C protected from light.

6.15 GC Column Performance Check Solutions.

     These solutions contain other TCDD isomers or TCDF isomers
     for the purpose of documenting the chromatographic
     resolution.  The 13C12-2,3,7,8-TCDD or 13C12~2'3,7,8-TCDF is
     also present.  The laboratory Is required to use tetradecane
     as the solvent and adjust the volume so that the final
     concentration does not exceed 100 pg/uL.  Table 3 summarizes
     the qualitative composition (minimum requirement) of these
     performance evaluation solutions.  The difference in
     concentrations of each isomer should be less than 35 percent
     of the concentration of the 2,3,1,8-substituted isomer.
                                      NCASI Method TCDD/F - 88.01
                          438

-------
                              -21-


6.16 Sample Fortification Solution (IS,  Table 2).

     This solution contains the two internal standards (
     2,3,7,8-TCDD and 13C12-2,3,7,8-TCDF)  at the  nominal
     concentration that is listed -in Table 2.  Ten uL of  solution
     IS is diluted in acetone and then is  spiked  into each
     wastewater sample aliquot prior to filtration and extraction.
     Ten uL of solution IS is spiked into  each solid type sample
     aliquot prior to soxhlet extraction.

6.17 Cleanup Recovery Standard Spiking Solution (CRS, Table 2).

     This solution contains the cleanup recovery  standard (37Cl4~
     2,3,7,8-TCDD) at the nominal concentration listed in Table
     2.  Ten uL of this solution will be spiked into each sample
     extract immediately after the extraction step and before any
     cleanup procedures are started.

6.18 Recovery Standard Spiking Solution (RS, Table 2).

     This solution contains the recovery standard (^C^-l* 2,3,4-
     TCDD) at the nominal concentration listed in Table 2.  Ten
     uL of this solution will be spiked into each sample  extract
     before HRGC/HRMS analysis.

7.  System Performance Criteria

     System performance criteria are presented below.  The
     laboratory will use the GC columns described in Section
     5.1.2.  It must be documented that all applicable system
     performance criteria specified in Section 7.1 were met before
     analysis for any sample is performed.  Table 6 provides
     recommended GC conditions that can be used to satisfy the
     required criteria.  During a typical  12-hour analysis
     sequence, the GC column performance and mass spectrometer
     resolving power checks must be performed at  the beginning
     of the 12-hour period of operation.  A routine calibration
     verification is required at the beginning and end of each
     12-hour period during which samples are analyzed.  A method
     blank or HRGC/HRMS solvent blank run is required between a
     calibration run and the first sample run.

7.1  GC Column Performance.

7.1.1     Inject the column performance check solution (Section
          6.14) and acquire selected ion monitoring (SIM) data as
          described in Section 5.1.3.

7.1.2     The chromatographic separation between 2,3,7,8-TCDD and

                                      NCASI Method TCDD/F - 88.01
                          430

-------
                             - 22 -

          the peaks representing any other TCDD isomers or between
          2,3,7,8-TCDF and the peaks representing other TCDF
          isomers must be resolved with a valley of <25 percent,
          where

               Valley Percent = (a/b) x 100

               a =  height of valley measured between 2,3,7,8-
                    TCDD or 2,3,7,8-TCDF and the closest TCDD or
                    TCDF eluting isomer, and

               b = the peak height of 2,3,7,8-TCDD or 2,3,7,8-
                    TCDF.

          It is the responsibility of the laboratory to verify
          the conditions suitable for the appropriate resolution
          of 2,3,7,8-TCDD from all other TCDD isomers and of
          2,3,7,8-TCDF from all other TCDF isomers.  Any
          individual selected ion current profile (SICP) (for the
          tetras, this would be the SICP for m/z 322 and m/z 304)
          or the reconstructed homologue ion current (for the
          tetras, this would correspond to m/z 320 + m/z 322 +
          m/z 304 + m/z 306) constitutes an acceptable form of
          data presentation.  An SICP for m/z 334 (labeled TCDD)
          and m/z 318 (labeled TCDF) is also required.

7.2  Mass Spectrometer Performance.

     All mass spectrometer tuning must be performed with the GC
     oven equilibrated at a temperature of 260° C.  This will
     provide carrier gas flow into the ion source at a rate
     consistent with the flow during TCDD data acquisition.

7.2.1     The mass spectrometer must be operated in the electron
          ionization mode.  It is recommended that the ionization
          potential be set to optimize sensitivity for the given
          column flow and source design.  A static resolving power
          of at least 10,000 (10 percent valley definition) must
          be demonstrated at appropriate masses before any
          analyses are performed (Section 11).  Static resolving
          power checks must be performed at the beginning of each
          12-hour period of operation.  However, it is recommended
          that a visual check (i.e., documentation is not
          required) of the static resolution be made by using the
          peak matching unit before and after each analysis.
          Corrective actions must be implemented whenever the
          resolving power does not meet the requirement.
                                      NCASI Method TCDD/F - 88.01
                         440

-------
                             - 23 -


7.2.2     Chromatography time for TCDDs and TCDFs exceeds the
          long-term mass stability of the mass spectrometer.
          Because the instrument is operated in the high-
          resolution mode, mass drifts of a few ppm can have
          serious adverse effects on the instrument performances.
          Therefore, a mass-drift correction is mandatory.  To
          that effect, use a lock-mass ion from the reference
          compound (PFK) used for tuning the mass spectrometer
          and monitor and record the lock-mass ion channel during
          SIM acquisitions.  The level of the reference compound
          metered inside the ion chamber during HRGC/HRMS analyses
          should be adjusted so that the amplitude of the selected
          lock-mass ion signal does not exceed 10 percent of  the
          full-scale deflection for a given set of detector
          parameters.  Under those conditions, sensitivity changes
          that might occur during the analysis can be more
          effectively monitored.

7.2.3     By using PFK molecular leak and an appropriate ion
          within the scan window, tune the instrument to meet the
          minimum required resolving power of 10,000 (10 percent
          valley).

7.2.4     Documentation of the instrument resolving power must be
          accomplished by recording the peak profile of the high-
          mass reference signal (m/z 380.9760).  The format of
          the peak profile must allow manual determination of the
          resolution; i.e., the horizontal axis must be a
          calibrated mass scale (amu or ppm per division).  The
          result of the peak width measurement (performed at  5
          percent of the maximum, which corresponds to the 10
          percent valley definition) must appear on the hard  copy
          and cannot exceed 100 ppm at m/z 380.9760 (or 0.038 amu
          at that particular mass).

8.  Calibration

8.1  Initial Calibration.

     Initial calibration is required before any samples are
     analyzed for 2,3,7,8-TCDD and 2,3,7,8-TCDF.  Initial
     calibration is also required if any routine calibration
     (Section 8.3) does not meet the required criteria listed in
     Section 8.4.

8.1.1     A minimum of five high-resolution concentration
          calibration solutions covering a minimum concentration
          range of three orders of magnitude must be used for the
          initial calibration.  The maximum difference between

                                      NCASI Method TCDD/F - 88.01
                                 441

-------
                             - 24 -

          the concentrations of one standard and the next higher
          concentration standard will be one order of magnitude.
          The lowest concentration for 2,3,7,8-TCDD and 2,3,7,8-
          TCDF must be 0.5 pg/uL or less.  If a second instrument
          is used for high concentration confirmations only,  then
          the low calibration standard can be at a higher
          concentration. . However, it is the responsibility of
          the laboratory to analyze and report only results that
          meet the target detection limits and which are within
          the calibrated range of the instrument used for the
          determination.

8.1.2     Tune the instrument with PFK as described in Section 7.2.

8.1.3     Inject the GC column performance check solution,
          (Section 6.14) and acquire SIM mass spectral data as
          described earlier in Section 5.1.3.  The laboratory
          must not perform any further analysis until it has
          demonstrated and documented that the column performance
          criterion listed in Section 7*1.2 was met.

8.1.4     By using the same GC and mass spectrometer conditions
          that produced acceptable results with the column
          performance check solution, analyze a 2 uL aliquot of a
          minimum of five concentration calibration solutions in
          duplicate with the following mass spectrometer operating
          parameters.

     8.1.4.1   The total cycle time for data acquisition must be
               <1 second.  The total cycle time includes the sum
               of all the dwell times and voltage reset times.

     8.1.4.2   Acquire SIM data for the ions listed in the
               descriptors in Table 5.

8.1.5     Integrate peak areas for all target analyte and standard
          peaks.  All the following criteria must be met for all
          individual injections.

     8.1.5.1   The retention time  (at maximum peak height) of the
               native analytes (i.e., the two ions used for
               quantitation purposes) must be within -1 and +3
               seconds of the retention time of the peak for the
               isotopically labeled internal standard at m/z
               corresponding to the first characteristic ion  (of
               the set of two; Table 5).

     8.1.5.2   The ion current responses for both ions used for
               quantitative purposed (e.g.,  for 2,3,7,8-TCDD: m/z
               319.8465 and 321.8936) must reach maximum

                                      NCASI  Method TCDD/F  - 88.01
4                            A O
                          ^i<~

-------
                             - 25. -

               simultaneously (± 1 second).

     8.1.5.3   The ion current responses for both ions used for
               the labeled standards (e.g.,  for ^C-TCDDs:  m/z
               331.9368 and 333.9339) must reach maximum
               simultaneously (± 1 second).

     8.1.5.4   The ratio of integrated ion currents for the M+ to
               the M+2+ ions belonging to the native analyte,  the
               carbon-labeled internal standards and the recovery
               standard must be between 0.65 and 0.89.

     8.1.5.5   For each SIM trace, all target analyte and labelled
               standard peaks must be detected with a minimum
               signal to noise ratio (S/M) £ 5.

8.1.6     Calculate the relative response factors (RRF) for
          unlabeled target analytes relative to the appropriate
          internal standards (Table 4) according to the following
          formula:

                                   Ax x QIS
                         RRF(n) =  	
                                   Ox x AIS
               where:
               Ay = the sum of the integrated ion
               abundances of the quantitation ions (Table 5) for
               unlabeled TCDD/TCDF.
               ATS = tne sum °f t^ie integrated ion abundances of
               the quantitation ions (Table 5) for the labeled
               internal standards.

               QIS = tne quantity of the internal standard
               injected (pg) .

               QX = the quantity of the unlabeled TCDD/TCDF
               analyte injected (pg).

               RRF(n) = relative response factor of a particular
               TCDD/TCDF isomer (n) relative to the internal
               standard as determined from one injection.

          The RRF(n) are dimensionless quantities; the units used
          to express QIS and gx and must be the same.
                                      NCASI Method TCDD/F - 88.01
                             443

-------
                              -26 -


8.1.7     Calculate the relative response factors (RRF) for the
          labeled 13C12 internal standards relative to the
          recovery standard according to the following formula:
                                    * QRS
                     RRF(ra) =      _
                               Qisra * ARS

               where :

               AJS™ = sum of the integrated ion abundances of the
               quantitation ions (Table 5) for a given internal
               standard  (m = 1-2).

               ARS ~ sum °^ tne integrated ion abundances of the
               quantitation ions (Table 5) for the recovery
               standard.

               QRS and Qism = quantities of the recovery standard
               (RS) and a particular internal standard (IS = m)
               respectively, injected (pg) .

               RRF(m) = relative response factor of a particular
               internal standard (m) relative to the recovery
               standard as determined from one injection.

     The RRF(m) is a dimensionless quantity; the units used to
     express Qjsm an<* QRS must ^e tne same.

8.1.8     Calculate the relative response factors (RRF) for  the
          labeled ^'CI^ cleanup recovery standard relative to the
          recovery standard according to the following formula:

                               ACS° x
                     RRF(o) =
                                    x  RS
               where :
                    =  the integrated ion abundances of the
               quantitation ion (Table 5) .

               ARS = sum °f tne integrated ion abundances of the
               quantitation ions  (Table 5) for the recovery
               standard.

               Qps and QCS° = quantities of  the recovery standard
               (RS) and the clean-up recovery standard injected
               respectively, (pg).
                                      NCASI Method TCDD/F - 88.01
                              444

-------
                             -  27  -


               RRF(o)  =  relative response  factor  the  Cleanup
               recovery  standard relative  to  the  recovery  standard
               as  determined from  one injection.

          The RRF(o)  is  a dimensionless  quantity;  the units used
          to express  Q^s and Qcs°  must be  the same.
8.1.9     Calculate the  RRF(n)s and their respective  percent
          relative standard deviations  (%RSD)  for  the five
          calibration solutions.


                RRF(n) =  1/10   j[  RRFj(n),


               where n represents the particular TCDD/TCDF  isomer
               (n=l-2),  and j  is the injection number (j=l-10)
               and RRF(n)  is the relative response factor as
               defined in  Section 8.1.6.

               RRF(rO =  calculated mean relative response factor
               of a TCDD/TCDF isomer (n)  relative  to  the internal
               standard  as determined from the 10  initial
               calibration injections (j).


8.1.10    Mean relative  response factors [RRF(m)]  to  be used  for
          the determination of the percent recoveries for the
          internal standards are calculated as follows:


               RRF(m) =   1/10


               where m represents the particular internal standard
               (m=l-2),  and j  is the injection number (j=l-10)
               and RRF(m)  is the relative response factor as
               defined in  Section 8.1.7.


               RRF(m) =  calculated mean relative response factor
               of a particular internal standard (m)  relative to
               the recovery standard as determined from the 10
               initial calibration injections (j).

8.1.11    Mean relative  response factors [RRF(o)]  to be used for
          the determination of the percent recovery for the clean-
          up recovery standard are calculated as follows:


                                     NCASI Method TCDD/F - 88.01
                           445

-------
                             - 28 -
                  RRF(o) =  1/10 V    RRFj (o),


               where o represents the clean-up recovery standard,
               and j is the injection number (j=l-10) and RRF(o)
               is the relative response factor as defined in
               Section 8.1.8.

               RRF(o) = calculated mean relative response factor
               for the clean-up recovery standard relative to the
               recovery standard as determined from the 10 initial
               calibration injections (j).

8.2  Criteria for Acceptable Calibration.

     The criteria listed below for acceptable calibration must be
     met before the analysis is performed.

8.2.1     The percent relative standard deviation for each of the
          mean response factors [RRF(n), RRF(m) and RRF(o)] from
          each of the 50 determinations (20 for the unlabeled
          standards, 20 for the labeled reference compounds and 10
          for the cleanup recovery standard) must be less than 20
          percent.

8.2.2     The S/N for the GC signals present in every SIM trace
          (including the ones for the labeled standards) must be > 5.

8.2.3     Isotopic ratios of all the M+ to M+2+ peaks must be within
          ± 15 percent of the theoretical value (i.e. 0.65 to 0.89).

               NOTE: If the criterion for acceptable calibration
               listed in Section 8.2.1-is met, the analyte-
               specific RRF can then be considered independent of
               the analyte quantity for the calibration
               concentration range.  The mean RRFs will be used
               for all calculations until the routine calibration
               criteria (Section 8.4) are no longer met.  At such
               time, new mean RRFs will be calculated from a new
               set of injections of the calibration solutions.

8.3  Routine Calibration.

     Routine calibrations must be performed at the beginning of a
     12-hour period after successful mass resolution and GC
     resolution performance checks and at the end of a 12-hour
     period following the analysis of samples.

                                      NCASI Method TCDD/F - 88.01
                               :G

-------
                             - 29 -


8.3.1     At the beginning of the 12-hour period,  inject 2 uL of
          the lowest or the next to lowest concentration calibra-
          tion solution (Table 4).  By using the same HRGC/HRMS
          conditions as used in Sections 8.1.3 and 8.1.4,
          determine and document an acceptable calibration as
          provided in Section 8.4.

8.3.2     At the end of the 12-hour period, inject 2 uL of the
          highest concentration calibration solution (Table 4).
          By using the same HRGC/HRMS conditions as used in
          Sections 8.1.3 and 8.1.4, determine and document an
          acceptable calibration as provided in Section 8.4.

8.4  Criteria for Acceptable Routine Calibration.

     The following criteria must be met before further analyses
     are performed.  If these criteria are not met, corrective
     action must be taken and the instrument must be recalibrated.

8.4.1     The measured RRFs [RRF(n) for the unlabeled standards]
          obtained during the routine calibration runs must be
          within 20 percent of the mean values established during
          the initial calibration (Section 8.1.4.6).

8.4.2     The measured RRFs [RRF(m) for the labeled standards RRF
          (o) for the clean-up recovery standard]  obtained during
          the routine calibration runs must be within 20 percent
          of the mean values established during the initial
          calibration (Section 8.1.4.7).

8.4.3     Isotopic ratios of all the M+ to M+2+ peaks must be within
          ± 15 percent of the theoretical value (i.e. 0.65 to 0.89).

8.4.4     If any one of the criteria above (Sections 8.4.1 through
          8.4.3) are not satisfied, a 'second attempt can be made.   If
          the second attempt fails to meet all the criteria,  then the
          entire initialization process (Section 8.1) must be
          repeated.

8.4.5     If the second analysis of the standard at the end of
          the 12-hour shift fails, then all samples run during
          that shift must be reanalyzed after the initialization
          process has been repeated.

9.  Quality Control

9.1  Method Blank.

     A method blank is performed by executing all of the specified

                                      NCASI Method TCDD/F - 88.01
                            4 4 7

-------
                             - 30 -

     extraction and cleanup steps, except for the introduction of
     a sample.  The method blank is also dosed with the internal
     standards, the cleanup standard and the recovery standard at
     the appropriate stages of the analysis.  For water samples,
     one liter of deionized or distilled water should be used as
     the method blank.

9.1.1     Before processing any samples, the analyst must
          demonstrate through the analysis of a method blank that
          all glassware and reagents are interferant-free at the
          method detection limit of the matrix of interest.   Each
          time a set of samples is extracted, or there is a  change
          in reagents, a method blank must be processed to
          document the absence of laboratory contamination.

9.1.2     A laboratory "method blank" must be run along with each
          set of 24 or fewer samples.

9.1.3     The Method Blank internal standard and cleanup recovery
          standard recoveries must be greater than 40 percent.
          If the recoveries fall below this minimum, the source
          of the problem will be identified and corrected.  The
          laboratory will repeat the analysis of all samples in
          that set which had either internal standard or cleanup
          recoveries below the 40 percent minimum objective.

9.2  Method Blank Native Spike.

     A method blank native spike is performed by executing all of
     the specified extraction and cleanup steps, except for the
     introduction of a sample.  The method blank native spike is
     dosed with the native analytes at the target detection
     limits, the internal standards, the cleanup standard and the
     recovery standard at the appropriate stages of the analysis.
     For water samples, one liter of deionized or distilled water
     should be used as the method blank native spike.

9.2.1     The laboratory will conduct a Method Blank Native Spike
          at the target detection limits (Section 9.4) whenever
          there is a change in the lot number of the acid alumina,
          Carbopack C or basic alumina to demonstrate acceptable
          recoveries of the native spikes.

9.2.2     All quality control criteria must be met before the
          reagent will be judged acceptable for use in the
          analysis of samples.

9.2.3     The Method Blank Native Spike internal standard and
          cleanup recovery standard recoveries must be greater
          than 40 percent.  If the recoveries fall below this

                                      NCASI Method TCDD/F -  88.01


                          448

-------
                             - 31 -

          minimum,  the source of the problem will be identified
          and corrected.   The laboratory will repeat the analysis
          of all samples  in that set which had either internal
          standard  or cleanup recoveries below the 40 percent
          minimum objective.

9.3  GC column performance must be  demonstrated initially and
     verified prior to analyzing any sample in a 12-hour period.
     The GC column  performance check solution must be analyzed
     under the same chromatographic and mass spectrometric
     conditions used for  other samples and standards.

9.4  Detection Limit Criteria.

     The target detection limits will be 1 ppt for pulps and
     sludges and 10 ppq for treated effluents.

9.4.1     If the laboratory has used a sample size smaller than
          the maximum recommended sample size listed in Table 1
          and fails to achieve the  target detection limit, the
          analysis  will be repeated using an appropriately larger
          sample size.

9.5  Isomer Specificity.

     The laboratory will  only report isomer specific data when
     2,3,7,8-TCDD or 2,3,7,8-TCDF is detectable.

9.5.1     Isomer specificity will be demonstrated on a daily basis
          according to the procedure outlined in Section 7.1.

9.5.2     If a non-isomer specific  column is used to screen for
          an analyte and  the analyte is not detectable above the
          target detection limit, the laboratory can report the
          results of this analysis  and re-analysis on the isomer
          specific  column will not  be required.  If the detection
          limit exceeds the target  detection limit, it will be
          left to the discretion of the laboratory as to whether
          the extract will be re-analyzed on the isomer specific
          column or a newjportion of sample will be extracted and
          analyzed  in order to meet the target detection limit
          criteria.

9.6  Native Spikes.

     In order to provide  an estimate of method accuracy, the
     laboratory will spike samples  with known concentration of
     2,3,7,8-TCDD and 2,3,7,8-TCDF and then analyze using this
     procedure to determine recovery.


                                      NCASI Method TCDD/F - 88.01


                          449

-------
                             - 32 -


9.6.1     The native spike recovery is calculated as follows:

                                   
-------
                             - 33 -


9.7.1     The relative percent difference (RPD)  is calculated as
          follows:
               RPD =
                         Si - S2
x  100
                       (Si + S2)/2

               where:

               S^ and S2 represent sample and sample duplicate
               results.

9.7.2     The client will designate the samples and frequency for
          duplicate or native spike duplicate determinations.  It
          is recommended that the minimum frequency of duplicate
          determinations is one sample in ten.

9.7.3     The relative percent difference between sample duplicate
          determinations must be < 50 percent.

9.7.4     The relative percent difference between matrix spike
          duplicates shall be < 50 percent.

9.8  For reliable detection and quantitation of 2,3,7,8-TCDF, the
     molecular ion of hexachlorodiphenyl ether must be monitored.
     The fragment ions of the hexachlorodiphenyl ether if present,
     will cause interferences for the ions monitored for 2,3,7,8-
     TCDF.  If a positive response on the hexachlorodiphenyl ether
     ion is noted at the 2,3,7,8-TCDF retention time, then the
     peak should be flagged and the report should indicate "sample
     not analyzable due to HCDPE interference".

10.  Extraction and Cleanup Procedures

10.1 Extraction procedures.

     Prior to extraction, all sludge, pulp and ash samples should
     be dried and blended according to the NCASI DIOXIN PROGRAM
     SAMPLE PREPARATION/PROCESSING PROTOCOL (Revision 4 or later).

10.1.1    Wastewater Samples.  If an aliquot of the sample is to
          be analyzed, proceed as described in step 10.1.1.1.  If
          the entire contents of the sample bottle are to be used
          for the analysis, proceed directly to step 10.1.1.3.

     10.1.1.1  Sub-sampling for analysis of an aliquot of the
               sample.  Mark the liquid meniscus on the side of
               the sample bottle.  Shake the sample bottle
               vigorously and measure out the desired volume of

                                      NCASI Method TCDD/F -  88.01
                          451

-------
                             - 34 -

               sample in a graduated cylinder.  Transfer the sample
               to suitable bottle.  Transfer the remainder of  the
               sample to a new bottle and refrigerate.   Rinse  the
               original sample bottle and the graduated cylinder
               used to measure the sample aliquot with three 20
               mL portions of methylene chloride.  Combine "the
               methylene chloride rinses in a suitably labeled
               bottle.  Refill the original sample bottle to the
               mark and transfer the liquid to a graduated
               cylinder.  Record the original sample volume to
               the nearest 5 mL.

     10.1.1.2  Calculate the percentage of the total sample which
               is to be used for the analysis.   Use the same
               percentage of the total sample volume for the
               volume of the methylene chloride rinses which
               should be included with the final sample extract
               (Section 10.1.1.9).  Proceed to Section 10.1.1.4.

     10.1.1.3  If the entire contents of the bottle are to be
               used for the analysis, mark the liquid meniscus on
               the side of the sample bottle for later
               determination of the sample volume (Section
               10.1.1.7).

     10.1.1.4  Internal Standard Addition.  Add 10 uL of the
               internal standard spiking solution of •   C]^-
               2,3,7,8-TCDD and 13C12-2,3,7,8-TCDF (IS, Table 2)
               to a test tube containing 1 mL of acetone.  Mix
               and transfer the solution into the sample bottle.
               Rinse the test tube with two one mL portions of
               acetone, adding the rinses to the sample.  Shake
               or stir the sample for a minimum of 30 minutes.

     10.1.1.5  Filtration.  Shake the sample vigorously and
               transfer the sample to the pressure filtration
               apparatus.  Add 50 mL deionized water to the sample
               bottle, cap, shake then filter to assure
               quantitative transfer of solids.  Repeat the
               deionized water wash a second time.  Pressure
               filter through Whatman GF/D  (or equivalent) glass
               fiber filter paper collecting the filtrate in the
               original sample bottle.

10.1.1.5.2     Filtration Alternative.  If using the vacuum
               filtration apparatus with the 90mm filter holder,
               perform the filtration as described in 10.1.1.5
               collecting the filtrate and rinsate in a 1L vacuum
               flask. Subsequent methylene chloride rinses in
               section 10.1.1.6 should include the vacuum flask

                                      NCASI Method TCDD/F - 88.01
                              452

-------
                        - 35 -

          in addition to the sample bottle.

10.1.1.6  Liquid-liquid Extraction.  Transfer the filtered
          wastewater into a 2 L separatory funnel.  Add
          sufficient deionized water to the filtered
          wastewater to bring the total sample volume to
          approximately one L.  After removing the glass
          fiber filter, rinse the pressure filtration
          apparatus with 60 mL methylene chloride, collecting
          the solvent rinses in the sample bottle. Seal the
          bottle and shake 30 seconds to rinse the inner
          surface.  Transfer the solvent to the separatory
          funnel and extract the sample by shaking the funnel
          for two minutes with periodic venting.  Allow the
          organic layer to separate from the water phase for
          a minimum of 10 minutes.  If the emulsion interface
          between layers is more than one-third the volume
          of the organic layer, the analyst must employ
          mechanical techniques to complete phase separation.
          Drain the extract and repeat this extraction two
          additional times with fresh 60 mL portions of
          methylene chloride.  Combine the methylene chloride
          extracts.

               Note: It is recommended that the analyst
               measure the volume of the solvent recovered.
               If the solvent recovery is less than 85
               percent, the extraction should be repeated a
               fourth time combining the fourth extract with
               the original three.

          Extract the sample one final time using 60 mL
          toluene.  Combine the toluene extract with the
          methylene chloride extracts.  Dry the organic layer
          by pouring through a funnel containing anhydrous
          sodium sulfate into a 'round bottom flask, wash
          with two 15 mL portions of hexane, and concentrate
          the extract solution to a volume of approximately
          15 mLs with a rotary evaporator (heated water baths
          required).

10.1.1.7  If the entire sample was used for the analysis,
          determine the original sample volume by refilling
          the sample bottle to the mark and transferring the
          liquid to a graduated cylinder.  Record the sample
          volume to the nearest 5 mL.

10.1.1.8  Soxhlet extraction of filtered solids.  Assemble
          and pre-extract a soxhlet apparatus for three hours
          with 68:32 ethanol/toluene.  Place the filtered

                                 NCASI Method TCDD/F - 88.01
                        453

-------
19,
                             -  36  -

               solids in the pre-extracted soxhlet thimble and
               break up the filter as much as possible with a
               spatula.  Place the thimble in a soxhlet charged
               with 96 mL toluene and add 204 mL ethanol to the
               soxhlet by passing it through the thimble.   Extract
               the filter for a minimum of 16 hours.

     10.1.1.9  Combine the methylene chloride extract concentrate
               from 10.1.5 with the ethanol/toluene extract.  If
               the sample was subdivided as described in Section
               10.1.1.1, measure out a volume of the methylene
               chloride sample bottle rinsate which corresponds
               to the proportion of the original sample which was
               used for the analysis and add it to the combined
               ethanol/toluene-methylene chloride extract.  Add
                  ul* of the cleanup recovery spiking standard of
                 rCl4~2,3,7,8-TCDD and ca 100 \LL of tetradecane
               (as a keeper) to the extract.  Concentrate to the
               tetradecane to remove the acetone and methylene
               chloride.  Using 5 mL of ethanol, redissolve the
               residue and pipet into 100 mL of hexane.  Proceed
               to Section 10.2.

10.1.2 Sludge Samples

     10.1.2.1  Assemble and pre-extract a soxhlet apparatus,
               including the thimble, for three hours with 68:32
               ethanol/toluene.  Discard the solvent and allow
               the thimble to dry.

     10.1.2.2  Weigh a representative (2g to lOg) portion of the
               sample into the pre-extracted soxhlet thimble.

     10.1.2.3  Internal Standard addition.  Add 10 ui of the
               internal standard spiking solution of   C^-
               2,3,7,8-TCDD and 13C12-2,3,7,8-TCDF (IS, Table  2).

     10.1.2.4  Soxhlet extraction. Place the soxhlet thimble
               containing the sample into the pre-extracted
               Soxhlet extraction apparatus charged with 68:32
               ethanol/toluene and extract for a minimum of 16
               hours.  Add 10 uL of the cleanup recovery standard
               spiking solution (37Cl4-2,3,7,8-TCDD) and ca 100
               UL of tetradecane (as a keeper) to the extract.
               Concentrate to the tetradecane.  Using 5 mL  of
               ethanol, redissolve the residue and pipet into  100
               mL of hexane.  Proceed to Section 10.2.
                       NCASI Method TCDD/F - 88.01
            454

-------
                             - 37 -


10.1.3  Pulps

     Partially bleached,  fully bleached and most paper products
     pulp samples are extracted as follows:

     10.1.3.1  Assemble and pre-extract a soxhlet apparatus,
               including the thimble,  for three hours with 95
               percent ethanol.  Discard the solvent and allow
               the thimble to dry.

     10.1.3.2  Weigh a representative  10 g portion of the sample
               into the pre-extracted  soxhlet thimble.

     10.1.3.3  Internal Standard addition.  Add 10 uL of the
               internal standard spiking solution of 13C^2-
               2,3,7,8-TCDD and 13C12-2,3,7,8-TCDF (IS, Table  2).

     10.1.3.4  Soxhlet extraction.  Place the soxhlet thimble
               containing the sample in the pre-extracted Soxhlet
               extraction apparatus charged with 95 percent
               ethanol and extract for a minimum of 16 hours.
               Add 10 uL of the cleanup recovery spiking solution
               of 37Cl4-2,3,7/8-TCDD and ca 100 uL of tetradecane
               (as a keeper) to the extract.  Concentrate to the
               tetradecane.  Using 5 mL of ethanol, redissolve the
               residue and pipet into  100 mL of hexane.  Proceed
               to Section 10.2.

10.1.4  Ash Samples

     10.1.4.1  Assemble and pre-extract a soxhlet apparatus,
               including the thimble,  for 3 hours with benzene or
               toluene.  Discard the solvent and allow the thimble
               to dry.

     10.1.4.2  Weigh a representative  (5g to lOg) portion of  the
               sample into the pre-extracted soxhlet thimble.   Add
               an equivalent weight of pre-extracted anhydrous sodium
               sulfate and mix into the sample.

     10.1.4.3  Internal Standard addition.  Add 10 uL of the  internal
                ftandard spiking solution of 13C^2-2,3,7,8-TCDD and
                3C12-2,3,7,8-TCDF (IS, Table 2).

     10.1.4.4  Soxhlet extraction.  Place the soxhlet thimble  contain-
               ing the sample in the pre-extracted Soxhlet extraction
               apparatus charged with benzene or toluene and extract
               for a minimum of 16 hours.  Add 10 uL of the cleanup
               recovery spiking solution of 3'Cl4-2,3,7,8-TCDD to the

                                      NCASI Method TCDD/F - 88.01
                               455

-------
                              -38 -

               extract and proceed to the Section 10.2.

10.2 Extract Cleanup Procedures.

10.2.1    Sulfuric Acid/Sodium Hydroxide Partitioning.

     10.2.1.1  Partition the solvent against 50 mL of concentrated
               sulfuric acid.  Shake for two minutes.  Remove and
               discard the aqueous layer (bottom).  Repeat the
               acid washing until no color is visible in the
               bottom layer (perform acid washings a maximum of
               four times).

     10.2.1.2  Partition the extract against 50 mL of distilled
               water.  Shake for two minutes.  Remove and discard
               the aqueous layer (bottom).

     10.2.1.3  Partition the solvent against 50 mL of 10 N NaOH.
               Shake for two minutes.  Remove and discard the
               bottom aqueous layer (perform the base washings a
               maximum of four times).

     10.2.1.4  Partition the solvent against 50 mL of distilled
               water.  Shake for two minutes.  Remove and discard
               aqueous layer (bottom).

     10.2.1.5  Dry the organic layer by pouring through a funnel
               containing anhydrous sodium sulfate into a round
               bottom flask, wash with two 15 mL portions of
               hexane, add ca 100 uL of tetradecane and
               concentrate the extract solution to the tetradecane
               with a rotary evaporator (heated water bath),
               making sure all traces of benzene (or toluene) are
               removed.  (Use of blowdown with an inert gas to
               concentrate the extract is also permitted).

10.2.2    NaOH-silica:silica:H2SO4rsilica/Acid alumina columns

     10.2.2.1  Place a glass wool plug in a 15 mm i.d.
               chromotography column followed by 1 g silica gel,
               4 g 33% 1 N sodium hydroxide/silica gel, 1 g silica
               gel, 8 g 44% sulfuric acid/silica gel,  2 g silica
               gel, and top with 1 cm sodium sulfate.  Place a
               glass wool plug in an 11 mm i.d. chromatography
               column followed by 6 g acid alumina and top with 1
               cm sodium sulfate.  Pre-rinse both columns with
               hexane and place the 15 mm i.d. column  above the
               11 mm i.d. column so that the eluant from  the top
               column goes directly onto the bottom column.


                                      NCASI Method TCDD/F - 88.01
                            456

-------
                             - 39  -


     10.2.2.2  Dissolve the sample residue in 2  mL of  hexane  and
               apply the hexane solution to the  top column.   Elute
               with 120 mL hexane.  (Discard this eluate).

     10.2.2.3  Remove the top column and elute the bottom column
               with 20 mL hexane.   (Discard this eluate).

     10.2.2.4  Elute with 20 mL 20% methylene chloride/hexane
               (collect this eluate).   Add ca 100 uL tetradecane
               to the eluate and blow down to the tetradecane
               with purified nitrogen.

10.2.3    Charcoal/silica gel column.

          The following cleanup procedure can be used  for bleached
          pulps but is not recommended for sample matrices which
          are more heavily contaminated.

     10.2.3.1  Cut 1 cm off the end tip of a 5 mL (7.5 mm i.d.)
               disposable pipet. Place a glass wool plug at  the
               2.5 mL mark, add 0.65 g charcoal/silica gel packing
               followed by another glass wool plug.

     10.2.3.2  Pre-elute the column with 5 mL toluene  and invert
               the column, continue the pre-elution  in the
               opposite direction  with 2 mL toluene, 2 mL 75:20
               methylene chloride/methanol, 1 mL 1:1 methylene
               chloride/cyclohexane, 5 mL hexane. (Discard
               eluates).

     10.2.3.3  Dissolve the sample residue in 2  mL hexane and
               transfer to the top of the charcoal/silica column.
               Rinse the sample vial twice with  2 mL of hexane
               and transfer each rinse to the column.  Elute  the
               column with 2 mL 1:1 methylene
               chloride/cyclohexane, 2 mL 75:20  methylene
               chloride/methanol.  (Discard eluates).

     10.2.3.4  Turn the column over and elute the column with 15
               mL of toluene. (Collect the eluate).

     10.2.3.5  Add 10 uL of tetradecane and concentrate the  sample
               with a stream of purified nitrogen or by vacuum
               evaporation. Transfer the sample  quantitatively
               into a 2 mL cone shaped vial and  evaporate to the
               tetradecane with a  stream of purified nitrogen. If
               the optional additional cleanup procedure described
               in section 10.3 is  used, proceed  directly to  that
               clean-up. If the extract is ready for analysis,

                                      NCASI Method TCDD/F -  88.01
                                 457

-------
                        - 40 -

          add 10 uL of the recovery standard (^C^-l>2,3,4-
          TCDD) to the extract and mix thoroughly.  Store the
          extracts protected from light.

10.2.4    Option to Charcoal/silica gel column 10.2.3.

     The following cleanup procedure is recommended for all
     wastewater, sludge and process pulp extracts.  It
     provides a higher capacity column for heavily
     contaminated sample extracts.

10.2.4.1  Cut 2 cm off the tip end of a 10 mL (8 mm i.d.)
          disposable sealogical pipet. Place a glass wool
          plug at the 1.0 mL mark, add 1.0 g of the 5 percent
          AX-21 carbon/silica gel packing followed by another
          glass wool plug.

10.2.4.2  Pre-elute the column with 5 mL 1:1 methylene
          chloride/cyclohexane and invert the column.
          Continue the pre-elution  in the opposite direction
          with another 5 mL 1:1 methylene
          chloride/cyclohexane. (Discard eluates).

10.2.4.3  Dissolve the sample residue in 1 mL hexane and
          transfer to the top of the AX-21 carbon/silica
          column. Rinse the sample vial twice with 2 mL of
          1:1 methylene chloride/cyclohexane and transfer
          each rinse to the column. Elute the column with an
          additional 6 mL 1:1 methylene chloride/cyclohexane
          and 5 mL 75:20:5 methylene chloride/methanol/
          benzene. (Discard eluates).

10.2.4.4  Turn the column over and elute the column with 25
          mL of toluene. (Collect the eluate).

10.2.4.5  Add 10 uL of tetradecane and concentrate the sample
          with a stream of purified nitrogen or by vacuum
          evaporation. Transfer the sample quantitatively
          into a 2 mL cone shaped vial and evaporate to the
          tetradecane with a stream of purified nitrogen. If
          the optional additional cleanup procedure described
          in section 10.3 is used, proceed directly to that
          clean-up. If the extract is ready for analysis,
          add 10 uL of the recovery standard (13C12-1,2,3,4-
          TCDD) to the extract and mix thoroughly. Store the
          extracts protected from light.
                                 NCASI Method TCDD/F - 88.01


                     458

-------
                             - 41 -

10.3 Optional Additional Clean-up.

     The following additional cleanup procedure can be used for
     samples which have chemical interferences that cause
     unacceptable detection limits.

10.3.1    Place a glass wool plug in an 11 mm c.d.  chromatography
          column followed by 5 g activated basic alumina and top
          with 1 cm of sodium sulfate.

10.3.2    Dissolve the residue in 2 mL hexane and apply to the
          top of the column.

10.3.3    Elute with 20 mL hexane (discard eluate).  Elute with 8 mL
          3% methylene chloride/hexane (save eluate).  Elute with 35
          mL 50% methylene chloride/hexane.  Collect the eluate.

10.3.4    Transfer the extract quantitatively into a 2 mL cone
          shaped vial.  Add 10 uL tetradecane and concentrate to
          the tetradecane with a stream of purified nitrogen.
          Add 10 uL of the recovery standard (13C12-1,2,3,4-TCDD)
          to the extract and mix thoroughly.  Store extracts
          protected from light.

11.  Analytical Procedures

11.1 Inject a 2 uL aliquot of the.extract into the GC, operated
     under the conditions previously used (Section 7.1) to produce
     acceptable results with the performance check solution.

11.2 Acquire SIM data according to Section 5.1.3.  Use the same
     acquisition and mass spectrometer operating conditions
     previously used to determine the relative response factors
     (Sections 8.1.4 through 8.4.4).

          NOTE:  A selected ion current profile (SICP) for the
          lock-mass ion will also be recorded.  It is recommended
          that the lock-mass ion SICP be examined for departure
          of the instrument's basic sensitivity and stability
          that could affect the measurements.

11.3 Identification Criteria.

     For a gas chromatographic peak to be identified as a 2,3,7,8-
     TCDD or 2,3,7,8-TCDF, it must meet all of the following
     criteria:
                                      NCASI Method TCDD/F - 88.01
                            4 5 3

-------
                             - 42 -

11.3.1    Relative Retention Times.

     11.3.1.1  The retention time (at maximum peak height)  of the
               sample components (i.e., the two ions used for
               quantitation purposes) must be within -1 and +3
               seconds of the retention time of the peak for the
               isotopically labeled internal standard at m/z
               corresponding to the first characteristic ion (of
               the set of two; Table 5 ) to obtain a positive
               identification .

     11.3.1.2  The ion current responses for both ions used for
               quantitative purposed (e.g., for 2,3,7,8-TCDD: m/z
               319.8465 and 321.8936) must reach maximum
               simultaneously (± 1 second).

     11.3.1.3  The ion current responses for both ions used for
               the labeled standards (e.g., for 13C-TCDDs:  m/z
               331.9368 and 333.9339) must reach maximum
               simultaneously (± 1 second).

11.3.2    Ion Abundance Ratios

     11.3.2.1  The integrated ion current for the two ions of the
               native analytes used for quantitation purposes
               must have a ratio (M+/M+2*) within ±15 percent of
               the theoretical value (i.e. 0.65 to 0.89).

     11.3.2.2  The integrated ion current for the two ions of the
               13-        internal standards (including ^    -
               2,3,7,8-TCDD,   C12~1'2'3'4~TCDD and   C12-
               2,3,7,8-TCDF) must have a ratio (M+/M+2+) within ±
               15 percent of the theoretical value (i.e. 0.65 to
               0.89) .

11.3.3    Signal-to-noise Ratio.  All ion current intensities
          must be > 2.5 times noise level for positive identific-
          ation of 2,3,7,8-TCDD or 2, 3 ,7 , 8-TCDF.

11.3.4    Internal Standard Recoveries.  Internal standard
          recoveries will be reported based upon the data obtained
          for the analysis used for the isomer specific
          determination of the analyte.  The internal standard
          recovery objective is 40 to 120 percent.

     11.3.4.1  If the internal standard recoveries are between 40
               and 120 percent, no further criteria or notations
               will be required.
                                      NCASI Method TCDD/F - 88.01
                             460

-------
                             - 43 -

     11.3.4.2  If the internal standard recoveries are less than
               40 percent but greater than 20 percent, the  ion
               current intensities for the internal standard ions
               must be >10 times the noise level and the S/N
               calculation must appear on the SIM trace above the
               GC peak in question.  If the internal standard S/N
               is not greater than 10:1, corrective steps will be
               taken and the sample will be rerun.  All results
               with internal standard recoveries between 20 and
               40 percent will be reported with the internal
               standard recovery given in brackets next to  the
               concentration detected or the detection limit (i.e.
               [30%]).

     11.3.4.3  If the internal standard recoveries are less than
               20 percent, the source of the low recovery will be
               determined, corrective steps will be taken and the
               sample will be rerun.  If the recoveries are still
               unsatisfactory, the results will be reported as
               "Detected - not quantifiable" where the
               identification criteria for the native compound
               were met, and as "analytical difficulties" where
               the identification criteria for the native compound
               were not.

12.   Calculations

12.1 Quantitation of 2,3,7,8-TCDD and 2,3,7,8-TCDF detected in
     samples.

     For gas chromatographic peaks that have met the criteria
     outlined in Section 11.3, calculate the concentration  of the
     2,3,7,8-TCDD or 2,3,7,8-TCDF using the formula:

                         Ax  x  QlS
                     AIS  x  W  x  RRF(n)

     where:

     Cx = concentration of unlabeled TCDD/TCDF in pg/g,

     Ax = sum of the integrated ion abundances of the quantitation
     ions  (Table 5) for the native TCDD/TCDF

     Ajg = sum of the integrated ion abundances of the quantita-
     tion ions  (Table 5) for the labeled internal standards,

     QlS = quantity, in pg (i.e. 2000 pg), of the internal
     standard added to the sample before extraction,

                                      NCASI Method TCDD/F - 88.01
                               461

-------
                             - 44 -


     W = weight, in g, of sample (solid or liquid),  and

     RRF(n) = calculated mean relative response factor from the
     initial calibration for the analyte [RRF(n) with n = 1-2:
     Section 8.1.9] .

12.2 Calculate the percent recovery of the internal  standards
     standard measured in the sample extract, using  the formula:
                                    AIS x
          Internal standard
          percent recovery  =  - x 100
                                    x ARS
     where:

     ATS = sum of the integrated ion abundances of the quantita-
     tion ions ( Table 5 ) for the labeled internal standard,

     Apg = sum of the integrated ion abundances of the quantita-
     tion ions (Table 5) for the labeled recovery standard;

     QlS = quantity, in pg (i.e. 2000 pg), of the internal
     standard added to the sample before extraction, and

     QRS - quantity, in pg (i.e. 2000 pg) , of the recovery
     standard added to the cleaned-up sample residue before
     HRGC/HRMS analysis, and

     RRF(m) = calculated mean relative response factor for the
     labeled internal standard relative to the recovery standard.
     This represents the mean obtained in Section 8.1.10.

12.3 Calculate the percent recovery of the cleanup recovery
     standard measured in the sample extract, using the formula:
        Cleanup Recovery                A^s x QRS
             Standard         =  - x 100
        percent recovery           Qcs x ARS x RRF(o)

     where :

     AQS = sum of the integrated ion abundances of the quant ita-
     tion ions (Table 5) for the cleanup recovery standard,

     ApS = sum °f tne integrated ion abundances of the quantita-
     tion ions { Table 5 ) for the labeled recovery standard;
                                      NCASI Method TCDD/F - 88.01
                            4B2

-------
                             - 45 -

         = quantity»  in P9 (i.e.  1000 pg) ,  of  the cleanup recovery
     standard added to the sample before clean up,  and
         = quantity,  in pg (i.e.  2000 pg) ,  of  the recovery
     standard added to the cleaned-up sample residue before
     HRGC/HRMS analysis, and
     RRF(o)  = calculated mean relative response factor for the
     cleanup recovery standard relative to the recovery standard.

     This represents the mean obtained in Section 8.1.11.

12.4 If the  concentration in the 20 uL final extract of either
     2,3,7,8-TCDD or 2,3,7,8-TCDF exceeds the upper method
     calibration limits  (MCL) listed in Table I, the linear range
     of response versus  concentration may have been exceeded.
     The extract should  be diluted to an appropriate volume and
     reanalyzed.  The report should include a notation of  the
     dilution used for the analysis.

12.5 Sample  Specific Estimated Detection Limit.

     The sample specific estimated detection limit (EDL) is the
     concentration of a  given analyte required to produce  a signal
     with a  peak height  of at least 2.5 times the background
     signal  level.

12.5.1    Samples presenting a response for either ion for native
          2,3,7,8-TCDD or 2,3,7,8-TCDF that is less than 2.5 times
          the background level.

     12.5.1.1  By using the expressions of EDL below, calculate
               an EDL for each isomer characterized by the absence
               of a response (i.e., S/N < 2.5).  The background
               level is  determined by measuring the range  of the
               noise (peak-to-peak) for the M+2"1" ion (Table 5) of
               a particular analyte, in the region of the  SICP
               trace corresponding to the elution of the internal
               standard, multiplying that noise height by 2.5,
              . and relating the product height to an estimated
               concentration that would produce that product
               height.

               Calculate the EDL using the following formula:

                                   2.5 x HtxM+2 x QIS
                           EDL =
                                  HtISM+2 x W x RRF(n)
                                      NCASI Method TCDD/F - 88.01
                               468

-------
                         - 46 -

          where:

          EDL = estimated detection limit for 2,3,7,8-TCDD
          or 2,3,7,8-TCDF.

          HtxM+2 = noise level height of the M+2+ ion (Table
          5.) for the native TCDD/TCDF

          HtlM+2 _ noiSe level height of the M+2+ ion (Table
          5_) for the labeled internal standard

          and W, RRF(n), and Qjs retain the same meanings as
          defined in Section 12.1.

12.5.1.2  When only one of the guantitation ion signals is
          below 2=5 tiroes the background level and the other
          ion signal is above, consider the positive ion
          channel as an interfering peak.  Calculate the
          estimated detection limit based on the single
          undetected ion using the corresponding M+ or M+2+
          internal standard ion as follows:

                              2.5 x Htx x QIS
                      EDL =
                             Htls x W x RRF(n)

          where:

          EDL = estimated detection limit for 2,3,7,8-TCDD
          or 2,3,7,8-TCDF.

          Htx = the noise level height of the undetected
          quantitation ion M+ or M+2+ (Table 5) for the
          native TCDD/TCDF

          Htjs = the noise level "height of the corresponding
          M+ or M+2+ ion (Table 5) for the labeled internal
          standard

          and W, RRF(n), and QrS retain the same meanings as
          defined in Section 12.1.

12.5.1.3  When the response of a signal having the same
          retention time as the analyte has a S/N in excess
          of 2.5 and does not meet any of the other
          qualitative identification criteria, calculate the
          estimated detection limit according to the equation
          given in Section 12.1 except using peak heights.
                                 NCASI Method TCDD/F - 88.01
                         464

-------
                             - 47 -

13.  Deliverables

13.1 Initial Calibration Data.   The laboratories standard
     operating procedure for the preparation of standards must be
     submitted prior to or simultaneously with the first
     submission of initial calibration data. The SOP need.not be
     submitted with subsequent  initial calibration data sets
     unless the procedure has been revised.   The information
     pertaining to the initial  calibration and described in the
     remainder of this Section  must be submitted prior to or
     simultaneously with the results of any sample analyses.  If
     the initiation process must be repeated, the information
     pertaining to the repeat analyses must be submitted prior to
     or simultaneously with the results of any sample analyses.

13.1.1    Documentation of the  mass spectrometer resolution as
          described in Section  7.2.4.

13.1.2    Documentation of the  isomer specificity as described in
          Section 7.1.2.

13.1.3    All SICP and area tables for all standard analyses
          presented in the format described in Section 13.3.

13.1.4    Tabular summaries containing the following information:

     13.1.4.1  The source, identification code and concentration
               of all standards.

     13.1.4.2  The ion ratios for native analytes and labeled
               standards for all calibration standard analyses.

     13.1.4.3  The native response factors relative to the
               appropriate internal standard (RRF(n)) for each
               calibration standard analysis.

     13.1.4.4  The relative response factors for each internal
               standard relative to the recovery standard (RRF(m))
               for each calibration standard analysis.

     13.1.4.5  The relative response factor for the cleanup
               recovery standard relative to the recovery standard
               (RRF(o)) for each calibration standard.

     13.1.4.6  The mean relative response factors for all
               determinations (RRF(n), RRF(ra), RRF(o)).

     13.1.4.7  The relative standard deviation (RSD) of each
               relative response factor (including RRF(n), RRF(m),
               RRF(o)) for all  calibration analyses.

                                      NCASI Method TCDD/F - 88.01
                             4G5

-------
                              - 48-


     13.1.4.8  The ±20 percent control limits to be used as
               criteria to determine the acceptability of the
               daily calibration checks.

13.2 Sample Data Packs.  Each submission of sample analysis
     results must include all information described in this
     Section.

13.2.1    A cover letter or case narrative describing the
          analytical procedure, identifying quality assurance
          samples, analyses conducted, appropriate references to
          Purchase Order Number, etc. should be submitted.

13.2.2    Pertinent copies of chain-of-custody records should be
          included.

13.2.3    Documentation of isomer specificity for each column and
          for each day analyses were performed as described in
          Section 7.1.2.

13.2.4    Documentation of mass spectrometer resolution for each
          instrument and day analyses were performed as described
          in Section 7.1.4.

13.2.5    Daily calibration data including the dates and times
          analyzed (alternatively, a copy of the appropriate page
          of an analysis log book which shows the sequence and
          times of injections can be submitted), instrument used,
          the identification code for the standards used cross-.
          referencing the initial calibration data set, all SICP
          formatted as described in Section 13.3, all native and
          labeled standard ion ratios, relative response factors
          (RRF(n), RRF(n) and RRF(o)), and the percent relative
          difference between the daily- calibration check relative
          response factors and the initial mean relative response
          factor (RRF(n), RRF(m) and RRF(o)).

13.2.6    Method Blank data including the dates prepared, sample
          size or volume, all SICP formatted as described in
          Section 13.3, tabular summaries of the estimated
          detection limits, internal standard ion ratios, internal
          standard recoveries and cleanup recovery standard
          recoveries and, where analyte(s) are detected, the
          analyte ion ratios and concentrations) detected.

13.2.7    Sample results including client sample identification
          number, contract laboratory identification number,
          sample weight, date sample preparation started, all
          SICP formatted as described in Section 13.3, tabular
                                      NCASI Method TCDD/F - 88.01
                                  4GG

-------
                              -49 -

          summaries of internal standard and analyte ion ratios,
          internal standard and cleanup recovery standard
          recoveries and concentrations detected or estimated
          detection limits where applicable.

13.2.8    Laboratory duplicate results including client sample
          identification number, contract laboratory
          identification number, sample weight,  date sample
          preparation started, all SICP formatted as described in
          Section 13.3,  tabular summaries of internal standard
          and analyte ion ratios, internal standard and cleanup
          recovery standard recoveries and concentrations detected
          or estimated detection limits where applicable and the
          calculated relative percent difference.

13.2.9    Native Spike recovery results including client sample
          identification number, contract laboratory
          identification number, sample weight,  date sample
          preparation started, exact spike level expressed as a
          concentration in the sample, all SICP formatted as
          described in Section 13.3 tabular summaries of internal
          standard and analyte ion ratios, internal standard and
          cleanup recovery standard recoveries,  concentrations
          detected or estimated detection limits where applicable
          and the native spike recovery.

13.3 Raw Data/SICP Reporting Format.

13.3.1    The report for isomer specific TCDD results should
          include separate pages plotting the following SICPs:

     13.3.1.1  M+ and M+2+ for native TCDD plus M+2+ for the
               internal standard.

     13.3.1.2  M+ and M+2+ for the TCDD internal standard plus M+
               for 37Cl4-2,3,7,8-TCDP.

     13.3.1.3  The lock mass check channel.

     13.3.1.4  M+ for native TCDD

          If the DB-5 column is used to screen for 2,3,7,8-TCDF
          and it is not detectable at or below the target
          detection limit, then the following additional SICPs
          should be included in the report:

     13.3.1.5  M+ and M+2+ for native TCDF plus M+2+ for the
               internal standard.

     13.3.1.6  M+ and M+2+ for the TCDF internal standard.

                                      NCASI Method TCDD/F -  88.01
                                 
-------
                             - 50-

     13.3.1.7  M+ for native TCDF

13.3.2    The report for TCDF results should include separate
          pages plotting the following SICPs:

     13.3.2.1  M+ and M+2+ for native TCDF plus M+2+ for the
               internal standard.

     13.3.2.2  M+ and M+2+ for the 13C12-1,2,3,4-TCDD recovery
               standard.

     13.3.2.3  M+2+ for native TCDF, M+ for hexachlorodiphenyl
               ether and the lock mass check channel.

     13.3.2.4  M+ for native TCDF

          If the DB-225 column is used to screen for 2,3,7,8-TCDD
          and it is not detectable at or below the target
          detection limit, then the following additional SICPs
          should be included in the report:

     13.3.2.5  M+ and M+2+ for native TCDD plus M+2+ for the
               internal standard.

     13.3.2.6  M+ and M+2+ for the TCDD internal standard plus M+
               for 37Cl4-2,3,7,8-TCDD.

     13.3.2.7  M+ for native TCDD

13.3.3    Each SICP plot should include in the header the client
          sample identification number, the contract laboratory
          identification number, the date, time and the instrument
          and column type (including internal diameter) used for
          the analysis.  If the date and time cannot be included
          in the header, the report should include a copy of the
          appropriate page of an analysis log book which shows
          the sequence and times of injections.

13.3.4    Noise levels used in the estimation of detection limits
          should be marked on all appropriate SICPs.

13.3.5    Noise levels and signal to noise level calculations
          should be recorded on all internal standard and cleanup
          recovery standard SICPs when the internal standard or
          cleanup recovery standard recoveries are below the
          target 40 percent.

13.3.6    Noise levels and signal to noise level calculations
          should be recorded on all native SICPs when the signal
          to noise is S 5 to 1.

                                      NCASI Method TCDD/F - 88.01
                                4G8

-------
TABLE 1
c:
            2.3,7,8-TCDD
            Lower MCLa
            Upper MCL

            2,3,7,8-TCDF
            Lower MCLa
            Upper MCL
           Recommended
           Sample Size
           IS Spiking Levels  (ng)
           13C12-2,3,7/8-TCDD
           13C12-2,3,7,8-TCDF
                                           TYPES OF MATRICES, SAMPLE SIZES  AND  2,3,7,8-TCDD
                                           AND  2,3,7,8-TCDF BASED METHOD  CALIBRATION  LIMITS
                                             Wastewaters
         Treated
         Effluent
       0.01 to 0.04
         10 to 40
Process
 Sewers
Sludges
Pulps
  1
Ash
0.01 to 0.1     1 to 5      1      1 to 2
10 to 100   1000 to 5000  1000  1000 to 2000
       0.01 to 0.04   0.01 to 0.1     1 to 5      1      1 to 2
         10 to 40     10 to 100   1000 to 5000  1000  1000 to 2000
      1.0 to 0.25L   1.0 to 0.10L   10 to 2g
          2.0 ng
          2.0 ng
           Cleanup Recovery  Spiking  Level
           37Cl4-2,3,7,8-TCDD              1.0  ng

           Recovery Standard Spiking Level
           13C12-1,2,3,4-TCDD              2.0  ng
           Final Extract Vol.  (y.L)
                                            20
  2.0 ng
  2.0 ng


  1.0 ng
 2.0 ng
 2.0 ng


 1.0 ng
 lOg    lOg to 5g
                     i
                     M
 2.0 ng    2.0 ng    (
 2.0 ng    2.0 ng


 1.0 ng    1.0 ng
                        2.0 ng       2.0 ng      2.0 ng    2.0 ng

                         20           20          20        20

-------
                               -  52  -
   TABLE 2   COMPOSITION OF THE SAMPLE FORTIFICATION SOLUTIONS
     Analvte
  13C12-2,3,7,8-TCDD
  Sample Fortification Solutions  (IS)
    Concentration  (pg/uL;  Toluene)

                200

                200
       (RS)
      Analyte
  13C12-1,2,3,4-TCDD
  Recovery Standard Spiking Solution
            Concentration
  	(pg/uL; Tetradecane)	

                200
    Analyte
37Cl4-2,3,7,8-TCDD
Cleanup Recovery Spiking Solution (CRS)
            Concentration
	(pg/uL; Toluene)	
                100
                                      NCASI Method TCDD/F - 88.01
                                470

-------
                       53   -
TABLE 3  GC COLUMN PERFORMANCE CHECK STANDARDS
        2,3,7,8-TCDD COLUMN PERFORMANCE
                CHECK STANDARD
       (for use with 60 m DB-5 columns)

                 1,2,3,7-TCDD
                 1,2,3,8-TCDD
                 1,2,3,9-TCDD
                 2,3,7,8-TCDD
        2,3,7,8-TCDF COLUMN PERFORMANCE
                CHECK STANDARD
      (for use with 30 m DB-225 columns)

                 2,3,4,7-TCDF
                 1,2,3,9-TCDF
                 2,3,7,8-TCDF
                             NCASI Method TCDD/F - 88.01
                       471

-------
                              -  54  -
            TABLE 4   CALIBRATION STANDARD SOLUTIONS
                          CONCENTRATION (pg/uL)
	Compound

   Native

2,3,7,8-TCDD

2,3,7,8-TCDF
Internal Standards

13C12-2,3,7,8-TCDD

13C12-2,3,7,8-TCDF


Recovery Standard

13C12-1,2,3,4-TCDD


Cleanup Recovery Standard

37Cl4-2,3,7,8-TCDD
CSS
1
0.5
0.5
2 3
5 50
5 50
4
250
250
5
500
500
                          100
100      100     100    100

         100     100   • 100
                                             100    100     100
                                              10
                 25
50
                                      NCASI Method TCDD/F - 88.01
                            472

-------
                             -  55  -
        TABLE 5  QUANTITATION,  INTERFERENCE AND LOCK MASS IONS
MONITORED
Accurate
Descriptor Mass
TCDF 303.9016
305.8987
315.9419
317.9389
375.8364
318.9793
TCDD 319.8965
321.8936
331.9368
333.9339
327.8855
318.9793
FOR HRGC/HRMS ANALYSIS OF TCDD/TCDFs
Ion
ID
M
M+2
M
M+2
M+2
LOCK
M
M+2
M
M+2
M
LOCK
Elemental
Composition
C12H435C140
C12H435C1337C10
13C12H435C140
13C12H435C1337C10
C12H435C160
PFK
C12H435C1402
C12H435C1337C102
13C12H435C1402
13C12H435C1337C102
C12H437C1402
PFK
Analyte
TCDF
TCDF
TCDF ( S ) a
TCDF ( S ) a
HxCDPEb

TCDD
TCDD
TCDD(S)a
TCDD(S)a
TCDD(SS)C

alnternal and/or Recovery Standard
bRexachlorodiphenylether interference monitoring ion
GCleanup Recovery Standard
                                      NCASI Method TCDD/F  -  88.01
                                -4^-10
                                4. / O

-------
                         -   56  -
      TABLE 6   RECOMMENDED GC OPERATING CONDITIONS
 Column Coating
 Film Thickness
 Column Dimension
 Injector Temperature
 Splitless Valve Time
 Interface Temperature
                DB-5
                .25 urn
                60 m x .25 mm id
                280°C
                40 Sec
                Function Of The Final Temp.
                   Temperature Program
Init. Temp.
   190
Init. Hold
Time  (min)
   1 min.
  Temp
  Ramp
(°C/min)
 Final
 Temp
 (°C)

  260
   Final
 Hold Time
  (min.)

    20
 Column Coating
 Film Thickness
 Column Dimension
 Injector Temperature
 Splitless Valve Time
 Interface Temperature
                DB-225
                .25 urn
                30 m x .25 mm id
                2508C
                40 Sec
                Function Of The Final Temp.
                   Temperature Program
  Init. Temp.
     190
 Init. Hold
 Time  (min)
   Temp
   Ramp
 (°C/min)
Final
Temp.
 (°C)

 235
  Final
Hold Time
  (min.)

    15
 It is recommended that the analyst bake out the DB-225 column
 between runs.
                                  NCASI Method TCDD/F - 88.01
                       4


-------
(1)
                         -   57   -

                 LITERATURE REFERENCES

"U.S. EPA/Paper Industry Cooperative Dioxin Screening
Study", EPA-440/1-88-025, (March 1988).
(2)   LaFleur,  L.E.,  Ramage,  K.,  Gillespie,  W.J.,  Miille,  M.J.,
     Luksemburg,  W.J.,  Valmores, S.,  "Optimization of Extraction
     Procedures for  the Analysis of TCDD/TCDF in Pulp, Paper Base
     Stocks,  and Pulp Industry Solid Wastes", presented at The
     8th International Symposium on Chlorinated Dioxins and
     Related  Compounds, Umea,  Sweden,(August 1988).
                                  4
                              •*•r;;.
                               ( U

-------
      U.S.  Environmental  Protection Agency

        National  Oioxin  Study - Phase II




Analytical  Procedures and Quality Assurance Plan

   for the  Determination of PCDD/PCDF in Fish
       Environmental Research Laboratory
              6201 Congdon Blvd
              Duluth, MN  55804
               July 26, 1988
                '47C

-------
                                  INTRODUCTION


      This document, "Analytical  Procedures and Quality Assurance Plan fur the

 Determination of PCDD/PCDF in  Fish"  has been drafted in response to the need

 for the Environmental  Research  Laboratory of Duluth (ERL-0) to perform analysis

 for tetrachloro-  to octachloro-  congeners/isomers of polychlorinated cibenzo-

 p-dioxins and dlbenzofurans  (PCOD/PCDF).

      These analyses are  limited  by lack of analytical  standards, however uomer

 specificity may  be  determined using  specially developed standards.   Analytical

 results  will,  therefore,  be  reported as concentration (pg/g) for each gas

 chromatography (GC)  peak  in  a congener class by making the assumption that

 the  response  for  the molecular  ion of all  isomers in that class is  equal to

 the  response  observed  for the isomer for which ERL-D does have a standard.

 The  target minimum  level  of  detection (HLO) for specific PCDD/PCDF  isomers is

 given in  Table 1  below.   This document is meant to be only a guideline for

 analyses  and may  be modified as  needed to satisfactorily analyze any sample.


                                   Table 1

.	PCDD/PCDF	Target Minimum Level  of Detection

     TCDD, TCDF                 :                    1 pg/g
     PeCDD, PeCDF                                   2 pg/g
     HxCDO, HxCDF                                   4 pg/g
     HpCDD, HpCOF                                  10 pg/g
     OCDD. OCDF	40 pg/g	
                               477

-------
                              TABLE OF CONTENTS

   I.   Sample  Preparation
       A.  Grinding
       B.  Extraction
       C.  Anthropogenic Chemical Isolation
       D.  Florisil Chromatography
       E.  PCDO/PCOF Isolation
       F.'  Reagents
      G.  Percent Lipid Determination

 II.  Gas Chromatography/Hass Spectrometry Analyses

III.  Analytical  Standard  Spiking Solutions

 IV.  Quality Assurance/Quality  Control
      A.  General  Procedures  of  Operation
      B.  Instrumental  Quality Checks
      C.  Evaluation of Data
          1.   Accuracy
          2.   Precision
          3.   Signal Quality Assurance  Requirements
      0.   Quality  Assurance Problems and Corrective Actions

 V.   Quantification Procedures
      A.   Method Efficiency
      B.   Quantification of PCOO/PCOF
      C.   Signal Quality
      0.   Quantification Standards
                                        478
                                     i i

-------
I.   Sample Preparation
    A.  Grinding:  Fish tissue  is  ground  frozen  in a stainless steel power
        meat grinder.   The  ground  tissue  is stored at -20°C in solvent
        rinsed glass  jars with  aluminum lined plastic lids.
    U.  Extraction:   Tissue  (20 g)  is  blended with enough anhydrous sodium
        sulfate to dry the  tissue  (100 g).  Two-thirds of the sample is
        placed in a glass Soxhlet  thimble, spiked with Standard Solutions A
        and  8  (Appendix A) and  the  remainder of the sample is added to the
        thimble.  The  sample  is  extracted at least eight hours with 1:1
        mixture of hexane and methylene chloride in a Soxhlet extractor.
        The  solvent is  removed with a  Kuderna-Danish (KD) apparatus, and
        percent lipid  is determined (see I.G.) before proceeding to the
        anthropogenic chemical  isolation step.
   C.   Anthropogenic Chemical  Isolation:   The sample extract is quanti-
        tatively  transferred to a 30 cm x 2.5 cm glass chromatography
       column  (MACRO-columns) fitted with a 300 ml reservoir on top,
       which has been  packed with plug of glass wool  (bottom to top) 2 g
       silica  gel, 2 g potassium silicate, 2 g sodium sulfate, 10 g celite/
       sulfuric acid and 2 g sodium sulfate, and previously washed with
       100 ml hexane.  The column is eluted with 100 ml benzene/hexane
       (5X)  and the element is collected  in a Kuderna-Oanish (KD)
       apparatus.  Isooctane (1.0 ml) is  added and the volume is  reduced
       and transferred to the florisil column.
   D.   Florisil Chromatography: A 1.0 cm  x 20.0 cm glass chromatography
       column  fitted  with a lt)0 ml reservoir is packed with a plug of glass

                                  -1-
                                 473

-------
     wool  (bottom to  top)  5.0 cm  (1.5 g) activated  florisil and  1.0 cm



     sodium sulfate.   The  column  is washed with 20 ml methylene  chloride



     followed  by 10 ml  hexane.  Sample and hexane rinses are quantita-



     tively applied in  small "plugs".  The column is eluted with 2C ml 2%



     methylene chloride/hexane,  nd the eluent discarded.  This wash is



     followed  by 50 ml methylene chloride which flows directly onto the



     micro  carbon/silca gel column for PCDO/PCOF isolation.



 E.   PCDD/PCDF Isolation:  Effluent from the previous step is passed onto



     a 4 nun x 200 mm column (micro-column) containing 300 mg silica gel/



     carbon (see  F.I.) which was fitted with a solvent reservoir.  After



     the sample has almost completely eluted from the micro-column, the



     reservoir is washed with 2 x 2 ml benzene/methylene chloride (25%)



     and the column is finally eluted with an additional  11 ml benzene/



    methylene chloride.  The PCDD/PCOF are then eluted with toluene



     (20 ml) in a reverse flow direction.  The toluene fraction  is



    collected in a pear shaped flask (25 ml) and reduced in volume to



    0.1 ml (60°C).  The sample is transferred to a microvial  using



    toluene to rinse  the flask.  The sample is stored in a freezer



    until  mass spectrometry analyses.



F.  Reagents:



    1.   Solvents:  Only pesticide grade distilled in glass solvents are



        to be  used.   They are:   hexane, isooctane, methylene chloride,



        benzene, toluene, acetone, and  methanol  (Burdick and Jackson,



     .  Fischer  Scientific or  Mallincrodt).






                               -2-
                                 480

-------
 2.   Sodium  Sulfate:  Sodium sulfate (Baker Chemical Company reagent
     grade anhydrous) is baked at 650°C in a furnace for 24 hours,
     cooled,  and  stored in an empty hexane solvent bottle.
 3.   Silica-Gel:  Silica-Gel-60 (Merck-Darmstadt), is Soxhlet
     extracted  eight hours with methanol , air dried for 12 hours,
     and  vacuum oven dried (125°C) for 24 hours.  It is then stored
     in an empty  hexane solvent bottle.  Just before use it is
     activated  at 105°C for 24 hours.
4.   Sulfuric Acid/Celite:  Sulfuric acid (Baker Chemical  Company,
     Ultrex)  (5 ml) is blended in a 250 ml beaker with Celite 545
     (Baker)  (10 g).
5.   Potassium Silicate:  High purity potassium hydroxide (Aldridge
     Chemical Company) (56 g) is dissolved in methanol  (300 ml).
     Silica-gel (100 g)  is added to the mixture and. stirred (1  hour,
     60°C).  The mixture is cooled, the solvent drained, and the
     solids transferred  to a Soxhlet thimble.  The solids are extracted
    with methanol for 4 hours, cooled, air dried in a hood, and
     stored in a solvent rinsed bottled until use.
6.   Silica Gel/Carbon;   Silica Gel-60 (100 g)  (Merck-Darmstadt)  is
     Soxhlet extracted with methanol  (200 ml) for 24 hours, air dried
     in a hood, and further dried in vacuum oven for 24 hours.   AMOCO
     PX-21 Carbon (5 g)  Js added and then blended until uniform in color.
    The Silica Gel/Carbon is stored in a closed jar at room
    temperature until  use.
                           -3-
                            481

-------
    7.  Florisil  60-100 mesh  (Baker  Analyzed)  is  soxhlet  extracted



        with methanol  for 24  hours,  air  dried  and activated  (stored)



        at 120°C.



G.  Percent Lipid  Determination:  The percent  lipid will  be  determined



    for all fish.   To  determine percent  lipid, the sample will be



    extracted as  in Section B of sample  preparation.  After  sample



    concentration,  the KD lower tube is  placed in a 60°C  water bcth



    under  a gentle  stream of dry filtered air.  After any remaining



    solvent has been evaporated, the lower tube and contents are weighed.



    The lipid is then  quantitatively transferred to the macro column as



    described in Section  C of sample preparation.  After  transfer,  the



    empty  lower tube and  boiling chips are weighed.  The  percent lipid



    is  calculated from  the weight-differences..
                               4
                              43

-------
  II.   GC/MS  Analyses
       All  gas  chromatography/mass spectrometry analyses (GC/MS) will be done
       on a Finnigan-HAT 8230 high resolution GC/high resolution MS (HRGC/
       MRMS)  system.   Instrumental parameters are given in Appendix 8.  Prior
       co analyses  each  sample will be spiked with Standard Solution C
       (Appendix  A) and  the  sample volume adjusted to 20 uL.

III.   Analytical Standard Spiking Solution
       Each sample  (20 g) will be spiked with 100 uL of Internal Standard
       Solution A (stable isotope labeled PCOO/PCDF) and Internal  Standard
       Solution B (Appendix  A) (surrogate analytes) prior to extraction, and
       20 uL of Internal Standard Solution C (Recovery Internal  Standard) just
       before MS analysis.   Appendix A provides details of the spiking solutions
      The surrogate analytes are used by the data reviewer to insure that
      calculated MLO values are reasonable.

IV.    Quality Assurance/Quality Control  (QA/QC)
      A.  General Procedures of Operation
          1.   Analysis of samples;  Samples will  be analyzed in sets of
              twelve consisting of:
              a.   Blank:  Method Blank (extraction apparatus) is  prepared in
                  the laboratory and subjected to the same sample preparation
                  procedures as environmental  samples.  The Method Blank will
                  be used in every sample set.
              b.   Fortified Matrix:   Native analyte (100 uL) (Appendix C)
                  are added to a blank sample matrix.  The levels of fortifi-
                  cation of native analytes in the matrix spike will be above
                  the target detection limit, to provide an estimate of the
                  method's  sensitivity,  and for determination of percent
                                      5
                                    488

-------
         accuracy of quantification.   This  sample may be substituted
         with a reference sample  that  has been analyzed by at least
         three labs  and  a mean  value of contamination has been
         established.
     c.   Detection Limit Verification  Sample:  An environmental
         sample with detectable amounts of  native analyte (determined
         from a previous analysis) will be  spiked with native analytes
         (Appendix C) and analyzed with the next sample set.  The
         addition of the QA/QC sample will  be done for only the
         first  three sample sets of any matrix type to establish
         that  the calculated MID is achievable.  If analytical
         results show difficulty in obtaining the MLD, then this
         QA/QC  sample must be in each set.  If no problem is
         experienced, then this QA/QC sample may be dropped.
    d.   Duplicate Sample:  Two separate portions of the same
         environmental sample will be processed and analyzed.
    e.   Environmental Samples:  The total number of environmental
         samples analyzed will  be eight if the Detection Limit
         Verification sample is used, otherwise nine samples will
         be analyzed.
2.  Sample Tracking  and Labeling of Samples
    a.  Logging Incoming Samples:  ERL-D completes the chain of
        custody forms and informs the Sample Control  Center (SCC)
        that samples arrived safely or informs SCC of any problems
        with the samples.  Each sample received  by ERL-D had
        previously been assigned two numbers by the Sample Control
        Center, the  sample number (SCC0)  and an  episode number.
                            6
                                48'<1

-------
         The SCCff number  is  unique  for each sample and provides a
         means for tracking  a  given sample throughout its analysis
         and its  permanent storage at the locker plant.  The samples
         are placed into  freezer  A upon arrival at ERL-Duluth,
         homogenized,  and an aliquot (100-500 g) is placed into
         freezer  B.  After the samples are extracted they are put
         into  freezer  C.  If all  the data meets QA requirements after
         mass  spectral analysis and quantification, the samples are
         transferred to a locker  plant for permanent storage (-20°C).
    b.   logging  and Labeling  Samples During Preparation:  A laboratory
         identification code (lab 10) is randomly assigned to each
         sample in  a set of  twelve at the start of sample preparation.
         The code consists of a letter, A through L, date of
         extraction, and two initials of the sample preparation
         chemist, i.e. A091587ML, and is used to identify the sampl-e.^
         throughout the analysis period.  The SCCJif, lab ID, sample
         description, weight of sample, and amount of analytical
         standards added to  each sample are all recorded in a sample
         preparation log book at the start of extraction.  The lab
         ID  is written on labeling tape which is transferred from
         beaker to  flask during sample preparation.  The lab ID is
        written  into the MS log book along with the mass spectra
         analysis number.
3.  Data System Sample Tracking:  ERL-0 has developed the National
    Oioxin Study (NOS) Phase  II:  Bioaccumulative Pollutant Study
    Sample Tracking Database to  facilitate record keeping and
    summary report generation for each sample on the DEC-VAX 11/785
                            7
                                485

-------
         (Digital  Equipment Corporation).  For each sample, including QA
         samples,  information pertinent to each sample is entered by the
         preparation chemist.  Quantification data (final concentration,
         ion  ratios, percent recovery, MLDs, and signal  to noise) are
         automatically uploaded to the database once all  QA criteria
         have been met.  Appendix 0 is an example of the  NOS database.
             The  first two letters of the SCC number indicate whatl«*»r
         the sample is an Environmental , Method or Matrix Blank, or
         Duplicate Sample.  All  environmental samples begin with the
         letter 0.  The Blank and Duplicate samples begin with the
         letter Q followed by a D or an R for duplicate or reference
         fish sample, respectively.  Appendix E lists the possible cades
         for the SCC number, and matrix type.  Episode numbers for
        Blanks and Fortified Matrix samples are entered  as 0000.
B.  Instrumental  Quality Control
    1.  Gas Chromatograph
        a.  Operation  and Maintenance:  Operation and maintenance of the
            gas chromatograph will be done according to  manufacturer's
            recommendations.
        b.  Column Performance:   GC columns  performance  will  be evaluated
            by:
            i.   Resolution of 1,2,3,4-TCDO from 2,3,7,8-TCDD
                 (Appendix F).
           ii.   Correlation  of  relative  retention time of all
                biosignificant  PCDD/PCDF, specifically calculated by
                relative  retention time  of 13C12l,2,3,6,7,8-HxCDD to
                37
                  C142,3,7,8-TCDO  (Appendix  F).
                                8
                                 486

-------
           iii.   Elution  of all  PCOD/PCDF during analysis will use a


                 GC window defining solution of select PCDD/PCDF


                 (Appendix G.).


     2.   Mass  Spectral  Performance:  The performance of the mass


         spectrometer will  be  evaluated for resolution, sensitivity and


         linearity.  Sens-itivity and linearity will be evaluated using


         calibration standards  (Appendix H).  The mass resolution used


         for these  analyses  will be a minimum of 5000 (10* valley
                                                       *

         definition).   The mass  spectrometer will be tuned each day to


         the required resolution according to the procedures established


         by the  instrument manufacturer.  The calibration curve must be


         linear  over the  range of concentrations used in the calibration


         standards.  The  percent relative standard deviations for the


        mean  response  factors must be "less than 20 percent.


C.  Evaluation  of  Data


    1.  Accuracy:   Accuracy, the degree to which the analytical


        measurement relects the true level present will be evaluated in


        two ways for each sample set.  These are:  the difference of


        measurement of a  PCDD/PCDF isomer added to a blank matrix, or


        difference  of measurement of a PCDO/PCDF from the level in an


        established reference material; and the efficiency for recovery


        of the  internal  standard added for each congener group.  QA


        requirements for  accuracy and method efficiency are provided in


        Appendix I.
                                    40 >-1
                                    C l

-------
     S Accuracy and 2 Method Efficiency are defined as follows:
             % Accuracy  =   measured value   X 100
                 amount native isomer added to blank matrix
    2 Method efficiency  =   measured value   X 100
               amount internal  standard added to each sample
2.  Precision:  Precision, a measure of mutual  agreement amcng
    individual measurements of the same pollutant in replicate
    samples, will be evaluated for each sample  set by the ratio  of
    the difference of duplicate values to their mean value.
    Appendix I provides QA requirements for precision.

    Precision is defined as follows:
    Precision  =   difference between duplicate" values    X 100
                      mean value for the duplicates

3.  Signal  Quality:   The quality of the mass spectral  signals used
    for qualitative  and quantitative analysis will  be evaluated
    using two parameters, the ion intensity ratio for the two ions
    monitored in each congener group, and the signal  to  noise (S/N)
    ratio.   Appendix I provides  QA requirements for  signal  quality.
    In  addition,  qualitative identification will  be based on
    coelution with the stable isotope labeled compound,  or relative
    retention time correlation (Appendix F).
                            10
                               488

-------
       0.  Quality Assurance Problems  and  Corrective Action:

                Problem                         Corrective Action
           MS performance outside QA
           GC column performance
            outside QA.

           Method efficiency outside
            of QA.
           Accuracy outside of QA for
            spiked matrix.
           Precision of duplicates
            outside QA.

           Detection of analyte in
            blank for 2.3,7,8-.TCOO,
            2,3,7,8-TCOF and
            1,2,3,7,8-PCOO

           For other analytes  in
            blank

           Analyte exceeds  calibration
            standard range.
          Method  efficiency  for
             blank outside  of QA or
             blank lost
Adjust MS parameters for resolution,
rerun initial  curve and  reanalyze
sample(s).

Reanalyze standards and  samples  on
modified or alternate column.

If 2378-TCDO method efficiency  <40%,
reanalyze sample set.  If method
efficiency <40% for analytes  other
than 2378-TCOD, flag and report  data.

If more than 20« of the  analytes are
outside of QA for accuracy and  pre-
cision, reanalyze the sample  set.
Reextract and reanalyze all  samples
for which the level  of contamination,
or MLO, is < 2.5X blank level.
Record blank concentration in  comment
field of samples.

Mea-sure method efficiency.  Dilute sample
100:1 respike with each standard
solution (A and B), adjust volume
and reanalyze.

Reextract and reanalyze all  positives
in set.
Note:  Because of  the  complexity of  these analyses types, it is not expected

that all analytes  will meet  all QA criteria.  Therefore, a complete review of

the data by a chemist  is  essential.  Responsibility for the evaluation of data

will be that of  the sample preparation chemist and the mass spectrometer operator.

Review of the data, including QA, and resolution of data quality problems will

finally be the responsibility of the Principal Investigator/Program Manager.

Resolution of data questions may require reanalysis of samples to include the

addition of confirmatory  ions or analysis on different types of GC columns.

                                       11
                                       489

-------
V.   Quantification Procedures
     Quantification of analytes will be accomplished by assigning  isomer
     identification, integrating  the area of mass specific GC peaks, c.id
     calculating  an analyte concentration based upon an ion relative
     response  factor between  the  analyte and standard.
     A.   Method Efficiency:   The  method efficiency for the recovery of stable
         isotope  labeled compounds is determined by calculating the amount of
         stable isotope labeled compound in the final extract and dividing by
         the amount spiked into the sample at the start of the cleanup procedure.
         This  is  done  by determining the relative response factor between 13C12
         1,2,3,4-TCDO  and the stable isotope labeled internal  standard (Eq. 1)
         and using  the response factor to calculate the concentration of the
         internal  standard in the final  solution (Eq. 2)._ The concentration in
         the final  solution times  the final  volume equals the total amount present.

              Example calculation shown for 37C14  2,3,7,8-TCDD:

        Determine Response Factor:
                      RF 328/334   =   A328 x C334             Eq. 1
                                     A334 x C328
        Calculate Internal  Standard  Concentration:
                            C328   =       A328 x C334          Eq. 2
                                     A334 x RF328/334
        Calculate Method  Efficiency:
                      % Recovery   =   C328 found    X 100
                                     C328 spiked
               where:    RF   =   response  factor
                          A    -   peak area
                          C    =   concentration of chemical

                                     1.2
                                         400

-------
 B.  Quantification of PCDD/PCDF:  The concentration of a natural
     PCDO/PCOF is  determined  by calculating a response factor between
     PCDD/PCOF and the stable  isotope labeled PCDO/PCOF for the congener
     group (see Section V.F.).  As shown in Eq. 3 in the example below,
     and  using the response  factor to calculate the level  of natural
     PCOO/PCOF in  the  sample;  (Eq. 4).

     Example  calculation  shown for 2,3,7,8-TCDO:
          Standard
                      RF322/334  =   A322 x C334
         Sample
                          V322
                                    A334 x C322
                                    A322 x S334
                                                      Eq.  3
Eq. 4
C.
                                 A334 .x RF322/334
        where:    RF  =  response factor
                  A   =  peak area
                  C   =  concentration in standard
                  S   =  spiking level in sample
                  V   =  level of analyte in sample

Signal Quality
1.  Minimum Level of Detection (HID):  Minimum Level  of Detection is
    defined as the concentration predicted from the ratio of baseline
    noise area to labeled standard area, plus three times the standard
    error of the estimate derived from the initial .calibration curve
    for the analyte of interest.
         Initial Calibration Based Method of MLO:  MID  will  be
    estimated from the ratio of the noise area to the isotopically
    labeled internal  standard area, plus three times  the standard
                             13
                              491

-------
 error of the estimate (SE)  for  the  area  ratio,  or  Y-axis, of the

 initial  calibration curve.   The Y-intercept  (INT)  is  subtracted

 from this  quantity, in  keeping  with  the  normal  formal is.-i  fcr

 "inverse prediction"  of a point on  the X, or concentrafiuii ratio

 axis, from a point  on the Y, or signal ratio axis.  The SE term

 is  derived from  an  analysis  of  variance  (ANOVA)  performed during

 the weighted least  squares  fit  of the initial calibration curve.

 This term  represents  the random error in the replicate iniecilons

 used to  generate  the  calibration curve,  the error  not accounted

 for by the linear model.  The weighting  is necessary  because of

 the relation, often observed in instrumental analysis, of increasing

 variance with increasing concentration.  MLO, according to this

 scheme,  is  defined  below, using 2,3,7,8-TCOD as  an example.
    MID  5-  [(An/A334) + (3 x SE). - INT]  x C334
                       RF322/334 x K

where:  An         =  noise area in the 2,3,7,8-TCOD window for
                      ion 322
        A334       =  labeled internal standard peak area in the
                      sample
        INT        =  the Y-axis intercept on the initial calibration
                . .   curve
        C334       =  labeled internal standard concentration
        K          =  constant to adjust for sample size and final
                      volume
        RF322/334  =  response factor for native/labeled
                      2,3,7,8-TCDD, the slope of the initial
                      calibration curve
        SE         =  standard error of the estimate of the initial
                      calibration curve
     In addition, fish tissue will be spiked with surrogate

analytes (see Internal Standard Solution B, Appendix A) prior to

extraction.  The surrogate analytes will serve as an added check
                         14
                                 49;

-------
         to  Insure  that HLO values calculated from the initial  calibration
                                                                             I
         curve,  as  discussed above, are reasonable.

     2.   Signal  to  Noise (S/N) :  The method of determining  the  signal to

         noise ratio  is shown  below.
                                                      Analyt*  signal
                                                      Hols*  signal
                                  Analjte  Sifnal Peak Area
                             —    __    __ «—___———— — __—
                             -   ———————————
                                   Noise Signal  Peak Area

        The noise area will be calculated  by Integrating over  a  peak

        width equivalent to the. anal yte signal,  typically  about  10  seconds.

0.  Quantification Standards:  Quantification standards were prepared  by

    Wright State University.  The concentration  of 2.3,7,8-TCOO  was

    checked against a primary standard obtained  from the U.S.  National

    Bureau of Standards.  A table of the concentrations of each  isomer in

    each standard is given in Appendix H.

E.  Qualitative Standards:  ERL-0 has developed  two qualitative  analytical

    standards, one containing all 75 PCDO's and  all 138 PCDF's was

    developed from an extraction of municipal Incinerator  fly  ash

    (Appendix J) and the other containing  only the biosignificant isomers

    was developed by exposure of fish to an extract of municipal

    incinerator fly ash and processing the exposed fish  for  PCOD/PCOF

    (Appendix F).  These standards will be used  to assign  structures  for

    isomer specific analyses.

                                   15


                                   433

-------
 F.   jm'tial  and  Daily Calibration of the HRMS:  An initial calibration of

     the  instrument  will be performed as needed.  This will include making
                                        *
     three  replicate injections of each calibration standard (see Appendix

     H).  Weighted least-squares linear regression will be used to generate

     a calibration curve for each analyte.   The weighting factor will  be

     inversely proportional to the variance among the replicate injections

     of each calibration standard.  The slope of the regression line will

     be the response  factor used to quantify the analyte.  At lesst two

     calibration standards will be injected daily to insure that any

     response factors used for quantification and recovery calculations do

     not deviate from the initial  calibration by more than 20J.  If the

     daily calibration generates values outside this margin, and less

     drastic corrective action does not solve the problem, a new set of

     initial calibration curves will  be generated and the old response

     factor libraries discarded.  An  example of a typical calibration

    curve, using 2,3,7,8-TCDO as  an  example, Is shown in Figure 1.

G.   Integration  of Automated  Data Processing and Quality Assurance:

    QA parameters for method  efficiency,  ion ratios, retention time

    correlations, signal/noise"ratio,  accuracy and  precision will  be

    monitored with the aid of software either developed in-house,  or

    modified  from existing programs  included with the HRMS data system.

    Raw data  will  be sorted and edited using the mass spectrometer's

    dedicated  data system, transferred to  the DEC-VAX system and processed

    using  software programs RFACTOR and DFQUANT (Figure 2.).  Data will

    be reviewed  by the  Project Director before entering into the NDS

    data  base.
                                 16
                                       494

-------
                                   Appendix A.

                      Internal  Standard  Solution A (100 ul)
                             Quantification Standards
 Compound
 37C14 2.3,7,8-TCDD

 13C12 2,3,7,8-TCOO

 i:C12 2,3,7,8-TCOF

 I3C1Z 1,2.3,7,8-PCDO

 13C12 1,2,3,7,8-PCOF

 13C12 1,2,3,4,7.8-HxCDO

 13C12 1,2,3,4,7,8-HxCDF

 13C12 1,2,3,4,6.7,8-HpCOO

 13C12 1,2,3,4.6,7,8-HpCDF

 13*C12 OCDO

 37C14 2,3,7,8-TCDF
           Concentration in Solution
                  pg/ul

                    2.0

                    5.0

                    5.0

                    5.0

                    5.0

                   12.5

                   12.5

                   12.5

                   12.5

                   25.0

                    2.0
Concentration in  Tissue*
         ppt

         10.0

         25.0

         25.0

         25.0

         25.0

         62.5

         62.5

         62.5

         62.5

        125.0

         10.0
            Surrogate Analytes Internal Standard Solution  B  (100  ul)
1,2,3,4-TCDO

1,2,4,7,8-PeCDO

1,2.3.4-TCOF

1,2,3,6,7-PeCOF
                    1.0

                    1.0

                    1.0

                    1.0


Recovery Internal  Standard  Solution C  (20 ul)
          5.0

          5.0

          5.0

          5.0
13C12 1,2,3,4-TCOO
                   50.0
         50.0
* Assuming analysis on a 20 g aliquot of tissue.
                                       17
                                        495

-------
                       Figure 1.
                    2,3,7,8-TCDD
                WEIGHTED CALIBRATION CURVE
 UJ
 ec
UJ
               CONCENTRATION/
          2          3
           CONCENTRATION
SLOPE = RESPONSE FACTOR

-------
                       Figure 2.
           DATA REDUCTION FOR PCDO/PCDF
               NATIONAL  DIOXIN STUDY
   DAILY
   CALIBRATION
   STANDARDS
      RFACTOR
     SOFTWARE
YES
                        INITIAL
                      CALIBRATION
                       LIBRARIES
                       DEC-VAX
                          OR
                        IBM-PC
                         MASS
                     SPECTROMETER
                         DATA
                        SYSTEM
                         BEGIN")
                         *^^^^^^^^^^
                 (BEGIN
'DATA
PASSES
 QA?
NO
CORRECTIVE
  ACTION
                                 SAMPLES
                                   DFOUANT
                                   SOFTWARE
                                              YES
DATA BASE
i
i


GENERATE
FINAL REPORT

                         STOP
                      49V

-------
                                Appendix  B.

                       HRGC/HRMS Operating Parameters

      Data Acquisition:   Multiple Ion  Selection  of  the  Following  Ions

       Compounds          Mass  Window             m/z Value
TCDF
37C14-TCDF
13C12-TCOF
TCDO
37C14-TCDO
13C12-TCDO
PCDF
13C12-PCOF
PC DO
13C12-PCDD
HxCDF
13C12-HxCDF
HxCDD
13C12-HxCDD
HpCDF
13C12-HpCDF
HpCDO
13C12-HpCdd
OCOF
13C12-OCDF
OCDO
13C12-OCDD
1
1
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
5
5
5
5
303.9016,
311.8898
317.9389
319.8965,
327.8847
333.9338
339.8597,
351.9000
353.8576,
367.8949
373.8207,
385.8610
389.8156,
401.8559
407.7817,
419.8220
423.7766,
435.8169
443.7398,
455.7801
457.7377,
471.7750
305.8986


321.8936


341.8567

355.8546

375.8178

391.8127

409.7788

425.7737

445.7369

459.7348

Note:  Nominal masses will be used for low resolution MS. .

Sample Introduction:  Capillary Column directly inserted into the ionizer.
lonization:  Electron Impact, 70ev, 1mA emission current
Source Pressure:  1 x 10~5 torr.
Ionizer Temperature:  250°C
Mass Resolution:  5000, 10X valley
Scan Rate:  1 MIS cycles per second
GC Column:  30 n DB-5, 60 m SP2330
Linear Velocity:  35 cm/sec Helium
Temperature Program:  180°C (2 min.), 3°/min. to 240°.  Z0°/min. to
                      280° (12 min.)

Mass windows are monitored sequentially during the temperature programs with
the windows defined by elution of standards.
                                    18

-------
 Comnound
                                   Appendix C.

                 Native PCDO and  PCOF  spiking solutions (100 ill)
  Concentration
(pg/ul  Tridecane)
 2,3,7,8-TCDD

 2,3,7,3-TCDF

 1,2,3,7,8-PCOD

 1,2,3,7,8-PCDF

 2,3,4,7,8-PCOF

 1,2.3.4.7.8-HxCOO

 1,2,3,6,7,8-HxCOO

 1,2,3,7,8,9-HxCDO

 1,2,3,4,7,8-HxCDF

 1,2,3,6,7,8-HxCOF

 2,3,4.6,7,8-HxCOF

 1,2,3,7,8,9-HxCOF

 1,2,3,4,6.7,8-HpCOO

 1,2,3,4,6,7,8-HpCDF

 1,2,3.4,7,8,9-HpCDF

OCOO

OCOF
c
0.5
0.5
0.5
0.5
0.5
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
2.5
2.5
Solutions
0
1.0
1.0
1.0
1.0
1.0
2.5
2.5
2.5
2.5'
2.5
2.5
2.5
2.5
2.5
2.5
5.0
5.0
E
1.5
1.5
1.5
1.5
1.5
3.75
3.75
3,75
3.75
3.75
3.75
2.75
3.75
3.75
3.75
7.5
7.5
                                       19

-------
                                  Appendix 0.

                 NDS  PHase  II:  Bioaccumulative Pollutant Study:
                            Sample Tracking System
                        ERL-0 loc: 25
 EPISODE  0:    0000
   Sampling  Information:
     Sampling  Office:
     State & City:
     Sampling  Contact:
     Date  Sampled:   O/ O/ 0
     Site  Location:
     Latitude:  N  0  0'  0"
     Analysis  Lab:  0
     Matrix Type:  R

Analytical:          PCOO/PCOF
  Extraction Date:   7/14/86
         GC/MS ID:   MAT86824
           LAB ID:   K071486LH
           Weight:   20.00
           XLipid:    5.2
SCC |:  QR071486
Longitude:  WOO1  0"
Date Received:  O/ O/ 0
Rerun:  0
Pesticide 4 Industrial
         O/ O/ 0
         Chemicals *
Comments:'  Reference Fish 86
                               Mass Lipid on GPC:
         0.00
         0.0
         OiOOOO
            Quality Assurance (QA) Information for PCODs/PCDFs
 Congener          Minimum QA 2 Recovery
  Tetra-PCDO/PCDF          50
  Penta-PCDO               35
  Penta-PCDF               35
  Hexa-PCDD/PCOF           35
  Hepta-PCDD/PCDF          35
  Theoretical
         0.76
         0.61
         1.53
         1.23
         1.02-
lon Ratio
   Information  pertinent  to  a  related  study.
                                       20

-------
EPISODE I:   0000
            Appendix  D.  continued

NOS Phase II:   Bioaccumulative  Pollutant  Study

            SCC I:  QR071486
DATA 1-iH BIOSIGNIFICANT  POLYCHLORINATED OIBENZOOIOXINS AND FURANS:
Analvte                CAS NO.:     I/R     S/N    2REC     OL
                                                            0.0000
                                                            0.9726
                                                            0.4863

                                                            0.0000

                                                            1.0892
                                                            1.6357
                                                            2.1784

                                                            4.0729

                                                            1.4654
                                                            0.7327
                                                            1.4654
                                                            0.7327
                                                            0.7327

                                                            1.3863
                                                            0.0000
                                                            1.3863

                                                            0.0000
                                                            0.0000

                                                            0.0000
2.3,/,8-TCOF
2,3,6,7-TCDF
3,4,6.7-TCDF
2,3,7,8-TCDD
1.2,?,7,8-PeCDF
2,3,',7,8-PeCDF
2.3,4,6,7-PeCDF
1.2,3,7,8-PeCDO
(1.2,3,4,7,8-HxCDF)
(1,2.3,4,6.7-HxCDF)
1,2,3,6,7,8-HxCOF
2,3,4,6,7,8-HxCDF .
1,2,3,7,8,9-HxCOF
1,2,3,4,7,8-HxCOO
1,2,3,6,7,8-HxCOD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
1.2,3.4,6,7,8-HpCDO
51207-31-9
1746-01-6
57117-41-6
57117-31-4
40321-76-4
70648-29-9

57653-85-7
19408-74-3
67562-39-4
55673-89-7
37871-00-4
32598-13-3


0.74
1.00
1.71
0.78
1.33
1.10
0.00
0.25
0.80
0.00
0.67
1.25
0.00
0.00
1.31
0.00
0.62
0.00
1.13
55.75
8.28
16.56
40.75
16.72
11.15
8.36
4.24
28.52
57.03
28.52
57.03
57.03
29.08
4.67
29.08
18.97
37.94
10.50
62
62
62
73
54
54
54
57
47
47
47
47
47
49
49
49
39
39
39
ERL-D Loc:    25
                                                Amount(pg/g)

                                                  5.26
                                                  NO
                                                  NO

                                                  15.63

                                                  NO
                                                  NO
                                                  NO

                                                  NO

                                                  NO
                                                  NO
                                                  NO
                                                  NO
                                                  NO

                                                  "NO
                                                  3.23
                                                  NO

                                                  NO
                                                  NO

                                                  5.93
          I/R  =  Ion  Ratio;  S/N = Signal  to  Noise;  OL
          (    )  =  Coeluting peaks on OB-5 column
                                 Detection Limit
                                         21
                                        501

-------
                        Appendix E.

           Codes  for  the SCC Number and Matix Type
 SCC  number  first letter options:
          D  -- Environmental Samples
          Q  — QA samples
 Second  letter options for Environmental  Samples:
      - Region 1
        Region
        Region
      - Region 4
      - Region 5
      - Region 6
              G - Region 7
              H - Region 8
              Y - Region 9
              J - Region 10
              T - All regional data
Second letter optoins for QA samples:

        B - Method or matrix blank
        0 - Laboratory duplicate
        R - Reference fish or fortified matrix
Matrix type:
PF - Predator Fillet
MB - Whole bottom
WP - Whole predator
R  - Reference
Y  - Blank
                         502
                            22

-------
I.
II.
                                 Appendix F.

                    GC Column Performance Quality Control

     Resolution of 1,2,3,4-TCDO from 2,3,7,8-TCDO will  be  used  to evaluate
     general column performance.  Resolution  (R)  must  be 0.75 or greater.
                                                      2d
                                                    ul
    Relative  retention  times of 13CJ2 1,2.3,6,7.8-HxCDO tci 13C12 2.3,7,8-
    ROO  (or    C,j 2,3,7,8-TCDO) wilt be used to evaluate column performance
    for  PCDO/PCOF analyses.  Relative retention time calculated for these
    standards  for an analysis set should not change by more than 51 from
    that established during the initial instrument calibration.  Relative
    retention  time of biological PCDO/PCDF are shown below.

         Relative Retention Times of Biosignificant PCOO/PCDF
    Compound

    2378-TCDF
    2367-TCOF
    3467-TCOF
    2378-TCDD
    12378-PCDF
    23478-PCDF
    23467-PCDF
    12378-PCOD
    123478-HxCOF
    123678-HxCOF
    123789-HxCDF
    234678-HxCOF
    123467-HxCDF
    123478-HxCOO
    123678-HxCDO
    123789-HxCDO
    1234678-HpCOF
    1234789-HpCDF
    1234678-HpCOO
                                  RRT DBS

                                   0.939
                                   0.973
                                   0.988
                                   1.000
                                   1.280
                                   1.359
                                   1.371
                                   1.400
                                   1.663
                                   1.676
                                   1.827
                                   1.744
                                   1.663
                                     764
                                     775
                                     802
                                   1.954
                                   2.043
                                   2.023
RRT SP2330

  1.263   •
  1.322

  1.000
  1.276
  1.775
  1.857
  1.373
  1.793
  1.817
  2.466
  2.717
  1.874
  1.989
    015
    199
    360
    174
  2.912
                                    23
                                   503

-------
                              Appendix G.



           GC Elution Window Defining Solutions  For DB-5  Column






Window             First Eluting Compound            Last  Eluting Compound



TCOO                   1,3,6,8-                        1,2,8,9-



TCDF                   1,3,6,8-                        1,2,8,9-



PeCDO              1,2,4,7,9- / 1,2,4,6,8               1,2,3,8,9-



PeCOF                  1,3,4,6,8-                       1,2,7,8,9-



HxCDO            1,2,4,6,7,9- /1,2,4,6,8,9-             1,2,3,4,6,7-



HxCDF                  1,2,3,4,6,8-                     1,2,3,4,8,9-



HpCDO                  1,2.3,4,6,7,9-                  1,2.3,4,6,7,8-



HpCOF                  1,2,3,4,6,7,8-                  1,2,3,4,7,8,9-
                                  504
                                   24

-------
en
o
•en
  ro
  en
         Calibration Standard
                        Appendix H


                   CALIBRATION STANDARDS


Concentrations in Calibration Solutions in pg/uL Tridecane


    Ul        W2        U3         U4       US         W6
                                                                                                       H7
W8
2,3,7,8-TCDD
2,3,7,8-TCOF
1,2,3,7.8-PeCDD
1,2,3.7.8-PeCDF
2,3,4.7,8-PeCDF
1,2,3,4,7,8-HxCDD
1,2.3.6,7,8-HxCDD
1,2,3,7.8.9-HxCDD
1,2,3,4.7,8-HxCDF
1.2,3.6,7.8-HxCDF
1,2,3,7,8,9-HxCDF
2.3.4.6,7.8-HxCDF
1,2,3,4,6.7.8-HpCDD
1,2.3,4,6.7.8-HpCDF
1,2,3.4.7.8,9-HpCDF
OCDD
OCDF
|Jc12-2,3,7,8-TCDD
3C o-2,3.7.8-TCDF
13CJ2-l,2,3.7,8-PeCDD
1 3 r i *5 i T o DnT nr
.,^19-1 ,2,3,7,8-PeLUr
13C o-l, 2.3.6. 7.8-HxCDD
13C o-l, 2,3,4, 7,8-HpCDF
13C ,-l. 2, 3.4.6.7 ,8-HpCDD
13c|2-l,2,3,4,6,7,8-HpCDF.
13C,o-OCDD
37Cn-2,3,7,8-TCDD
37ClJ-2,3,7.8-TCDF
13^5-1,2,3,4-1000
200
200
200
200
200
500
500
500
500
500
500
500
500
500
500
1000
1000
50
50
50
50
*t u
125
125
125
125
250
20
20
50
100
100
100
100
100
250
250
250
250
250
250
250
250
250
250
500
500
50
50
50
50
125
125
125
125
250
20
it
50
50
50
50
50
125
125
I 125
,|125
|! 125
125
125
125
125
125
250
250
50
50
50
50
125
125
125
125
250
20
20
50
25
25
25
25
25
62.5
62.5
62.5
62.5
62.5
62.5
62.5
62.5
62.5
62.5
125
125
50
50
50
50
125
125
125
125
250
20
it
10
10
10
10
10
25
25
25
25
25
25
25
25
25
25
50
50
50
50
50
50
125
125
125
125
250
20
it
5
5
5
5
5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
25
25
50
50
50
50
125
125
125
125
250
20
it
2.5
2.5
2.5
2.5
2.5
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
6.25
12.5
12.5
50
50
50
50
125
125
125
125
250
20
18
kJ \J
1
1

1

2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
5
5
50
50
50
50
125
125
125
125
250
20
it

-------
                               Appendix  I.



                       Quality  Assurance  Parameters
Ion Ratio
TCDD
PCDD
HxCDO
HpCDD
OCDO
TCOF
PCDF
HxCDF
HpCDF
OCOF
0.
0.
1.
.76+15*
.61+151
,23+152
1. 02*152
0.
0.
1.
1.
1.
1.
88+152
76+152
53+152
23+152
02+152
53+152
Method1-
Efficiency
>402,
>40%,
>402,
>402,
>401.
>40%.
>40I,
HOI.
>40J,
>40J,
<1202
<1202
<1202
<1202
<120I
<120X
<120X
<120I
<1202
<120X
Accuracy1-
at 10 pg/g
j+50%
+;50l
+1002
+1002
j+2002
^501
+;50l
+1002
^5002
j+5001
Precision^-
at 10 pg/g
+502
+501
+1002 .
+1001
+1002
+502
+501
+1002
+5002
+5001
S/N
Minimum
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
Variance of measured value from actual.



Variance of difference of duplicates from mean.
                                   26

-------
                          Appendix  J.
 Relative Retention  Times  for 4-8 Chlorine  PCDD/PCDF  Isomers
 Compound
 1368
 1379
 1369
 1378
 1469
 1247
 1248
 1246
 1249
 1268
 1478
 1279
 1234-
 1236-
 1269-
 1237-
 1238-
 2378-
 1239-
 1278
 1267-
 1289-
-TCDD
-TCOD
-TCDO
-TCOO
•TCOO
•TCDO
•TCOD
•TCOO
•TCOO
•TCOO
TCDO
TCOO
TCDO
TCOO
TCDO
TCOO
TCOO
TCOO
TCOO
TCOO
TCDO
TCDO
12468-
12479-
12469-
12368-
12478-
12379-
12369-
12467-
12489-
12347-
12346-
12378-
12367-
12389-
 •PCDO
 •PCOO
 •PCOO
 •PCOO
 PCOO
 PCDO
 PC DO
 PCOO
 PCDO
 PCOO
 PCOO
 PCOO
 PCOO
 PCOO
RRT 085

 0.814
 0.838
 0.861
 0.912
 0.912
 0.912
 0.912
 0.921
 0.921
 0.934
 0.940
 0.960
  .985
  ,985
  ,985
 0.993
 0.993
  .000
  ,009
  .028
  .048
0.
0.
0.
1.079

1.224
1.224
1.265
1.293
1.308
1.320
1.348
1.348
1.348
1.368
1.368
1.400
1.415
1.443
                                               RRT SP2330
0.826
0.871
0.948
0.916
1.072
0.948
0.948
1.014
1.014
0.972
0.990
1.027
1.014
1.027
1.105
1.014
1.014
1.000
  088
1.072
1.130
1.216
                       1
                       1.111
                       1.111
                       1.268
                       1.148
                       1.188
                       1.209
                       1.307
                       1.321
                       1.321
                       1.268
                       1.352
                       1.288
                       1.363
                       1.463
                             27
                               50'

-------
                     Appendix J.  continued
  Compound

  124679-HxCDO
  124689-HxCDO
  123468-HxCDD
  123679-HxCOD
  123689-HxCDO
  123469-HxCDD
  123478-HxCDO
  123678-HxCOD
  123467-HxCOD
  123789-HxCDO

  1234679-HpCOO
  1234678-HpCOD

  12346789-OCDO
                       RRT 085

                        1.620
                        1.620
                        1.673
                        1.700
                        1.700
                        1.700
                        1.764
                        1.775
                        1.802
                        1.802

                        1.976
                        2.023

                       2.234
                        RRT SP2330

                        1.473
                        1.473
                        1.473
                        1.546
                        1.546
                        1.681
                        1.604
                        1.618
                        1.789
                        1.721

                        2.135
                        2.297

                        3.225
'  1368
  1468
  2468
  1247
  1347
  1378-
  1346-
 2368-
 1367-
 1348-
 1379-
 1268-
 1248-
 1467-
 1478-
 1369-
 1237-
 2467-
 1234-
 2349-
 1236
 1469-
 1238-
 1278-
 1349-
 1267-
-TCDF
-TCDF
-TCDF
-TCDF
-TCDF
-TCDF
-TCDF
-TCDF
-TCDF
-TCDF
-TCDF
•TCDF
•TCDF
TCDF
•TCDF
TCDF
TCDF
TCDF
TCDF
TCDF
TCOF
TCDF
TCDF
TCDF
TCOF
TCDF
0.
0.
0.
0.
 0.730
 0.752
 0.763
  ,782
  .782
  ,782
  .782
 0.782
 0.801
 0.801
 0.801
 0.835
 0.835
 0.853
 0.853
 0.863
 0.863
 0.863
 0.380
 0.880
 0.880
 0.880
 0.880
0.902
0.920
0.920
 0.777
 0.875
 0.989
 0.885
 0.865
 0.853
 0.919
 1.071
 0.881
 0.900
 0.853
 0.943
 0.919
 0.989
 0.943
 0.943
 0.943
 1.109
 0.977
 0.977
0.989
 1.061
0.989
 1.017
 1.013
1.049
                             28
                                  508

-------
                   Appendix  J.   continued
Compound
2378-
2348-
2347-
2346'
1246-
1249
1279
2367
1239
1269
3467
1289
•TCOF
•TCDF
•TCDF
•TCOF
•TCOF
• TCDF
•TCOF
•TCOF
•TCDF
•TCDF
•TCDF
•TCDF
13468-
12468-
23479-
12368-
12473-
13467-
12467-
13478-
13479-
23469-
12479-
13469-
23468-
12469-
12347-
12346-
12348-
12378-
12367-
23489-
12379-
23478-
12489-
13489-
12369-
23467-
12349-
12389-
  PCDF
  PCDF
  PCOF
  PCDF
  PCDF
  PCDF.
  PCOF
  PCOF
  PCDF
  PCDF
  PCDF
  PCDF
  PCDF
  PCDF
  PCDF
  PCOF
  PCOF
  PCOF
  PCDF
  PCDF
  PCDF
  •PCDF
  •PCDF
  • PCOF
  •PCOF
  •PCOF
  •PCOF
  -PCDF
RRT DBS

 0.939
 0.939
 0.939
 0.939
 0.939
 0.939
 0.939
 0.973
 0.988
 0.988
 0.988
 1.071
 123468-HxCDF
 134678-HxCDF
 124678-HxCDF
 134679-HxCDF
   120
   120
   190
   202
 1.202
 1.202
 1.202
 1.202
 1.217
 1.217
 1.233
 1.253
 1.253
 1.253
 1.253
 1.253
 1.280
 1.280
 1.295
 1.309
 1.309
 1.359
 1.359
  1.359
  1.359
  1.371
  1.392
  1.446

  1.556
  1.570
  1.570
  1.570
                                                  RRT  SP2330
1.169
1.175
1.140
1.193
0.940
1.071
1.049
1.206
  140
1.162
1.264
1.341
1
1.008
1.028
1.065
1.103
1.121
1.142
1.160
1.083
1.103
1.173
1.142
1.204
1.278
1.278
1.173
1.231
1.216
 1.216
 1.252
 1.388
 1.237
 1.557
 1.446
 1.350
 1.373
 1.612
 1.420
 1.590

 1.336
 1.370
 1.348
 1.348
                              29

                            503

-------
                    Appendix J.   continued
 Compound
 124679
 124689
 123467
 123478
 123678-
 123479-
 123469-
 123679-
 123689-
 234678-
 123789-
 123489-
•HxCOF
•HxCDF
HxCDF
HxCDF
HxCDF
HxCDF
HxCDF
HxCDF
HxCDF
HxCDF
HxCDF
HxCDF
1234678-HpCDF
1234679-HpCOF
1234689-HpCDF
1234789-HpCDF

12346785-OCDF
RRT DBS

 1.602
 1.621
 1.663
 1.663
 1.676
 1.676
 1.712
 1.730
 1.744
 1.744
 1.827
 1.827
                    1.954
                    1.979
                    2.024
                    2.043
                    2.240
 RRT  SP2330

 1.428
 1.521
 1.533
 1.489
 1.502
 1.489
 1.668
 1.562
 1.668
 2.012
 1.871
 1.940

 1.936
 2.001
2.161
2.463

3.165
                            30
                              510

-------
    APPENDIX B





ITD SPECIFIC TABLES
      511

-------
TABLE  1: COMPOSITION OF THF. SAMPLE  FORTIFICATION  AND RECOVERY
                       STANDARDS SOLUTIONS
                     (concentrations  in pg/uL)
A n a 1 y t e
Sample Fortification
       Solution
Recovery Standards
      Solut i on
1 3Ci 2 -2,3,7 ,8-TCDD (IS)
' 3Ci 2 -1 , 2, 3 ,4-TCDD ( RS )
37Cl4 -2,3,7,8-TCDD (SS)
» 3Ci 2-1 ,2,3,7 ,8-PeCDD (IS)
i 3Ci 2 -1 ,2,3,6, 7.8-HxCDD (IS)
i 3Ci 2 -1,2,3,7,8,9-HxCDD ( RS )
1 3Ct 2 -1 ,2,3,4 ,6,7,8-HpCDD (IS)
l3Ci2-OCDD (IS)
i 3Ci 2-2,3,7,8-TCDF (IS)
i3Ci2-l,2,3,7,8-PeCDF (IS)
i3Cl2 -1 ,2,3,4, 7,8-HxCDF (IS)
i3Ci2-l,2,3,7,8,9-HxCDF (SS)
i 3Ci2'-l ,273,4 , 6,7,8-HpCDF ( IS)
100
-
100
100
100
-
100
200
100
100
100
100
100
-
500/100*
-
-
-
500/100*
-
-
_
-
-
-
~
IS  =  Internal  Standard; SS  =  Surrogate Standard;  RS
Standard
                              = Recovery
Note: The volume of  the  recovery  standards solution added  to  the
      final  extract before GC/MS analysis is  20  uL.

    * 500 pg/ul for sludge sample and 100 pg/ul for water samples
                              51

-------
TABLE 2:    COMPOSITION OF THE INITIAL CALIBRATION SOLUTIONS
                 USED FOR HIGH LEVEL SAMPLES

Compound
So 1 . Number
Unlabeled
Analy tes
2,3,7,8-TCDD
2,3,7, 8-TCDF
1 ,2,3,7,8-PeCDD
1 ,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4 ,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,7,8-HxCDF
1,2,3, 6,7, 8-HxCDF
. 1,2,3,7,8,9-HxCDF
" 2 , 3 , 4', 6 , 7 ", 8-HxCDF
1,2,3,4,6,7,8-HpCDD
1 ,2,3,4,6,7,8-HpCDF
1, 2,3,4,7,8,9-HpCDF
OCDD
OCDF
Internal
Standards
13Ci 2 -2,3,7,8-TCDD
»3Ci2-l ,2,3,7,8-PeCDD
13 Ci 2-1, 2, 3, 6, 7, 8-HxCDD
13 Ci 2-1, 2, 3, 4, 6, 7, 8-HpCDD
1 3Ci2 -OCDD
13Ciz -2,3,7 , 8-TCDF
13 Ci 2-1, 2, 3,7, 8-PeCDF
1 3 Ci 2 - 1 , 2 , 3 , 4 , 7., 8-HxCDF
i3 Ci 2-1, 2, 3, 4,6, 7, 8-HpCDF
Surrogate
Standards
37C1< -2,3,7,8-TCDD
13C12-1 ,2,3,7,8,9-HxCDF
Recovery
Standard
1 3Ci 2 - 1 ,2, 3 , 4-TCDD
1 3G, 2 -1 , 2.3,7,8.9-HxCDD
1

5
5
25
25
25
25
25
25
25
25
25
25
25
25
25
50
50


100
100
100
100
200
100
100
100
100


5
5


100
100
Concen Lra t. i ons (pg/uL)
2 3 -5 5

50
50
250
250
250
250
250
250
250
250
250
250
250
250
250
1000
1000


100
100
100
100
200
100
100
100
100


50
50


100
100

100
100
500
500
500
500
500
500
500
500
500
500
500
500
500
1000
1000


100
100
100
100
200
100
100
100
100


100
100


100
100

500
500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
5000
5000


100
100
100
100
200
100
100
100
100


500
500


100
i no

1000
1000
5000
5000
5000
5000
5000
5000
5000
5000
5000
5000
5000
5000
5000
10000
,10000


100
100
100
100
200
100
100
100
100


1000
1000


100
inn
                          513

-------
TAI1LE 3.  COMPOSITION OF THE  INITIAL CALIBRATION'  SOLUTIONS
                      USED TOR  LOW LEVEL SAMl'LES

Co in pound
So L . Number
Un labe Led
Analy tes
2,3,7 ,8-TCDD
2,3,7,8-TCDF
1,2,3,7 ,8-PeCDD
1 ,2., 3,7,8-PeCDF
2,3,4,7 ,8-PeCDF
1,2,3,4,7 ,8-HxCDD
1,2,3,6,7 ,8-HxCDD
1 , 2,3,7,8,9-HxCDD
1,2,3,4,7, 8-HxCDF
1 ,2,3,6, 7,8-HxCDF
1,2,3,7,8,9-HxCDF
'2,3,4', 6,7','8-HxCDF
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7 ,8-HpCDF
1,2,3,4,7 ,8,9-HpCDF
OCDD
OCDF
Internal
Standards
l3Ciz -2,3 ,7,8-TCDD
13C12 -1,2,3,7 ,8-PeCDD
13 Ct 2-1, 2,3, 6, 7, 8-HxCDD
13Ci2 -1 ,2,3,4 ,6,7,8-HpCDD
l3Ct 2 -OCDD
1 3Ct 2 -2,3,7,8-TCDF
13d 2 -1 ,2,3,7,8-PeCDF
1 3d 2 -1 ,2,3,4 ,7,8-HxCDF
I3d 2 -1,2, 3,4, 6, 7 ,8-HpCDF
Surrogate
Standards
37C14 -2,3,7,8-TCDD
1 3C, 2 -1 ,2.3,7 ,8,9-HxCDF
Recovery
S tandard
1 3 C, 2 -1 ,2 , 3 ,4-TCDD
1 3C, 2 -1 ,2 , 3. 7 ,8 ,9-HxCDD
TAHLK •) . iOr.'-AIUJNDANCK RATIO

Conce
1


0.5
0.5
2.5
2.5
2.5
2.5
2.5
2.5
2. 5
2.5
2.5
2.5
2.5
2.5
2.5
5
5


100
100
100
100
200
100
100
100
100


0.5
0.5


100
100
ACCEPT
514
2


1
1
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10


100
100
100
100
200
100
100
100
100


1
1


100
100
AHI.I-:

ntrat i
3


5
5
25
25
25
25
25
25
25
25
25
25
25
25
25
50
50


100
100
100
.100
200
100
100
100
100


5
5


100
100
RANiCI-

ons ( pg/uL)
4


50
50
250
250
250 '
250
250
250
250
250
250
250
250
250
250
500
500


100
100
100
100
200
100
100
100
100


50
50


100
100
:s FOR

5


100
100
500
500
500
500
500
500
500
500
500
500
500
500
500
1000
1000


100
100
100
100
200
100
100
100
100


100
100


100
100
I'cnn

6


500
500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
5000
5000


100
100
100
100
200
100
100
100
100


500
500


100
100
s AND


-------
TABLE

Number of
Ch
At.







a)
b)
1. o r i n e
oms
4
5
6
6a
7»
7
8
Used only
Used only
I on
Typo

M/M + 2
M+2/M+4
M+2/M+4
M/M+2
M/M+2
M+2/M+4
M+2/M+4
for i^C-HxCDF
for l3C-HpCDF
Theore
Ra 1.

0.
1 ,
1 ,
0.
0.
1 .
0.


t i ca 1
i.o

77
55
2-1
51
44
04
89


Con
trol
Lowe r

0.
1 .
1 .
0.
0.
0.
0.



65
32
05
43
37
88
76


L i in
i ts
l/'ppe r

0
1
1
0
0
1
1



.89
.78
.43
.59
.51
. 20
.02


  515

-------
TABLE 5.   ELEMENTAL COMPOSITIONS AND  EXACT MASSES OF  THE IONS
            MONITORED  I'.Y HIGH-RESOLUTION  MASS  SPECTROMETRY  FOR
                        PCDDs AND PCDFs

Descriptor Accurate
Number Mass*
1 303.9016
305.8987
315.9419
317 .9389
319.8965
321 .8936
327 .8847
331.9368
333.9339
375.8364
[354.9792]
. 2 339.8597
'" " '""' 341 .8567
351 .9000
353.8970
355 .8546
357.8516
367,8949
369.8919
409.7974
[354.9792]
3 373.8208
375.8178
383.8639
385.8610
389.8157
391.8127
401.8559
403.8529
445-7555
[430.9729]
4 407.7818
409.7789
417.8253
419.8220
423.7766
425.7737
435.8169
437.8140
479.7 165
[•130.9729]
Ion
Type
M
M + 2
M
M+2
M
M + 2
M
M
M + 2
M + 2
LOCK
M + 2
M + 4
M+2
M + 4
M + 2
M + 4
M+2
M+4
M + 2
LOCK
M+2
M+4
M
M+2
M + 2
M+4
M + 2
M + 4
M+4
LOCK
M+2
M+4
M
M + 2
M + 2
M + 4
M+2
M+4
M + 4
LOCK
Elemen ta 1
Compos i U. Lon
d 2H4 3SCl4O
C, zH,3 SC1337 CIO
1 3Ci 2H4 3SCl40
1 3Ci 2H« 3 SC1337C1O
Cl2H43SCl402
C12H43SC1337C102
C,2H437C1402
1 3Ci 2H, 3 SC14O2
1 3Ci2H4 3 5C1337C1O2
Ci2H43 5Cls37ClO
C$ Fl 3
Ci2H3 3SC1437C1O
Ci2H33SCl337Cl2O
1 3Ci2H33 SC1437C1O
l3Ci2H3 3SC1337C120
Ci2H335Cl437ClO2
Ci2H33 SC1337C12O2
13Cl2H33SCl437C102
l3Ci2H33SCl33-7Cl2O2
Ci2H33SCl637ClO
C9Fi3
Ci2H23SCls37C10
Ci2H23SCl437Cl2O
13Ci2H23SCl60
1 3Ci2H2 3SCls37ClO
Ci2H235Cls37C102
Ci2H2 3SCl4 37C12O2
1 3Cl2H23SCls37C102
1 3Ct 2H2 3 SC14 37CL2O2
C12H2 3SC1637C120
C9F, 3
Ci 2 H35C16 37C1O
Ci 2H3SCls37Cl20
1 3Ci 2H3sci7O
1 3Ci 2H3SC1637C10
Ci2H3sCl637C102
Ct 2H3 SC1S37C12O2
1 3d 2II3 SC1637C102
1 3C, 2 H3scis37Cl2O2
C, 2 H3SC17 37C120
C, F,7
Analy te
TCDF
TCDF
TCDF (S)
TCDF (S)
TCDD
TCDD
TCDD (S)
TCDD (S)
TCDD (S)
HxCDPE
PFK
PeCDF
PeCDF
PeCDF (S)
PeCDF (S)
PeCDD
PeCDD
, PeCDD (S)
PeCDD (S)
HpCDPE
PFK
HxCDF
HxCDF
HxCDF (S)
HxCDF (S)
HxCDD
HxCDD
HxCDD (S)
HxCDD (S)
OCDPE
PFK
HpCDF
HpCDF
HpCDF (S)
HpCDF (S)
HpCDD
HpCDD
HpCDD (S)
HpCDD (S)
NCDPE
PFK
                             516

-------
Tab.
   Con t. i nued
f) 4 4 1
•143
457
459
469
471
513
(442
.7428
.7399
.7377
.7348
.7779
.7750
.6775
.9728]
M+2
M+4
M + 2
M+4
M+2
M + 4
M + 4
LOCK
C,
d
d
C,
1 3 Ci
1 3 Ci
Ci
Ci
2
2
2
2
2
2
2
0
3
3
3
3
3
3
3
F
5 C 1 7
5CU
SC17
SGU
SC17
'CU
sCle
1 7
3
3
3
3
3
3
3

7
7
7
7
7
7
7

CIO
C120
C102
CljOj
C102
C1202
C120

OCDF
OCDF
OCDD
OCDD
OCDD
OCDD
DCDPE
PFK




(S)
(S)


a) The following  nuclidic masses  were  used:
H
C
13C
F
 1 .007825
12.000000
13.003355
18.9984
- 15
- 34
= 36
994915
968853
965903
S = Labeled Standard
                                i '7
                                -L (

-------
TABLE 6: GAS CHROMATOGKAPHY CON1!) IT IONS
Column type
Length (m)
i.d.  (mm)
Film Thickness (urn)
Carrier Gas
Carrier Gas Flow (mL/min)
Injection Mode
Valve Time (s)
Initial Temperature (* C)
Program Temperature
DD-5
 60
0.25
0.25
He 1ium
 1-2
splitless
30
150
150* C to 190° C ballistically
then 3° C/min up to 300° C.
                                518

-------
TABLE 7.    INITIAL AND CONTINUING  CALIBRATIONS
                     MINIMUM REQUIREMENTS
RESPONSE  rACTORS

Compound
Unlabeled
Analytes
2,3,7,8-TCDD
2,3,7,8-TCDF
1,2, 3, 7,8-PeCDD
1,2,3, 7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4 ,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4, 7, 8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
•2,3;4',6,"'7",8-HxCDF
1,2, 3,4,6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
1,2, 3,4,7,8, 9-HpCDF
OCDD
OCDF

Internal
Standards
i s Ci 2 -2,3,7,8-TCDD
13Ci2-l,2,3,7,8-PeCDD
is Ci 2-1, 2, 3, 6, 7, 8-HxCDD
i3Ci2-l,2,3,4,6,7, 8-HpCDD
l3Ci2-OCDD
i3d 2 -2,3,7,8-TCDF
l3Ci2-l,2,3,7,8-PeCDF
i3Ci2-l,2,3,4,7,8-HxCDF
i3Ci2-l,2,3,4,6,7,8-HpCDF
Surrogate
Standards
37C14 -2,3,7,8-TCDD
13 Ci 2-1, 2, 3, 7, 8, 9-HxCDF
RelaL i ve
I-Cal %RSD

25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
30



25
30
25
30
30
30
30
30
30


25
30
Response Factors
Con-Cal %Delta

25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
30
t


25
30
25
30
30
30
30
30
30


25
30
                          519

-------
  Determination of Tetra-,  Hexa-,  Hepta-,  and
  Octachlorodibenzo-p-dioxin  Isomers in  Particulate Samples  at
  ivarts per Trillion  Levels
     L. Lamparski* and T. J. Nestrick
  Analytical Laboratories. 574 Building, Dow Chemical US.A, Midland. Michigan 48640
  An analytical procedure Is presented which permits the Iso-
  mer-specific determination ol tetra-, hexa-, hepta-, and octa-
  chlorodlbenzo-p-dioxins simultaneously at parts per trillion
  concentrations. Typical data are presented to establish Its
  applicability on a variety of environmental particulate samples.
  The use of a highly specific sample clean-up procedure based
  on multiple chromatographies Is  shown to permit the Iso-
  mer-speciflc determination of 2,3,7,8-tetrachlorodibenzo-p-
  dloxin (2378-TCDD) by packed-column gas chromatogra-
  phy-tow-resolution mass spectrometry In the presence of any
  or all other TCDD Isomers.
    The determination of parts per trillion (10*12 g/g, pptr)
  concentrations of chemical residues generally requires the use
  of either highly selective  sample purification  procedures
  and/or very specific detectors (1,2). As detection limits are
  lowered, the number  of possible interferences present  at
  significant concentrations increases dramatically (3). Don-
  aldson (4) has surmized that every known organic chemical
  \uld be detected in water at a level of 1(T15 g/g or higher.
<-u>45imuarly, considering an analysis at the 10 pptr concentration
  level in a sample matrix that is 993% pure, intcrfL rt ~.ces from
  kas many as 10* compounds at concentrations 10s times higher
  Jthan the component of interest are possible. Naturally the
  addition of interferences from sources other than'the sample
  matrix can make this task formidable.  Such contamination
  of laboratory reagents by a multitude of compounds has been
  reported (5-79). Indeed, in some cases, the controlling factor
  in determining the limit of detection (LoD) for a given analysis
  is not the instrumental sensitivity of the detector but the
  apparent response observed in reagent blanks (20-22).
     This paper reports the development of an analytical pro-
  cedure which permits the isomer-specific determination of
  2378-TCDD at low parts per trillion concentrations, even in
  matrices that have been intentionally fortified with equivalent
  amounts of each of the other 21 TCDD  isomers.  Higher
  chlorinated dioxins, including hexachlorodibenzo-p-dioxins
  (HCDDs, 10 possible isomers), heptachlorodibenzo-p-dioxins
  (H7CDDs, 2 possible isomers), and octachlorodibenzo-p-dioxin
  (OCDD), can also be determined at low parts per trillion levels
  by using this technique. In regards to the isomer-specific
  determination of 2378-TCDC, the other 21 TCDD isomers
  may also be considered as possible interferences. Several
  publications have recently appeared which demonstrate CDD
  determination capabilities but do not provide complete TCDD
  isomer specificity (23-32).
               EXPERIMENTAL SECTION
  .   Reagents. The preparation of 44% concentrated sulfuric acid
  Jon silica, 10% silver nitrate on silica, basic alumina, and purified
   nitrogen (Femtogas) have been described (1).
     Silica. This adsorbent is prepared from chromatographic grade
>   silicic acid as described for the preparation of 44% sulfuric acid
   on silica (7).
     33% 1 M Sodium Hydroxide on Silica. The silica support
   is prepared as described (1).  Activated silica is weighed into an
   appropriately sized glass bottle.  On the basis  of the support
weight, the amount of 1 M aqueous sodium hydroxide necessary
to yield a reagent containing 33% by weight is added in a stepwise
fashion with shaking to produce a uniformly coated, free-flowing
powder.
  Chemicals and Solvents. All solvents used are Burdick and
Jackson, distuled-in-glass quality. Laboratory chemicals (H^SO^
AgNOa, NaOH) are ACS reagent grade. These materials are tested
by subjecting them to the analytical procedure to verify the
absence of contamination. Spectrophotometric grade Gold-label
n-hexadecane was obtained from Aldrich Chemical Co. (Mil-
waukee, WI) and was purified by passage through basic alumina.
  Expendables. Pyrex glass wool, silica boiling stones, and
disposable pipettes are cleaned before use. Glass wool and boiling
stones are Soxhlet extracted *~ 1 h consecutively with the following
solvents:  methanol, chloroform + benzene (1:1 by volume),
benzene, and methylene chloride. They are then dried in a hot
air oven at ~160 °C for ~1 h. Disposable pipettes are cleaned
ultrasonically in deionized water and then methanol and finally
methylene chloride prior to drying at ~160 °C.  Final sample
residues are stored in Reacti-Vials obtained from Pierce Chemical
Co. (Rockford, EL). The vials are cleaned by washing with de-
tergent and water and  then boiled sequentially in benzene +
chloroform + methanol (1:1:1 by volume), benzene + chloroform
(1:1 by volume), benzene, and finally methylene chloride. They
are air-dried and again rinsed with methylene chloride imme-
diately before use.
   Dioxin Standards. The primary standard of 2378-TCDD was
prepared by W. W. Muelder (Dow Chemical Co.) and its structure
was confirmed by single-crystal X-ray diffraction techniques (33).
Purity was assessed at 98% by mass spectrometry. Standards
of other TCDD isomers were synthesized and isolated as previously
described (34). Primary standards of 1,2,3,4,6,7,8-heptacnloro-
dibenzo-p-dioxin (1234678-HrCDD) and OCDD were synthesized
by H. G. Fravei and W. W. Muelder (Dow Chemical Co.). A
standard containing two HCDD isomers was prepared by Aniline
 (35). Standards of 1234679-HTCDD and the 10 HCDD isomers
were synthesized and isolated in a manner similar to that reported
for TCDDs (34).  Isotope-enriched 13C-2378-TCDD and UC-
 123478-HCDD were synthesized by A. S. Kende (University of
Rochester, Rochester, NY).  Mass spectrometric analysis indicated
 these standards to be 86 atom  % and 43 atom % "C, respectively.
 Perchlorination of the  "C-2378-TCDD provided 13C-OCDD.
   Apparatus. Reverse-Phase High-Performance Liquid Chro-
 motognphy (RP-HPLO. Residues containing chlorinated dioxins
 are injected into the RP-HPLC system:  column, two 6.2 X  250
 mm Zorbax-ODS (DuPont Instruments Division, Wilmington, DE)
 columns in series; isocratic eluent, methanol at 2.0 mL/min; pump,
 Altex Model 110A; column temperature, 50  °C; UV detector,
 Perkin-Elmer Model LC-65T liquid chromatographic column oven
 and detector operated at 0.02 aufs at 235 nm; injector, Rheodyne
 Model 7120 with 50-pL sample loop.
   Normal-Phase Adsorption High-Performance Liquid Chro-
 matography (Silica-HPLC).  Residues containing TCDDs are
 injected into the sflica-HPLC system: column, two 6.2 X 250 mm
 Zorbax-SIL (DuPont Instruments Division) columns in series;
 isocratic eluent, hexane at 2.0 mL/min; pump, Altex Model 110A;
 column temperature, ambient; UV detector. Laboratory Data
 Control Model 1204 variable-wavelength detector at 0.05 aufs at
 235 nm; injector, Rheodyne Model 7120 with 100-jiL sample
 injection loop. The columns were activated by the procedure of
 Bredeweg et aL (36).
   Packed-Column Gas Chromatography--Low-Resolution Mass
 Spectrometry (GC-LRMS).  Chlorinated dioxin quantification
                                      Reprinted from Analytical Chemistry, 1980, 52, 2045.
                 Copyright © 1980 by the American Chemical Society and reprinted by permission of the copyright owner.
                                                             520

-------
                2046 • ANALYTICAL CHEMISTRY, VOL. 52. NO.  13. NOVEMBER  1980
'
—
was accomplished by GC-LRMS using a Hewlett-Packard Model
5992-A operating in the selected ion mode (SIM) at unit resolution:
column, 2 mm Ld. x 210 cm silylaced glass; packing. 0.60% OV-L7
siiicone -t- 0.40% Poly S-179 on 80/100 mesh Pennabond Methyl
Silicone-10 cycle (HNU Systems Inc, Newton. MA); injection port
temperature. 280 °C on-column; carrier gas. helium at 14 cm1/mm:
separator, single stage glass jet operating at column temperature;
electron  energy, 70 eV.  TCDD analyses conditions:  column
temperature, 246 °C isothermal; ions monitored, native TCDDs
at mfe 320. 322. and 324, and 13C-2378-TCDD internal standard
at m/e 332.   Higher chlorinated dioxin analyses  conditions:
column temperature, programmed  from 230 to  300 "C at 10
"C/min and hold at Tnaiimirm- ions monitored, native HCDDs
at m/e 388, 390, and 392. native H,CDDs at m/e 422. 424. and
426, and native OCDD at m/e  458. 460. and 462, "C-123478-
HCDD and ^-OCDD are monitored at m/e 398 and 470, re-
spectively.
  Environmental Paniculate Samples.  Industrial Dust.
Particulates were removed from the air intake filtration system
from a research building located in Midland, ML
  Electrostatically Precipitated Fly Ash.  Particulates  were
collected from the ash-removal system associated with the elec-  .
troatatic precipitator on the Nashville Thermal Transfer Corp.
refuse incinerator located in Nashville, TN.
  Activated Municipal Sludge. Representative samples  were
removed from the center of a commercially purchased 20-kg bag
of Milwaukee Milorganite.
   Urban ParticulateJAatter. Standard Reference Material No.
1648 was obtained from the National Bureau of Standards (NBS).
  European Fly Ash. Particulate emissions from a municipal
trash incinerator were collected on filter paper by a nonisokinetic
sampling procedure.  The location of the sampling port was
downstream from the electrostatic precipitator. This incinerator
was not operated to recover energy for'power generation.
 '  Sample Preparation Prior to GC-LRMS STM quantification,
the sample is prepared by using five basic steps: (1) chlorinated
diozins removal from the matrix via hydrocarbon extraction, (2)
chemically modified adsorbent txeafinent of th* extract to remove
easily oridizable species, (3) adsorbent treatment to remove
common chemical interferences, (4) RP-HPLC residue fraction-
ation to remove residual chemically aim tin* interferences and to
separate diozins present into groups according to their degree of
chlorination, and (SXsflica-HPLC refractionation of the RP-HPLC
TCDD fractions to provide a second high-efficiency chromato-
graphic separation having different selectivity to remove residual
interferents  and to permit TCDD isomer specificity.
   An appropriately sized all-glass Soxhlet extraction apparatus
equipped with a water-cooled condenser, a 43 X 125 mm glass
thimble with coarse frit, a 250-mL boiling flask, and a tempera-
ture-controlled heating mantle is assembled. Each of the-parts
is thoroughly scrubbed with an aqueous detergent solution, rinsed
with deionized water followed by acetone, methanol, and meth-
ylene chloride, and finally air-dried. Depending on the particulate
sample  size  (larger samples require most), 5-15 g of silica is
 charged into the thimble followed by a plug of glass wool large
 enough to cover the silica bed completely. The assembled system
 (thimble installed) is charged with benzene (~250 mL) and al-
 lowed to reflux at a recycle rate of ~20 mL/min for a minimum
 period of 2 b. Following this preextraction period, the system
 is permitted to cool and the total benzene extract is discarded.
 The extraction thimble is removed and allowed to drain completely
 on a. clean wire stand in a fume hood.  The glass wool plug is
 removed with clean forceps while  a representative particulate
 sample, ranging from 50 mg for filtered airborne particulates to
 100 g for heavy soils, is quickly charged on top of the silica bed.
 The glass wool plug is replaced and the thimble returned to the
 Soxhlet extractor body. At this Hma aliquots of isooctane internal
 standard solutions containing isotopically labeled 2378-TCDD,
 123478-HCDD,  and OCDD are introduced directly  into the
 particuiatea  bed. The system is recharged with fresh benzene and
 exhaustively extracted at  the rate previously described for a
 minimum period of 16 n. Each sample or set should have at least
 one system  treated as  described for the sample to serve as a
 reagent blank.
    Upon completion of the prescribed extraction period, the flask
 mnfj»inint? t.h<» honTpnp »YT.rarr. is rpmovorl nnfi fittwi with a t.hrw-
                                                                                                      r
                                                                                                      31
                                                                                           I/
                            u
                           6 mm
Figure 1.  Liquid chromatograprtic dean-up columns.


to six-stage Snyder distillation column. The volume of the extract
solution is then reduced by atmospheric pressure distillation of
the benzene solvent to a final volume of approximately 25 mT..
The concentrated benzene extract is then diluted with a roughly
equal volume of hexane when cooL
  Bulk matrix (benzene extractables other thm^ CDDs) removal
is accomplished by passing the residue extract solution through
a Super-Macro chromatographic column (see Figure 1) prepared
as follows. The column is thoroughly washed and dried just prior
to use via the g»tn« procedure described for the Soxhlet extracted
A glass wool plug is inserted into the end of the column to se^H
as a bed support, and the following reagents are then carefiu^P
weighed directly into the column:  1.0 g of silica (bottom layer),
2.0 g of 33% 1M sodium hydroxide on silica, 1.0 g of silica, 4.0
g of 44% concentrated sulfuric acid on silica, and 2.0 g of silica
(top layer). The freshly packed column is then immediately
prewashed with 30 mL of hexane and the effluent discarded. The
residue extract is then passed through the  column followed by
3 x 5-mL hexane rinses of the boiling flask vessel Following these
rinses an additional 30 mL of hexane is  passed through the
column. The total effluent is collected in a 150-mL beaker and
then evaporated to dryness under a stream of Femtogas nitrogen.
A single drop of n-hexadecane (~25 mg) is added to the reagent
blank prior to its evaporation to dryness as a means of improving
internal standard recovery.
   Common chemical interferences are removed by passage of the
residue through a dual column system consisting of a top Macro
chromatographic column draining into a bottom High Aspect
column.   (See Figure 1.)  Each of these columns is cleaned as
previously described and a glass wool bed support inserted just
prior to use. The Macro column is packed with Ld g of 10% silver
nitrate on silica and prewashed with 15 mL of hexane prior to
use.  The High Aspect column  is packed with 5.0 g of basic
alumina.  When the top Macro column prewash has drained, it
 is positioned over the High Aspect column reservoir.  The sample
 residue is dissolved in  ~ 15 mL of hexane and introduced into
 the top column followed by 3 x 5-mL hexane beaker rinses.
 Following the rinses, an additional 30 mT^ of hexane is passed
 through the system. When drained, the top column is discarded.
 After the hexane has drained to bed  level in the High Aspect
 column, 50 mL of 50% (v/v) carbon tetrachloride in hexane is
 passed through. The total effluent to this point can be discarded
 A  25-mL glass vial (cleaned same as chromatographic columM
 is  used to collect the total effluent after 22.5 mL of 50% (v^
 methylene chloride in hexane is introduced into the  column.
 When elution is complete this fraction which contains chlorinatea
 diozins is evaporated to dryness under a stream of Femtogas
 nitrogen  (1).
    RP-HPLC fractionation of the residue is initiated by calibration
 of the appropriate collection zones for TCDDs. HCDDs. H-rCDDs.
 and  OCDD.  This is accomolished bv iniectine a calibration

-------
                                                          -—-••*-'-
ANALYTICAL CHEMISTRY. VOL. 52. NO. 13. NOVEMBER 1980 « 2047
standard containing approximately 10-20 ng each: 2378-TCDD,
HCDD(s), H-rCDD(s), and OCDD in no more than 30 u.L of
chloroform.  In accordance with the chromatogram obtained,
appropriate collection zones are established for each of these
species (see Discussion section). Following calibration, the injector
is rinsed with copious quantities of chloroform, to include multiple
consecutive injections of 50 *iL of chloroform into the column to
ensure that no residual chlorinated dioxins remain.
  The residue is prepared for  RP-HPLC fractionation by
quantitative transfer to a 0.3-mL Reacti-Vial Quantitative in-
jection requires complete residue solubility in 30 jiL or less of
chloroform. Larger injections of chloroform into this RP-HPLC
system severely reduce column efficiency. An aliquot of no more
than 30 iiL can be fractionated if the sample residue requires
greater amounts  of  chloroform to be dissolved.  Appropriate
chlorinated dioxin fractions are collected in 25-mL volumetric
flasks, equipped with ground glass stoppers, containing ~ 1 mL
of hexane. The chlorinated dioxins are recovered by addition of
2% (w/v) aqueous sodium bicarbonate.  The hexane layer is
transferred to a 5-mL gla«« vial and the aqueous phase is extracted
three additional times with — 1 mL of hexane. The combined
extracts are then evaporated to dryness under a stream of Fern-
togas nitrogen.   HCDD, H,CDD, and  OCDD fractions are
 quantitatively transferred to 0.3-mL Reacti-Vials and diluted to
 appropriate volumes for determination by GC-LRMS.
   Regarding the case for an isomer-specific 2378-TCDD deter-
 mination, additional silica-HPLC fractionation of the RP-HPLC
 2378-TCDD fraction is required (see Discussion section). Cali-
 bration of the appropriate collection zone is accomplished by
 injecting approximately 10 ng of 2378-TCDD into the silica-HPLC
 in 60-80 fiL of hexane and monitoring the chromatogram obtained.
 Adequate isomer specificity is obtained when the silica-HPLC
 columns  are sufficiently dry so as to provide a 2378-TCDD re-
 tention tm"» ranging from a minimum of 12J5 min to maximum
 of 17 min (24). Following injection of the residue fraction, the
 chromatogram is monitored and the appropriate 2378-TCDD
  fraction is collected in a 5-mL glass vial This fraction is then
  evaporated to dryness under a stream of Femtogas nitrogen and
  diluted to appropriate volume for determination by GC-LRMS.
  This procedure can also be used to collect other TCDD isomers
  aa described in the Discussion section; see Figure 2.

                     DISCUSSION
    The purpose of this paper is to demonstrate the feasibility
  of using a single multiple-step procedure to accomplish the
  isomer-specific determination of TCDDs, HCDDa, HtCDDs,
  and OCDD at low part per trillion concentrations in a variety
  of environmental particulate samples.  There were two
  prerequisites for our development of the methodology. First,
  the sample cleanup must be capable of recovering each of the
  listed chlorinated dioxin (CDD) groups from a single sample
  and from a single workup. And second, all procedures must
  use the  least sophisticated and most reliable instrumentation
  possible so that such analyses could be conducted in the
  greatest number of analytical facilities. These prerequisites
  have determined the means by which the described analyses
  <-an be  accomplished. That is, a neutral or acid extraction
  procedure must be used.  Any treatment of either the sample
  or its extracts with strong bases is known to cause degradation
  of the higher chlorinated dioxins (21,37). In accordance with
  ease of handling and the general solubility characteristics of
  higher chlorinated dioxins (least soluble species), continuous
  benzene extraction was found to be adequate for all particulate
  samples examined.  The selection of packed-column gas
  chromatography-low-resolution mass spectrometry as opposed
   to capillary column gas chromatography-high-resolution mass
   spectrometry represents our attempt to use the least so-
   phisticated instrumentation for CDD determination. Because
   packed-column GC-LRMS is inherently more subject to
   possible interference than Capillary column GC-HRMS, a more
   rigorous sample  preparation is required.  The approach of
   combining classical extraction and adsorbent clean-up tech-
   niques with consecutive RP-HPLC and silica-HPLC residue
                                                        5
                   [  a ICM i eri t sex H urr SXTP^CT : c :i
                    O IT I £O_C LASSj >A L_ "
     — iTCSO lT»CtLOn»i-
                            fMCT:CNAT:C.1 £— tMICHCR
          HP-ZJ7S TCOQs
            - [SO* 1 TCSO»
                                               ID t MATED C00«r—j
                             M7»*c7:s:iAT:c!i

nn-nmo
1478-TC3O
ap-isoil/siiii

1«7-TCDD
-i «--".' 1










JJ71--C50
RP-127a/5tL»L
UJ8-TCM
1247-TCIO

RP-I378/SXLI2
127B-TCOO
RP-2J7I/SXLI]
1244-TCOO
U24»-TCOOI
im-TCBO
11JJ-TC9D
RF-217B/SILI4
CU4«-TC3DI
1249-TCSO

136I-TC30

1234-TCCO |

1 «-"" ' 	 '


     Figure 2. Block diagram for COO sample preparation.

     fractionations can be one solution to this problem. Under
     these circumstances a significant portion of the method ca-
     pabilities to prevent MS interferences during the identification
     ajd. quantification of CDDs is relegated to the cleanup rather
     than to the final gas  chromatographic separation. This can
     be advantageous when dealing with highly contaminated
     samples because the chromatographic capacity of the clean-up
     steps is usually much greater *nn" that of the GC column,
     'especially when capillaries are used. In addition, this approach
     incorporates the consecutive RP-HPLC and silica-HPLC steps
     that we have published for the separation and isolation of the
     22 TCDD isomers (34). Their described application in this
     procedure permits the analyst to predetermine which possible
     TCDD isomers can  be present in a given residue fraction.
      Hence, the necessity of using a capfllary GC column to obtain
      improved TCDD isomer separations is eliminated. This ca-
      pability may be of utmost importance as the authors are not
      aware of any published date suggesting  that all 22 TCDD
      isomers can be separated simultaneously using a single ca-
      pillary GC column. The described methodology will address
      this problem.
         It is to be understood that this procedure has been de-
      veloped and used for survey purposes on a variety of different
      environmental particulates.  A complete method validation
      including controls,  fortifications, and replicates would be
      required for each specific matrix before  its absolute degree
      of reliability can be established.  The inclusion of isotopically
      enriched TCDD, HCDD, and OCDD internal standards pro-
      vide a reasonable degree of reliability under the circumstances
      of its described uses.
         The samples LO g of NBS urban particulate matter, 1.0 g
       of industrial dust, LO g of electrostatically precipitated fly ash
       from a municipal burner (fly ash), 16.7 g of Milorganite, and
       0.3968 g  of European flyash were Soxhlet  extracted with
       benzene  for —16 h and the resulting  residues processed
       through the preliminary liquid chromatographic clean-up
       steps. Each sample, to include a reagent blank, was fortified
       with 5-20 ng of isotopically enriched internal standard CDDs
       (^ enrichment) prior to analysis. After transfer to a 0.3-mL
       Reacti-Vial and evaporation of the solvent, ail samples yielded
       a visible white residue. Each of these was then quantitaritvelv

-------
                 2048 « ANALYTICAL CHEMISTRY, VOL. 52. NO. 13. NOVEMBER 1980
                  Table I. TCDD Isomer RP-HPLC Fractionation Scheme
                  and Specific Retention Indices
                                                                          Table II.  HCDDs, H,CDDs, and OCDD Retention Indices
'••'
">. --.:?!
 --. .i*
 t' "-s
  »
 -r-
$$'.
•iSife;:
  to?
   •fc~"
   W
                 TCDD isomer


                  1269
                  1469
                  1267/1239

                  1268/1279

                  1369/1478



                  1246/1249

                  2378
                  1236/1239

                  1278
                  1237/1238

                  1247/1248
                    1378
                    1379
                    1368
                    1234
RP-HPLC
abs RT,°
min
RP-Iso No.
11.6-13.0
11.6-13.0
12.2-12.9
12.2-12.9
13.3-13.9
13.3-13.9
13.3-13.9
13.3-13.9
RP-2378
13.7-14.5
13.7-14.5
13.8-14.5
13.8-14.4
14.4-15.2
14.0-14.7
14.0-15.0
14.0-15.0
14.2-15.1
14.2-15.1
silica-
HP LC C
rel RT6
1 Fraction
1.702
1.497
1.623
1.795
1.238
1.291
1.220
1.340
Fraction
1.323
1.411
1.000
1.356
1.350
1.288
1.100
1.128
1.154
1.199
                                 RP-Iso No. 2 Fraction
                                14.9-15.7   1.000
                                14.9-15.9   0.940
                                1*9-16.8   0.977
                                15.8-16.8   1.248
                                                     GC packed column
                                                         rel RTe
0.998
0.912
1.081
1.200
0.956
1.065
0.802
0.907
0.896
0.898
1.006d
1.037
0.969
0.893
0.979
0.990
0.854
0.857


0.858
0.771
0.729
0.960
                    a RP-HPLC abs RT = absolute retention time (±0.1 min)
                  to collect peak. * Silica-HPLC rei RT = retention time
                  relative to 2378-TCDD (±0.010).  c GC-packed column
                  rel RT = retention time relative to "C:2378-TCDD
                  (*0.005). d Native 2378-TCDD elutes slightly later than
                  "C-2378-TCDD.
                 subjected to reverse-phase high-performance liquid chroma-
                 tography fractionation. The resultant liquid chromatograms
                 monitored by a UV detector at 235 nm ^X^ for TCDDs)
                 and 0.02 aufs are shown in Figure 3b-f. Shown in Figure 3a
                 is the cfaromatogranr obtained for a CDD calibration standard
                 by RP-HPLC.  Although the appropriate CDD collection
                 zones, denoted by dotted lines, were initially established by
                 individual injections of 22 TCDD isomers, 10 HCDD isomers,
                 2 HvCDD isomers, and OCDD, we routinely compute their
                 location from the observed retention times of only a few se-
                 lected species. The specific RP-HPLC retention indices for
                 TCDDs are given in Table I and those for HCDDs, HrCDDs,
                 and OCDD are listed in Table U.
                   As indicated, all 22 TCDD isomers can be fractionated from
                 a sample residue by collecting the column effluent beginning
                 at — 1L5 and ending at ~ 17.0 min. The initial stage of TCDD
                 iaomer specificity is achieved by collecting the 22 isomers in
                 three separate fractions as shown. TCDD Iso No. 1 (RP-
                 HPLC TCDD iaomer fraction no. 1) can contain the following
                 isomers: 1269-, 1469-, 1267-, 1289-, 1268-, 1279-, 1369-, and
                 1478-TCDD. The TCDD 2378 fraction contains 1246-, 1249-,
                 2378-, 1236-, 1239-, 1278-, 1237-, 1238-, 1247-, and 1248-TCDD.
                 TCDD Iso No. 2 contains the remaining four isomers: 1378-,
                 1379-, 1368-, and 1234-TCDD. Preliminary evidence, gained
                 by fortifying samples with roughly equal amounts of all 22
                 TCDD isomers at approximately the 150 pptr concentration
                 level, has indicated that three of the possible isomers in TCDD
                 Iso No. 1 must be sacrificed in order to ensure quantitative
                 collection of 2378-TCDD in  the following fraction. This
                 consequence will be discussed later. Its occurrence is related
                 to the RP-HPLC retention times for the isomers: 1369-TCDD,
                  1478-TCDD, and one of the pair 1268- or 1279-TCDD having
                 Sil rel RT 1.238 (normal-phase silica HPLC retention time
        CDD isomer

 HCDDs
  123469-HCDD
  123467-HCDD
  124679/124689-HCDD
  124679/124689-HCDD
  123678/123789-HCDD
  123679/123 639-HCDD
  123679/123689-HCDD
  123678/123789-HCDD
  123478-HCDD
  123468-HCDD
 H7CDDs
  1234679-H,CDD
  1234678-H,CDD
 OCDD
                                                                                                             RP-
                                                                                                     silica-   HPLC GC-packe;
                                                                                                     HPLC    abs    column
                                                                                                    rel RT°   RT6   rel RTC
                                                                  i
1.081
1.192
0.958
0.972
1.060
0.970
1.039
0.974
0.941
0.890
19.23
19.47
19.62
19.70
20.07
20.20
20.23
20.85
21.02
21.37
0.954
1.077
0.805
0.306
1.103
0.903
0.908
1.016
1.006d
0.861
                                                                                                              24.00
                                                                                                              24.65
                                                                                                              29.40
  " Siiica-HPLC rel RT= retention time relative to 2378-
TCDD (±0.010).  6 RP-HPLC abs RT = absolute retention
(±0.1 min) at peak maximum.  c GC packed column rel
RT = retention time relative to "C-123478-HCDD.
d Native 123478-HCDD elutes slightly later than "C-
123478-HCDD.
relative to 2378-TCDD). Their retention times are very close
to the fraction boundary separating Iso No. 1 and 2378 and
are split rather irreproducibly between these fractions. Al-
though these isomers do not necessarily interfere  with the
quantitation of the isomers expected to the present in the
TCDD 2378 fraction, then1 quantitation essentially becomes
impossible. For cases where quantitation of these
TCDDs is required, a second aliquot of sample residue '
be fractionated by RP-HPLC in such a manner so as to <
pand the Iso No. 1 fraction-to ensure their collection.
  The 10 HCDD isomers are collected in accordance with
Figure 3 and Table fl. Although isomer-specific HCDD de-
terminations are possible by using essentially the same
chromatography procedures described for TCDDs (Le., RP-
HPLC — sflica-HPLC — GC), we have not yet applied this
system to samples. Similarly, the two H7CDD isomers are
collected in a single fraction, as is OCDD.
   The RP-HPLC residue fractionation chromatograms in
Figure 3 are typical of those associated with particulate sam-
ples. The presence of higher chlorinated species, such as
HrCDDs and OCDD, can often be observed at this point in
the analysis. Although the UV detector has been adjusted
for TTm-rJTtiiim sensitivity for TCDDs, under these conditions
                   a detectable response for HCDDs, H^ODs, and OCDD is
                   obtained for approximately 5 ng. Similarly, heptachlorodi-
                   benzofurans (HfCDFs) and octachlorodibenzofuran (OCDF)
                   may also be observed in the RP-HPLC fractionation.  Because
                   of the lack of availability of authenticated chlorinated di-
                   benzofuran (CDFs) standards, we have made no attempt to
                   quantitate these species. Via collection of appropriate RP-
                   HPLC fractions, and capillary GC-EC and GC-LRMS, we
                   have established the possible presence of four H^CDF isomers
                   and OCDF in a variety of particulate samples.
                     Refractionation of the RP-HPLC TCDD  fractions via
                   normal-phase HPLC (silica-HPLC) is the final stage of the
                   sample cleanup prior  to GC-LRMS  analysis. Normally
                   monitoring of these chromatograms with a UV detector at 0-J
                   aufs and 235 nm does not produce observable peaks with t^
                   exception of the "C-2378-TCDD internal standard. For this
                   reason example chromatograms are omitted. Table  I lists the
                   individual TCDD isomera contained in each RP-HPLC TCDD
                   fraction. Included are the RP-HPLC, silica-HPLC, and GC
                   packed column retention indices for each species.  By use of

-------
                                                    ANALYTICAL CHEMISTRY. VOL. 52. NO. 13. NOVEMBER 1980 • 2049
0*1,
****
\

!




rcDOi
" ~ ~~ *""""
an '*
"°\\
\ > ,
\ :A:/i





•n
4-







~-/^-









OCOO



^N.








 O  1  •  •  t 10  II 1* I*  It JO
      3* RP-HPUC fractionation chroinutOQrarns:
European fly ash, (f) NBS urban partculates.
(a) calibration standard, (b) Industrial dust, (c) electrostatic fly ash, (d) municipal sludge, (e)
this information, appropriate fractions can be collected from
the sflica-HPLC which permit isomer-specific GC-LRMS
identification and quantitation.
  The sflica-HPLC TCDOs fractionation scheme in Table HI
is designed to provide maximum isomer-apedfic information
when using our packed-column GC-LRMS analysis, while
minimizing the total number of fractions collected. Remem-
bering that the primary goal was to provide the highest quality
analytical data for 2378-TCDD, this scheme is adequate.
Examination of the GC packed column relative retention times
(GC rel RT, TCDD retention time relative to I3C-2378-TCDD)
for all TCDDs present in the RP-2378-TCDD fraction indi-
cates that four other TCDDs have GC rel RTs within ±0.050
(~ 12 s for 4 min absolute retention time for "C-2378-TCDD)
of 2378-TCDD.  Arbitrarily defining GC rel RT ±0.050 as the
minimum GC pakced  column separation for qualitative
identification of a TCDD isomer from 2378-TCDD and then
direct GC-LRMS analysis of the RP-2378-TCDD fraction
would yield a 2378-TCDD value which could include a max-
imum of four other TCDD isomers (2378-TCDD + 4). How-
ever, examination of the silica HPLC relative retention times
(Sil rel RT, TCDD retention time relative to 2378-TCDD) for
these TCDDs indicates that 2378-TCDD is the first isomer
to elute. The next isomer to elute is 1237/1238-TCDD (Sil
rel RT 1.10); however, even at the minimum acceptable sil-
ica-HPLC retention time for 2378-TCDD which is ~1£5 min,
this isomer is separated by ~ 1.75 min. The remaining nine
TCDD isomera, other than 2373-TCDD, present in the RP-
2378-TCDD fraction can be determined as single isomera with
the exception of those in Sil Fraction No. L Although 1237-,
                  1238-, 1247-, and 1248-TCDD are essentially baseline sepa-
                  rated by sflica-HPLC, attempts to collect them in individual
                  fractions under conditions where the species cannot be ob-
                  served by a UV detector would be difficult. Hence a single
                 ^ fraction is collected for GC-LRMS analysis. As indicated by *
                 'the respective GC rel RTs, these isomers can be determined
                  as a total for 1237- and 1238-TCDD and a total for 1247- and
                  1248-TCDD.
                    Three of the TCDD isomers present in RP-Iso No.  1 are
                     ificed in order to ensure ma-rimnm recovery of 2378-TCDD
                  in the following RP-HPLC fraction. The consequence of this
                  situation is the possible presence of 1268/1279-TCDD (Sil rel
                  RT 1.238), 1369-TCDD, and 1478-TCDD in the RP-2378-
                  TCDD fraction.  Regarding their effect upon the isomer-.
                  specific determination of 2378-TCDD, it can be observed that
                  no interference occurs by virtue of both their respective sil-
                  ica-HPLC rel RTs and their GC-packed column rel RTs.
                  However, under circumstances where the 1268/1279-TCDD
                  (Sil rel RT 1.238) isomer is relatively high in concentration,
                  it could be misidentified as 1237- and 1238-TCDD present
                  in Sfl Fraction No. 1 of the RP-2378-TCDD fraction. This
                  interference results from similar GC rel RTs for these isomers
                  as indicated in Table HL  The 1369/1478-TCDD (Sil rel RT
                  1.220) will not cause any similar interference problems with
                  those  TCDDs  present in RP-2378-TCDD  fraction—Sil
                  Fraction No.  1 because of its GC rel RT of 0.802. The re-
                  maining isomer, 1369/1478-TCDD (SU rel RT  1.340), if
                  present in high concentration may interfere with 1246/
                  1249-TCDD (Sil rel RT 1.411) in RP-2378-TCDD fraction-Sil
                  Fraction No.  3.

-------
*• •:-
• -.1
i
•ST
i>
**"*
r^«-
*«"*r.
 f* 7
                 2050 • ANALYTICAL CHEMISTRY. VOL. 52. NO. 13. NOVEMBER 1980

                                            (o)
                   (mM 3
                  Figure 4. Isomer-specific 2378-TCDO GC-LRMS mass cftromatograrns: (a) caffljratton standard, (b) reagent blank, (c) industrial dust, (d) electrostatic
                  fly ash, (e) municipal sludge, (f) European fly ash.
  GC-LRMS mass chromatograms for the isomer-specific
2378-TCDD fractions of each paniculate sample analyzed are
shown in Figure 4.  Native 2378-TCDD is monitored at m/e
320,322, and 324 and "C-2378-TCDD at 332. The calibration
standard (Figure 4a shown is typical for a 2-nL injection of
a reference standard containing 100 pg/^L of native 2378- .
TCDD and 500 pg//*L of l3C-2378-TCDD.
  The GC-LRMS mass chromatograms in Figure 5 compare
the analysis of the RP-2378-TCDD fraction from electro-
statically precipitated fly ash for 2378-TCDD, before and after
silica-HPLC refractionation. As a means of ensuring homo-
geneity, a 2-g portion of sample was processed through the
cleanup including RP-HPLC fractionation. At this point the
RP-2378-TCDD fraction was divided into two equal portions,
each equivalent to 1 g of original sample. One portion was
analyzed directly by GC-LRMS as illustrated in Figure 5a.
The other portion was further fractionated by silica-HPLC,
the Sil Fraction 2378 collected, and this residue analyzed by
GC-LRMS (Figure 5b). Comparison of 2378-TCDD quanti-
tation for these residues yields 1500 pptr before silica-HPLC
refractionation, and 430 pptr after. The value obtained before
silica-HPLC refractionation must be qualified as being the
concentration of 2378-HPLC plus four possible unseparated
isomers (see Table IV).
   Isomer-specific TCDD analysis data for each of the de-
scribed particulate  samples appear in Tables IV and V.
Quantitation of TCDDs was accomplished by averaging the
observed response at m/e 320,322, and 324 for all cases except
where denoted. Instrumental calibration for all TCDD isomers
was based upon the observed responses for a primary standard
of 2378-TCDD.  The listed concentrations for 2378-TCDD
have been corrected for recovery of the "C-2378-TCDD in-
ternal standard as given in Table V.  Concentrations given
for  all  other TCDD isomers  represent absolute observed
values.  The limit of detection (LoD) for all species was defined
as 2.5 X peak-to-valley noise in a region nearby the expected
elution tune. Observed concentrations less than the LoD are
listed as not detected (ND).


                                               525
                                                                                  FS-41
                                                                                              "C • 2378 • TCOD
                                                                                           m/i* 333
                                                                                  FS-18
                                                                                              intM TCDOt
                                                                                            m/«" 324
                                                                                  FS-28
                                                                                             /e' 322
                                                                                            m/e' 320"
                                                                                                            FS-28
                                                                                                                        ;'C-2378-TCI
                                                                                                                       ml*' 332
                                                                                                                       m/t- 324
                                                                                                            FS-8
                                                                                                                       ntlt 372
                                                                                                            FS-3
                                                                                                                       m/e' 320
                                                                                 (mini 3
                                                                                                           (mini 3
                                                                               Hgure 5.  Comparative 2378-TCDD GC-LRMS mass chromati
                                                                               (or electrostatic fly ash (a) after RP-HPLC (RP-2378 fraction) (b) al
                                                                               silica-HPLC (silica-2378 fraction).

                                                                                  As a means of investigating the degree of reliability asso-
                                                                               ciated with the isomer-specific determination of 237S-TCDD
                                                                               in a sample containing equivalent concentrations of all 21 other

-------
                                                      ANALYTICAL CHEMISTRY, VOL. 52. NO. 13. NOVEMBER 1980 « 2051
Table III.  TCDD Isomer Silica-HPLC Fractionation
Scheme and Specific Retention Indices

                                              GC
                       silica-                packed
                       HPLC  Sil collection  column
     TCDD isomer     rel RT" zone rel RT°  rel RT*
             RP-Isol No. 1 Fraction TCDDs
  Sil fraction no. 1
    1268/1279-TCDD

    1369/1478-TCDD

  Sil fraction no. 2
    1269-TCDD
    1469-TCDD
    1267/12S9-TCDD


              RP-2378
  Sil fraction 2378
    2378-TCDD
  Sil fraction no. 1
    1237/1238-TCDD*

    1247/1248-TCDD*

  Sil fraction no. 2
    1278-TCDD
  Sil fraction no. 3
    1246/1249-TCDD
    1236/1239-TCDD

  Sil fraction no. 4
    1246/1249-TCDD
        1.455-1.850
        1.180-1.370
1.238C
1.291
1.220C
1.340C

1.702
1.497
1.623
1.795

Fraction TCDDs
        0.950-1.050
1.000
        1.050-1.244
1.100
1.128
1.154
1.199
     ,   1.244-1.300
1.288
        1.300-1.385
1.328
1.356
1.350
        1.385-1.450
1.411
             RP-Iso No. 2 Fraction TCDDs
  Sil fraction no. lf             0.900-1.050
    1368-TCDD        0.940
    1379-TCDD        0.977
    1378-TCDD        1.000
  Sil fraction no. 2f             1.210-1.288
    1234-TCDD        1.248
0.956
1.065
0.802
0.907

0.998
0.912
1.081
1.200
1.006d

0.979
0.990
0.854
0.857

0.893

0.896
1.037
0.969

0.898
                     0.729
                     0.771
                     0.858

                     0.960
  " Silica-HPLC rel RT = retention time relative to 2378-
TCDD (±0.010).  * GC packed column rel RT= retention
time relative to "C-237S-TCDD (*0.005).  « See text for
recovery information. d Native 2378-TCDD elutes
slightly later than "C-2378-TCDD.  * Related isomers
typically reported as a total.  f Fractions typically com-
bined prior to GC-LRMS analysis.

TCDD isomers, we intentionally fortified a second portion of
muncipal sludge with each TCDD isomer at the levels shown
in Table VI  Neither 1237- or 1238-TCDD was added due
to their natural presence at 230 pptr (see Table V).  Analysis
of the fortified sample yielded the recovery data shown in
Table VL Regarding the 2378-TCDD data, the amount found
was corrected for the recovery of the "C-2378-TCDD and also
for the 20 pptr natural 2378-TCDD previously observed in
                                             ft • ITS  oat «SO
                                             F5-MS  m/« 4M
                                              FS-131  TO. 424
                                                                             "'COOi
                                              FS - TOt  mlf 330
                                                        I     m/«"

                                                      X.
                                                               10  It 12
 Figure 6. Higher chlorinated dloxin GC-LRMS mass chromatograms
 for electrostatic fly ash.

 the sample. These data indicate that no other TCDD isomer
 interferes with the determination of 2378-TCDD. when this
 analytical procedure is used.  Recovery values given for all
 other TCDD isomers represent absolute observed values and
 were corrected for natural levels when necessary as listed in
 Table VL
   Typical temperature programmed GC-LRMS mass chro-
 matograms for the determination of higher chlorinated dioxins
 appear in Figure 6.  For the analysis of electrostatically
, precipitated fly ash the RP-HPLC HCDDs, HTCDDs, and
 OCDD fractions were combined prior to GC-LRMS exami-
 nation (see Figure 3c). As a means of overcoming problems
 associated with samples having relatively large amounts of
 native chlorinated diorins compared to the 1-20 ng of fortified
 internal standards, a complete method validation study was
 conducted for HCDDs, HTCDDs, and dCDD ranging from 50
 pptr to 10 ppm (jig/g) and from 10 pptr to 5 ppb for 2378-
 TCDD. The control particulate sample used was a sandy loam
 sofl, to which was  added ~150 fiL of Mobile 1 synthetic engine
 lubricant per 20 g, as a means of increasing the total organics
 content to better  simulate typical particulates. The following
 native CDD standards were used for sample  fortification:
 2378-TCDD, 123678-HCDD, 123679/123689-HCDD (Sil rel
 RT 1.039), 1234678-HTCDD, and OCDD. The results of this
Table IV.  Chlorinated Dioxina Observed in Environmental Particulate Samples
                                                                      parts per billion

                                   reagent
              CDDs               blank, ng

      2378-TCDD + 4 isomers"      ND (0.06)
      other TCDDs (17 isomers)     ND (0.04)
      HCDDsc (10 isomers)         ND (0.18)
      1234679-H.CDDC             ND (0.14)
      1234678-H,CDDe             ND (0.14)
      OCDD*                      ND(0.29)

  " RP-HPLC RP-2378 fraction analyzed directly by GC-LRMS and not isomer specific as described in text.  b Sample fully
 fractionated for isomer-specific results given in Table V.  c Observed values without correction run as part of validation
 work reported in Table VII.  d For "semi" isomer specific see Table VIII.  e "C-2378-TCDD recovery 78% and value listed
 has been corrected, see Table V for others, and ND = compound not detected at limit of detection in parentheses and no
 parentheses indicates detected signal > 10X limit of detection.
industrial
dust
. . .6
. . .»
ND(14)
200
220
4000
electrostatic
flyash
1.5*
. . .ft
14
11
17
30
municipal
sludge
. . .6
. . .»
2.1
14
15
180
European
flyash
. . .6
. . .&
550d
470
570
650
NBS urban
particulates
0.12(0.12)'
0.16(0.08)
2(2)
16
• 18
210
                                                       r or
                                                       U .^u

-------
                 20S2 • ANALYTICAL CHEMISTRY. VOL 52. NO. 13. NOVEMBER 1980
                  Table V.  Isotner-Specii'ic TCDD Analyses of Environmental Particulate Samples
i  I
* .1
 •
 •-11
Sit--*
                                                                                         parts per trillion
         TCDD isomer

2373-TCDD"
1269-TCDD
1469-TCDD
1267/12S9-TCDD Sil rel RT 1.623
1267/12S9-TCDD Sil rel RT 1.795
1268/1279-TCDD Sil rel RT 1.233
1268/1279-TCDD Sil rel RT 1.291
1369/1478-TCDD Sil rel RT 1.220
1369/1478-TCDD Sil rel RT 1.340
1273-TCDD
1236/1239-TCDD Sil rel RT 1.356
1236/1239-TCDD Sil rel RT 1.350
1237/123S-TCDD Sil rel RT 1.100
1237/123S-TCDD Sil rel RT 1.128
1246/1249-TCDD Sil rel RT 1.323
1246/1249-TCDD Sil rel RT 1.411
1247/1248-TCDD Sil rel RT 1.154
1247/1248-TCDD Sil rel RT 1.199
1378-TCDD
1379-TCDD
1368-TCDD
1234-TCDD
total TCDDs
"C-2378-TCDD %"recovery
 reagent
blank, pg

 ND (40)
 ND(20)
 ND (20)
 ND (20)
 ND (20)
 ND (30)
 ND(30)
                                                       ND(60)
                                                       ND (60)
                                                       ND (60)

                                                      [ ND (60)

                                                      }ND(60)

                                                      | ND (60)

                                                       ND(20)
                                                       ND (20)
                                                       ND(20)
                                                       ND (20)
                                                       ND
                                                      63%
industrial
dust
1100
ND(40)
ND (50)
ND (50)
ND(50)
_ . .6
ND(50)
electrostatic
flyash
430 (110)
190(60)
ND (50)
100 (60)
120(60)
190 (90)c
310(90)
municipal
sludge
20(2)
ND(2)
ND(2)
ND (2)
ND(2)
.. . .
3(3)d
European
flyash
2300
1000 (140)
250(140)
300 (140)
500 (140)
1000C
1500
                ND (40)
                ND(60)
                ND(60)

                240 (50)*

                ND(60)e

                140(50)

                560(110)
               1340
               2780
                180
               6340
                59%
 ND (80)
 2SO(110)
 150(110)

 720*

 730(110)e

 310(70)

1370(150)
1160(150)
1320 (150)
 370 (150)
7750
 54%
                                                                                                    ND(3)
                                                                                                    ND(3)
                                                                                                    ND(3)

                                                                                                    230e

                                                                                                    ND (3)e

                                                                                                      8(2)

                                                                                                     23(5)
                                                                                                     13(5)
                                                                                                     13(5)
                                                                                                    ND (30)
                                                                                                    310
                                                                                                    61%
 3100
 1500
  800 (400)

 8500*

 2000*
 1500

 6900

13200
 7000
16200
 2100
69800
  56%
                    0 Corrected for "C-2378-TCDD recovery and all other isomers are absolute observed.  * • • • = not recovered as described
                  in text.  c Observed but recovery questionable.  * Detected on in/e 322 only.  * Possible isomer interference as described
                  in text.


                  Table VI.  Isomer-Specific TCDD Analysis of Municipal Sludge after Fortification
                                                                                concn in pptr
                                         TCDD isomer

                               2378-TCDD
                               1269-TCDD
                               1469-TCDD
                               126-f£L289-TCDD Sil rel RT 1.623
                               1267/1289-TCDD Sil rel RT 1.795
                               1268/1279-TCDD Sil rel RT 1.238
                               1268/1279-TCDD Sil rel RT 1.291
                               1369/1478-TCDD Sil rel RT 1.220
                               1369/1478-TCDD Sil rel RT 1.340
                               1278-TCDD
                               1236/1239-TCDD Sil rel RT 1.356
                               1236/1239-TCDD Sil rel RT 1.350
                               1237/1238-TCDD Sil rel RT 1.100
                               1237/1238-TCDD Sil rel RT 1.128
                               1246/1249-TCDD Sil rel RT 1.328
                               1246/1249-TCDD Sil rel RT 1.411
                               1247/1248-TCDD Sil rel RT 1.154
                               1247/1248-TCDD Sil rel RT 1.199
                               1378-TCDD
                               1379-TCDD
                               1368-TCDD
                               1234-TCDD
                                                        added

                                                          143
                                                          150
                                                          166
                                                          150
                                                          171
                                                          137
                                                          140
                                                          143
                                                          151
                                                          160
                                                          147
                                                          146
                                                           141
                                                           151
                                                           131
                                                           163
                                                           171
                                                           171
                                                           101
                                                           143
                                      found

                                      140
                                      108
                                      122
                                      126
                                      145
                                      . . .*
                                       69
                                      104
                                      103
                                       80

                                     (180)d

                                     }220e

                                     [203*

                                      151
                                      138
                                       45
                                      122
                       % recovery

                          98°
                          72
                          73
                          84
                          85
                          49
                          65
                          70
                          55
                          75

                          69

                          88
                          81
                          45
                          85
                    0 Corrected for recovery of "C-2373-TCDD (72%) and native 2378-TCDD present given in Table V, all other isomers are
                   absolute observed.  6 • • • = not recovered as described in text.  c Total not added. High native concentration given in
                   Table V.  d Absolute amount observed in this sample. * Total.
                   study appear in Table VTJ.   These data indicate that the
                   average recoveries of HCDDs, H7CDDS, and OCDD over the
                   described concentrations range are reasonably constant and
                   are between 70 and 80%.  Because typical paniculate samples
                   contain higher chlorinated CDDs within this range, recovery
                   factors derived from the validation can be used.  Since 13C-
                                                                                                           :vj«
                                                                                                            1
                      labeled internal standards are added to all samples, whenev^
                      very low native concentrations are  observed appropri
                      correction factors can be applied. Note that recovery valu<
                      reported for TCDD have been corrected for the observed
                      l3C-2378-TCDD internal standard recoveries after RP-HPLC
                      fractionation.

-------
                                                       ANALYTICAL CHEMISTRY, VOL, 52. NO. 13. NOVEMBER 1980 • 2053
Table VII.  Chlorinated Dioxin Recovery and Precision Data for Fortified Sandy Loam Soil"

            2373-TCDD*                HCDD                      H.CDD
sample   added, found,
  no.      pptr   pptr
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

xallc
o_ allc
x prec0
a prec'
  10
  20
  20
  50
  50
  50
  50
  50
  50
  50
  50
 100
 100
5000
5000
50
  13
  23
  21
  49
  45
  51
  53
  50
  50
  47
  52
  97
 109
5350
5400
49.6
2.6
130
140
105
 98
 90
102
106
100
100
 94
104
 97
109
107
108

106
13
99.2
5.2
added.
pptr
50
100
100
250
250
250
250
250
250
250
250
500
500
1 x 10*
1 X 10*


250

found.
pptr
30
72
57
160
180
170
170
190
160
ISO
170
410
440
8.1 X 10s
9.1 X 10'


173
10.4

%
60
72
57
64
72
68
68
76
64
72
68
82
88
87
91
73
10
69
6.0
added.
pptr
50
100
100
250
250
250
250
250
250
250
250
500
500
5 x 10*
5 x 10«


250

found,
pptr
46
75
65
170
200
200
170
210
160
180
160
430
460
4.5 X 10°
4.7 X 10*


181
19.6

%
92
75
65
68
80
80
68
84
64
72
64
86
92
90
94
78
11
72
10.8
OCDD
added,
pptr
200
400
400
1000
1000
1000
1000
1000
1000
1000
1000
2000
2000
10 X 10"
10 X 10"


1000

found,
pptr
160
330
260
730
820
7SO
720
880
700
690
690
1900
2060
8.4 x 10"
9.0 x 10'


751
69.4

%
80
83
65
73
82
73
72
88
70
69
69
95
103
84
90
80
11
75
9.2
  0 Data for all species obtained by GC-LRMS analysis of appropriate RP-HPLC fractions.  2378-TCDD values corrected for
 "C-2378-TCDD internal standard recovery, other CDOs are absolute observed. fc Corrected for 1JC-2378-TCDD where    _
 average recovery was 59.8% for all samples. c x all and a all represent the mean and standard deviation of all samples.  d x
 prec and a prec represent the mean and standard deviation of samples 4-11 to determine precision of the analysis.
 Table VTIL "Semi" Isomer-Specific HCDD Analysis Data for European Flyash, Absolute Values Reported

                                                                               parts per billion

                           HCDD isomer"
             124679/124689-HCDD Sil rel HT 0.958
             124679/124689-HCDD Sil rel RT 0.972
             123468-HCDD
             123679/123689-HCDD Sil rel RT 0.970
             123679/123689-HCDD Sil rel RT 1.039
             123469-HCDD
             123478-HCDD
             123678/123789-HCDD Sil rel RT 0.974
             123678/123789-HCDD Sil rel RT 1.060
             123467-HCDD
                                                          reagent blank

                                                          }ND(0.13)6-C

                                                           ND (0.13)

                                                          \ND(0.13)C


                                                          }ND(0.13)C

                                                          }ND(0.13)C
   " HCDD Sil rel RT = retention time relative to 2378-TCDD by silica-HPLC (Table II).
 limit of detection in ppb based on flyash sample size.  c Total.
                                                                        European flyash

                                                                             82C

                                                                              9(9)

                                                                          .  260C


                                                                            110e

                                                                             85(9)c

                                                                   ND (0.13) is not detected with
   GC-LRMS analysis data for higher chlorinated CDDs ap-
 pear in Tables IV and VTU. Table VUI illustrates a format
 for HCDD determination that is "semi"-isomer specific. In
 this case, the total RP-HCDDs fraction was analyzed directly
 by packed-column GC-LRMS. However, because GC rel RTs
 have been experimentally determined (see Table II) for all
 10 individual HCDD isomers, we  can separate the HCDDs
 observed into five distinct groups. Within each group only
 a limited number of isomers are possible. These analyses are
 accomplished by using isothermal column condition (~270
 °C) so as to ma-rimi^g the separation power of the column and
 to improve relative retention time measurements.

                    CONCLUSIONS
   Although this paper demonstrates the applicability of a
 multiple-step procedure to isomer-specifically determine a
 variety of CDDs in environmental particuiate samples, we have
 also applied the technique to many other matrices successfully.
 Simple modification of the preliminary matrix extraction has
 permitted the analysis of tissues, human  milk, vegetable
 matter, chemical products, and wastes without sacrificing high
 sensitivity or isomer specificity.  This procedure, utilizing
 packed-column GC-LRMS, has provided reliable results for
 several heavily contaminated matrices where the combination
                                                   of a less sophisticated cleanup followed by both packed and
                                                   capillary column GC-HR MS has failed. Interested individuals
                                                   may request a more thorough discussion of the method de-
                                                   velopment experiments from the authors.

                                                                  ACKNOWLEDGMENT
                                                     The authors express their gratitude to O. Hutzinger for
                                                   graciously suppling the European fly ash sample and to R.
                                                   Bumb, W. Crummett, and V. Stenger for their help in re-
                                                   pairing this manuscript.

                                                                   LITERATURE CITED
                                                     (1) Lamparskl, l_ !_; Nestrick. T. J.; Stert. H. H. Anal. Cham. 1979. St.
                                                        1453.
                                                     (2) Baugftman. R W.; Mesetson. M: 5. EHP. Environ. Health Parspact.
                                                        1973, S. 27.
                                                     (3) Widmark. G.  Adv. Ctiem.  Ser. 1971. No. 104. 1.
                                                     (4) Donaldson. W. T. Environ. 3d. Tgctinol. 1977. 11. 348.
                                                     (5) Lavi. I.; Nowicki. T. W. Bull. Environ. Contain. Toxicol. 1972, 7, 193.
                                                     (6) Scfiwaitz. 0. P. J. Chromatogr. 1978.  752. 514.
                                                     (7) RouKe. 0. R: Mueiter. W. ?A Yang, R S. H. J. Assoc. Off. Anal. Cham.
                                                        1977. 60. 233.
                                                     (8) Oennay.  0. W.: Karasek. F. W. J. Chromatogr. 1978.  151. 75.
                                                     (9) ASTM Method 11104-04-T3T, Health Lab. So. 1974,  11. 218.
                                                    (10) Giam. C. S.: Chan, H. S.; Nert. G. S. Anal. Cham. 1975. 47. 2225.
                                                    (11) Singmaster. J. A.; CrosSy. 0. G. Bull. Environ. Comam.  Toxicol. 1976.
                                                         76. 291.
                                                    (12) Glam. C. S.: Wong, M. K. J. Chromatogr. 1972. 72. 283.
                                                    (13) Millar. J. M.: Kirchner. J. G. Anal. Cham. 1952.  24. 1480.
                                                         528

-------
                                            Anal. Chem. 1980. 52. 2054-2057
 (14) Stanley, W. L: Vanmer. S. H.; Gemilli, a. J. Assoc. Oil. Anal. Cham.
     1957. 40. 282.
 (15) Bowyer. C. E.: Usat W. M. ¥.; Howard. A. N.: Gresnam. G. A. BJocnam.
     J. 1963. MS. 24P.
 (16) Brown. T. 1_; Benjamin. J. /»na/. Cham. 1964. 36. 446.
 (17) Amos. ft. J. Chromatogr. 1970. •«. 343.
 (18) Bevenuo. A.: Keftoy. T. W.: Hylin. J. W. J. Chromatogr.  1971. 54. 71.
 (19) Hutzingor. O.; Safo. S.: Zhko. V. "The Chemistry o( PC3s": Chemical
     Ruober Co. Cross:  Cleveland. OH. 1974.
 (20) foots. C. F.: wajberty. 0. G. J. Chromatogr. 1977.  /32. 511.
 (21) Lamoarskf. l_ 1_: MaWe. N. H.; Shaaoff. l_ A. J. Agric. Food. Chem.
     1978. 25. 1113.
 (22) Pe«H. 0. A. Mature (London}  197S. ^54. 324.
 (23) Bceman. G. A.; Clement. R £.: KasaseK. F. W. Anal. Cham. 1979. .57,
     2343.
 (24) CSemetn. R E; Sceman. G, A..- Karasek. F. W.; Sowars. W. O.; Parsons.
     M. l_  J. Chromatogr. 1980.  189. S3.
 (25) Matsumura. P.; Wart. C. T. Project No. OWRT A-058-W1S: Wisconsin
     University:  Maaison. WI. 1978.
 (26) Buser. H. R: SossnaroX H. P. J. Chromatogr. 1974, SO. 71.
 (27)  Buser. H. R; Bossharot. H. P. J. Assoc. Off. Anal. Cham. 1978. •
     562.
 (28)  Raooe. C.; Markiund. S.: Buser. H. R.; Bossnarot. H. P. Chamospne
     1978. 3. 269.
 (29)  Busor. M. H. Anal. Cham. 1977. 43. 918.
 (30)  Buser. H. R. Chemosanere 1979. a. 251.             j
 (31)  Tha Oow Chemical Co.. Trace Chemistries o« Ftre. 1978.  '
 (32)  Blair. E. H.. Ed. Adv. Chem. Ser. 1973. No. 120.
 (33)  Boer. F. P.; Van Remoonaro. F.: Moewaf. W. W. J. ,4m. Cham  Soc
     1972. 34. 1006.
 (34)  Nestrick, T. J.; Lamparsfci. l_ l_: Start), R H. Anal. Cham. 1979. 5.
     2273.
 (35)  Aniline. O. /4oV. C/wn. Ser. 1973. Mo.  720. 126.
 (36)  Breoeweg. R. A.; flotnman. L. O.; Prettier. C. 0. Anal. Chem. 1979. 5:
     2081.
 (37)  Firestone. 0. J. Assoc. Off. Anal. Chem. 1977, SO, 354.

RECEIVED for review January 28,1980. Resubmitted May 14.
1980. Accepted July 31, 1980.
Secondary Ion Mass Spectra  of  Diquaternary Ammonium  Salts
Timothy M. Ryan, Robert J. Day, and R. Graham Cooks*
Department of Chemistry. Purdue University, West Lafayette, Indiana  47907
Molecular dications emitted by momentum transfer processes
are observed In secondary (on mass spectra (SIMS) of dl-
quatemary ammonium salts. The relationship between mo-
lecular structure and the observation of dlcations Is explored.
Large (ntercharge separations, corresponding to lessened In-
tramolecular couiombfc repulsions, are observed to correlate
with dication detection.  Fragmentation with charge separation
is facilitated by small (ntercharge distances and can preclude
observation of the dlcauon.  Electron attachment to yield the
monocatlon is an alternative to dlcatlon emission when the
structure of the dicatton facilitates reduction. This occurs, for
example, for the herbicide dlquat (M^-etnyiene-2^-ofpyrldyl
dibromide) which Is detected as Its monocatlon. Complete
spectra of dlquaternaries can be taken with nanogram size
samples.
  Secondary ion mass spectrometry (SIMS) has recently been
shown to be a sensitive method for the characterization of
organic salts (1-4). Reported here is the observation of intact
organic dications emitted from diquatemary ammonium saits
upon sputtering.  This constitutes the first observation of
multiply charged organic molecular ions in SIMS. The result
is of interest with regard to both analytical applications of
SIMS and the fundamentals of ionization during sputtering.
Specifically, some biologically important compounds, such as
the herbicides paraquat and diquat and the curare alkaloids,
have the diquatemary structure, so that SIMS may facilitate
their characterization. In addition, studies on organic dications
reflect the degree to which electron attachment occurs during
sputtering. This process yields observable charged products
for dications, but neutrals are sputtered when monocations
are reduced during ion bombardment.

             EXPERIMENTAL SECTION
  All compounds were synthesized by using standard methods
for the preparation of quaternary ammonium salts. The organic
salts were burnished onto a 1 cm2 roughened foil of either silver
or platinum prior to SIMS analysis using argon primary ions at
5 keV and 0.3-0.5 nA primary ion current. Beam diameter was
approximately 1 mm and pressures in the ultra-high-vacuum
chamber remained below 1 X 10"* torr during the course of j
experiments.
  All spectra  were taken with Riber SIMS system usir'	
quadrupole mass analyzer, Channeltron electron multiplier, and
pulse-counting electronics.
  Ihtercharge distances were measured by using Dreiding models;
charge localization on nitrogen was assumed and the ma-rimnm
distance in the unstrained molecule is reported. Intercharge
distances (r) were used to calculate coulombic repulsive energies
(T) from T (eV) =» 14.6/r (A).

            RESULTS AND DISCUSSION
   The SIMS spectrum  of  WV/-bis(dimethyl)-4,4'-tri-.
methylenedipiperidine diiodide (1) is shown in Figure L  This
spectrum provides both the molecular weight (inferred from
the highest mass doubly charged ion, 26S2'*') and structural
information on the compound.  Emission of the doubly
charged species is confirmed by the observation of the 13C
isotope peak one-half mass unit above the dication peak (m/z
134.5 in Figure 2).  Changing the counterion does not affect
the SIMS spectrum; for example, the dibromide and diiodide
of compound 1 gave identical SIMS spectra.
   Analogous  results  were obtained for AfJV'-bis(ethyl-
methyl)-4,4'-trimethylenedipiperidine diiodide (2) and for the
aromatic compounds ^V^v"'-bis(trimethyl)-4,4'-methylenedi-
aniline diiodide  (3)  and ^^V'-bis(dimethylethyi)-4,4'-
methyienedianiiine diiodide (4). The spectrum of compound
3 is shown in Figure 3; the dication, 2842* at  m/z 142 ia of
relatively low abundance, but its 13C isotope is well resolved
in high-resolution scans.
   A considerable number of diquatemary salts (5-19, Table
I) did not exhibit observable dications. Compounds 18 an5
 19, while they did not yield molecular dications, did show i
 corresponding singly charged ions in their SIMS spectral
 Compounds 5-17 may fail to exhibit dications because they
 fragment by a favorable charge separation route, M'2* — MI*
 + Mo*. This is indicated by the absence of both singly and
 doubly charged molecular ions for these samples.
                                0003-2700/80/0352-2054301.00/0  © 1980 American Chemical Society

-------
Chemosphere,  Vol.19,  Nos.1-6,  pp  27-31,  1989
Printed  in Great Britain
0045-6535/89  $3.0O  +
Pergamon Press pic
.OO
                                                                                                                       asw
             NOVEL EXTRACTION DEVICE FOR THE DETERMINATION OF CHLORINATED
                      DIBENZO-P-DIOXINS (CDDs) AND DIBENZOFURANS (CDFs)
                                  IN MATRICES CONTAINING WATER
                                   L. L, Lamparski and T. J. Nestrick

                                       The Dow Chemical Company
                             Michigan Applied Science & Technology Laboratories
                                     Analytical Sciences, 1602 Building
                                     Midland, Michigan  48674  USA
           ABSTRACT

The efficient extraction of CDDs I CDFs from paniculate matrices containing measurable amounts of water is compli-
cated by the need to remove the water from the particles before beginning the extraction process. Traditionally, this has
been done by pre-extracting the sample with a water miscible solvent such as isopropanol or by air drying for an ex-
tended period of time over various desiccants such as sulfuric acid or phosphorus pentoxide.  These techniques cau suffer
from three possible problems: potential loss ofanalytes, accidental sample contamination, or an unacceptably long
sample drying time. In response to these drawbacks, we have developed an extraction device that combines the simulta-
neous removal and measurement of water from a particulate sample with an exhaustive Soxhlet extraction of the dried
particulars.

           KEYWORDS

Chlorinated dibenzo-p-dioxin and dibenzofuran; Soxhlet-Dean-Stark extractor (SDS); wet particulate matrices; water
determination in solids; drying procedure; aromatic solvents; Soxhlet extractor; Dean-Stark receiver.

           INTRODUCTION

In the field of trace-level determination of chlorinatedjdibenzo-p-dioxins and dibenzofurans (CDDs/CDFs), it is fre-
quently necessary to measure the analytes in solid matrices of environmental, biological, or industrial origin. When
the solid matrix cannot be dissolved in an appropriate solvent, optimum extraction of the CDD/CDF analytes is often
achieved by subjecting the sample to an exhaustive Soxhlet extraction using an aromatic solvent such as benzene or
toluene (1,2). While this approach is known to be reliable for relatively dry solid matrices, the often encountered
presence of a significant quantity  of water in a sample can render typical benzene Soxhlet extraction procedures
ineffective for CDDs/CDFs removal. Because high and variable levels of water in samples can distort the apparent
analyte concentration, it is often desirable to report quantitative findings based upon the dry weight of the sample
matrix. An obvious solution to this problem is the elevated temperature drying of an aliquot of wet sample to provide a
reasonably quick determination of water content-by weight-loss measurement. This dried sample can then be sub-
jected to Soxhlet extraction to remove the CDD/CDF analytes. Unfortunately, this procedure  can yield unreliable
results due to loss of CDDs/CDFs  during the elevated temperature drying. This phenomenon  (codistillation of CDDs
with water) has been described previously for the steam-stripping of CDDs from particulates (3).

The currently accepted method for determining CDDs/CDFs in solid sample matrices containing appreciable amounts
of water (e.g., sediments) requires dividing the sample into two portions of assumed identical composition.  One portion
is then subjected to elevated temperature drying (typically  12 to 24 hours at 105°C) to provide apparent water content
by weight loss measurement The second fraction is subjected to ambient temperature desiccation (typically requiring
in excess of 24 hours for relatively small aliquots of 1 to 10 g), followed by particle size reduction and homogenization,
and finally Soxhlet extraction of the dried material for 12 to 24 hours with an aromatic solvent to remove and concen-
trate the CDDs/CDFs (4). Alternative methods of water removal include: lyophilization (5), batch extraction with
hexane/acetone (6), mixing with anhydrous sodium sulfate followed by column extraction (7), or Soxhlet extraction (8),
or Soxhlet extraction with a water miscible solvent followed by Soxhlet extraction with methylene chloride (9,10).
                                             1

                                                 27
                                                              i > u

-------
28
While these previously described procedures for determining CDDs/CDFs in wet solids can provide reliable quantita-
tive results, they may suffer from the following problems. (A) Water and CDDs/CDFs content are determined on
separate aliquots of the original solid matrix which requires the assumption of sample homogeneity. (B) Desiccation
procedures for the aliquot to be subjected to Soxhlet extraction typically require extended time periods even for rela-
tively small sample sizes.  (C)  Desiccation procedures can lead to loss of CDD/CDF analytes.  (D) Desiccation proce-
dures can lead to inadvertent sample contamination.  (E) Certain extraction solvents which are suitable for removing
water from the sample matrix may not efficiently extract CDDs/CDFs from some types of particulates.

In response to these problems, we have combined two classical analytical techniques, Soxhlet extraction (11) and Dean-
Stark azeotropic distillation (12). into a single process which effectively removes both water and CDDs/CDFs from wet
particulate samples. In doing so, this Soxhlet-Dean-Stark (SDS) extraction apparatus permits the determination of
both water and CDDs/CDFs on the same sample aliquot, it eliminates the separate time period required for sample
desiccation prior to Soxhlet extraction, it eliminates the possible loss of analytes during desiccation, and it substan-
tially reduces the potential for sample contamination during the separate desiccation step prior to extraction.

In the course of submitting a patent application for the SDS extraction device we discovered a similar apparatus for
use on biological matrices described by H. E. Wistreich in US Patent 3,170,767, dated 23FEB65. Although combining
azeotropic distillation and collection of water with solvent extraction of the sample, this unit does not employ Soxhlet
conditions of matrix extraction and is therefore not expected to be as efficient an the described SDS design for inorganic
matrices.
           EXPERIMENTAL
Apparatus

The SDS apparatus for the extraction of wet, solid, sample matrices is shown in Figure 1.  The extractor consists of
three basic parts: the Soxhlet extractor body, including extraction thimble and reflux flask; the Dean-Stark azeotropic
distillation receiver; and the water-cooled condenser. The Soxhlet extractor (Ace Glass, Vineland, NJ 08360; Cat. No.
6730-10) is designed to hold a 43 x 125 mm glass thimble with a porosity-A glass frit (Ace Glass; Cat No.  6813-22) and
is equipped with a 250-mL boiling flask. The Dean-Stark receiver was constructed in the Dow Glass Fabrication
                                              Laboratory with 55/50 standard-taper fittings as shown;  and the
                                              water receiver was made to hold -35 mL of water and, if necessary,
                                              could be graduated to directly measure water volume collected.
                                              The extractor condenser is also of standard design (Ace Glass; Cat.
                                              No. 6740-10) for use with the Soxhlet apparatus. Heat is applied
                                              to the reflux flask by a standard heating mantle and variable
                                              transformer.

                                              Extraction Procedure

                                              Typical operation of the SDS extractor in our laboratory consists of
                                              the following procedure. The individual components  of the extrac-
                                              tor are thoroughly cleaned and dried prior to use. The extraction
                                              thimble is charged with 5.0 g of 100/200-mesh silica (previously
                                              cleaned and activated as described in  reference  13) and 100 g of
                                              50/70-mesh white, quartz sand (available from Aldrieh Chemical
                                              Company, Milwaukee, WI 53233; Cat. No. 27,473-9; used as
                                              received). Care should be taken to avoid disrupting the layers of
                                              the two adsorbents. The thimble is then placed in the extractor
                                              body; 230 mL of benzene is placed into the reflux flask and 35 mL
                                              into the Dean-Stark receiver; and the apparatus is assembled as
                                              shown in Figure  1.  Power is applied to the heating mantle, and
                                              the entire system is pre-extracted for approximately four hours.
                                              After the pre-extraction period, the apparatus cooled and dis-
                                              mantled; the solvent is discarded; and the receiver, extractor
                                              body, and reflux flask are rinsed with methylene chloride and
                                              allowed to air dry. Excess benzene is removed from the extraction
                                              thimble, sand, and silica gel by means of a slight vacuum applied
                                              to the bottom of the thimble through a small teflon-covered rubber
                                              cone. At this point, the wet sample (up to -35 g) is loaded into the
                                              thimble and manually mixed into the sand layer with a clean
                                              metal spatula (taking care not to disrupt the  silica layer below the
                                              sand) until a reasonably uniform, porous mixture is obtained. This
                                              mixture is then fortified with [1SC,,]-CDD/CDF internal standards.
  Figure 1. Typical operating configuration for
  Soihlet-Dean-Stark (SDS) extractor apparatus.
                                                531

-------
                                                                                                         29
The extractor apparatus is reassembled with a fresh charge of benzene loaded into the reflux flask and Dean-Stark
receiver.  Extraction of the sample is commenced by applying power to the heating mantle. (Note: If the sample has a
large amount of water, benzene flow through the sand bed may be somewhat restricted for the first few extraction
cycles.  To prevent this, it is useful to have made a small opening in the bed for the solvent to flow through with rela-
tively little restriction. As the sample bed dries, this opening will gradually collapse allowing efficient extraction of the
analytes. It is also helpful to slow the solvent reflux rate to match the rate of percolation through the sand and silica
beds until water removal lessens the restriction to benzene flow, otherwise thimble overflow may occur.)

Water removal from the bed usually begins with the first pass of hot benzene through the bed. In this process, water
and extractable components removed from the wet solid sample are transported to the reflux flask along with the
warm benzene. Because water and benzene distill as an azeotrope (containing -4% water and 96% benzene) at a lower
temperature than pure benzene, the water that has been extracted from the sample and returned to the reflux flask,
will be transported overhead where it will condense and  separate from the benzene and be collected as a lower phase in
the Dean-Stark receiver. The relatively dry benzene drains back into the extractor body to repeat the extraction
process. Because CDDs/CDFs do not azeotropically distill with benzene, they will remain and be concentrated in the
reflux flask.  Water removal is usually complete within eight hours and for an additional eight hours, the extraction of
the dried sample is continued to assure removal of the CDD/CDF analytes.

Following extraction, the reflux flask is removed and the crude benzene extract solution is processed through the
remainder of the sample preparation procedure. The collected water can be removed from the Dean-Stark receiver and
measured volumetrically or gravimetrically to determine the dried sample weight

           RESULTS

Our primary reason  for developing the SDS extractor was to maximize the efficiency of extraction of CDDs/CDFs from
wet, solid matrices.  Therefore, initial evaluations of the  device were designed to compare results from SDS extraction
to two different, standard, Soxhlet extraction procedures using wet sediment samples collected from the Biltic Sea as
the test material. The three extraction tests can be summarized as follows:

     SI'S - Wet. SDS extraction of -25 g of wet sediment using the procedure described in the experimental section.

     Soxhlet - Wet.  Soxhlet extraction of-25 g of wet sediment The sediment was spread as a thin film on Whatman
     GF/C glass microfiber filters, fortified with internal standard solutions, and extracted for 16 hours with benzene.

     Soxhlet - Dry.  Soxhlet extraction of pre-dried sediment. An aliquot of the wet sediment (-35 g) was spread on
     the inner surface of an 800-mL beaker and placed  under a stream of purified nitrogen for -48 hours to dry. At
     this time, the dried material could easily be ground into a  fine powder which was fortified with internal stan-
     dards and Soxhlet extracted with benzene for -16 hours.
 All of the extracts were then processed through identical sample preparations using a


   Table 1. Comparison of extraction techniques for TCDD and TCDFin sediment.
Sample
Identification'
Sediment # 4


Sediment # 5


Sediment # 6


Reagent Blank

—
Benzene
Extraction
Technique
Soxhlet - Wet
Soxhlet - Dry
SDS - Wet
Soxhlet - Wet
Soxhlet - Dry
SDS - Wet
Soxhlet - Wet
Soxhlet - Dry
SDS - Wet
Soxhlet - Wet
Soxhlet - Dry
SDS - Wet
PARTS PER TRILLION
(based on dry sediment weight)
2378-TCDD
ND
4.9
5.5
ND
0.8
1.0
ND
1.6
1.6
ND
ND
ND
(0.6)
(0.3)

(0.4)
(0.4)
(0.2)
(0.5)
(0.3)
(0.3)
(3pg)
(5pg)
(3pg)
%Rec
84%
86%
85%
69%
78%
64%
66%
85%
69%
67%
79%
73%
2378-TCDF
1.0
52
59
1.9
8.9
9.6
2.6
16
16
ND
ND
ND
(0.2)


(0.3)


(0.4)


(3pg)
(4pg)
(2pg)
%Rec
82%
91%
77%
91%
71%
76%
77%
69%
71%
84%
78%
74%
       (*  See Table 2 for water content of these samples according to number.)
modification of our previously
   reported procedure (13). The
   analytes and internal stan-
   dard recoveries were meas-
   ured by capillary column gas
   chromatography-low resolu-
   tion mass spectrometry using
   a Hewlett-Packard Model
   5987-A quadrupole instru-
   ment operating in the se-
   lected-ion-monitoring mode at
   unit resolution:  column, 0.17
   mm ID x  20 m J&W DB-5
   fused silica capillary with  a
   0.40 urn film thickness (J&W
   Scientific Inc., Folsom, CA
   95630); injector, splitless  con-
   figuration operated at 280°C
   with a 1.0 minute splitless
   valve time; carrier gas,
   helium at -30 cm/second
   linear velocity; GC/MS
   interface, column coupled
   directly to the ion source with
   the interface operating at
   250°C;  ion  source, electron
                                           I
                                           '

-------
•. ?:

.1
"•:  -a
'v  I
               30
impact mode operated at 300°C and 70 eV ionization potential; ion masses monitored for TCDD = 320, 322,334;  ion
masses monitored for TCDF = 304,306,318,376. During the examination of sample residues for 2378-TCDF, the
parent ion for hexachlorodiphenyl ether (m/z 376) was also monitored in order to demonstrate no interference from
such species.

The results of our comparison of sediment extractions for 2378-TCDD and 2378-TCDF are shown in Table 1. Three
sediment samples and a reagent blank are compared for each of the three extraction procedures studied. The concen-
trations of each analyte in parts per trillion, based on the dry sediment weight (i.e., corrected for water content), are
given. An ND signifies that the analyte was not detected at the limit of detection given in parentheses. These results
have been corrected for the internal standard recoveries reported in Table 1.

A secondary criterion for the applicability of the SDS extractor was the accuracy of its sample water content measure-
ment We compared results from a series of wet sediment samples that were analyzed for water content by the two
techniques given below. Results for these experiments are presented in Table 2.

     SDS Extraction. The water content of a sample aliquot used for TCDD/TCDF determination was measured
     gravimetrically after 16 hours extraction time.

     Oven Drying. A separate aliquot of each sample was oven-dried at 10S°C for 16 hours and the weight loss
     measured gravimetrically to calculate the water content.

           DISCUSSION

As can be seen from the data in Table 1, the SDS extraction procedure yields results for TCDD and TCDF in wet
sediment samples that are comparable to those obtained using the currently accepted method of air drying followed by
Soxhlet extraction of the homogenized, dried powder. The significant difference is that the SDS procedure does not
require the -48 hour drying period prior to extraction. Both of these methods of extraction are much more effective
than a Soxhlet extraction of the wet sediment with benzene. We have used this technique (Soxhlet extraction of wet
particulate samples) successfully when the total amount of water in the matrix was less than -5 g. This amount of
water could be absorbed by the silica  gel in the extraction thimble or distributed throughout the apparatus via
azeotropic distillation; hence, the sample matrix is dried and then efficiently extracted with benzene. When a greater
amount of water is present,  the Soxhlet extraction system's capacity for redistributed water is easily exceeded and the
sample never becomes adequately dry to allow efficient extraction of CDDs/CDFs adsorbed on the particulates.  This
occurrence may not be apparent since added internal standards, which are not actually adsorbed on the total surface of
the sample, may be recovered quantitatively (as is the case in Table 1). Because we have not attempted to determine
the exact water capacity of the Soxhlet extraction apparatus and we do not usually know the water content of a solid
	   sample before beginning the extraction, we have adopted the use of the SDS
                                =   extractor for all solid samples which would otherwise be extracted by the
                                      classical Soxhlet technique.
               Table 2. Comparison of water content
               determinations on wet sediment.

                             WEIGHT % WATER
                                      The data in Table 2 show that the SDS extractor can be used to determine
                                      the water content of the actual sample aliquot being extracted for CDDs/
                                      CDFs analysis. This can be an important advantage when only a limited
                                      amount af sample is available or there is some concern about the homogene-
                                      ity of the water content of the bulk sample. Closer examination of the data
                                      in Table 2 shows that the water levels determined by the SDS procedure are
                                      consistently lower than the standard oven drying technique by -5% relative.
                                      Although we do not believe that this is a significant difference for typical
                                      samples, we suspect that it is due to small amounts of water that adhere to
                                      various parts of the apparatus and are not collected in the Dean-Stark
                                      receiver. When a greater degree of accuracy in the water determination is
                                      required, it may be possible to dislodge these water droplets from the
                                      glassware and collect them in the Dean-Stark receiver.

Operation of the SDS extraction system in our laboratory over the past year indicates it to be a convenient method of
efficiently extracting wet, solid samples for CDDs/CDFs determinations while eliminating the need for separate proce-
dures to dry and measure moisture content in a given matrix.  In addition to the convenience, we have discovered other
advantages: (A) Wet solids can be extracted directly with benzene or other aromatic solvents without external drying
procedures that may lead to either loss of analytes or contamination of the sample.  (B) Total analysis time is signifi-
cantly reduced by eliminating the usual sample drying procedure. (C) Extraction of organics entrained in the solid
sample matrix may be more efficient than with externally dried samples because the water is removed slowly by
dissolving it in warm benzene. Hence, as the water is removed it is replaced by solvent, therefore, the matrix never
becomes devoid of a solvent and does not tend to solidify into an "impenetrable* mass as it does when the original
water is permitted to evaporate directly. When combined with the dispersion of the sample as a thin film on a clean
supporting surface, this new mode of simultaneous drying and extraction should be beneficial for recovery of various
Sample
Number
1
2
3
4
5
6
Oven Drying
at 1058C
23.3
47.4
49.1
70.0
72.0
70.2
SDS
Extraction
22.2
45.0
46.6
65.9
69.5
66.6
                                                              538

-------
                                                                                                         31
  organic species from wet, solid matrices.




            REFERENCES





  1.  R. M. M. Kooke, J. W. A. Lustenhouwer, K. Olie and 0. Hutzinger, Anal- Chem. 5JL461-463 (1981).



  2.  R. E. Clement, A. G. Viau and F. W. Karasek, Can, J..  Chan. 82, 2629-2633 (1984).




  3.  D. I. Townsend, L. L. Lamparski and T. J. Nestrick, Chemospherp Ifi. 1753-1757 (1987).




  4.  G. F. VanNess, J.  G. Solch, M. L. Taylor and T. O. Tiernan, Chemospherg 2, 553-563 (1980).




  5.  H. Hagenmaier, H. Brunner, R. Haag and A. Berchtold, Chemosphere 15, 1421-1428 (1986).




 6.  L. A. ShadofF, R. A. Hummel, L. L. Lamparski and J. H. Davidson, BulL Environ. Contarp. and Toxieol. 12,478-
     485 (1977).





 7.   L. M. Smith, D. L.  Stalling and J. L. Johnson, Anal. Chem. 5ji, 1830-1842 (1984).




 8.  R. C. G. Wegman, J. Freudenthal, G. A. L. deKorte, G. S. Groenemeijer and J. Japenga, Chemosphere 15, 1107-
    1112 (1986).




 9.  J. M. Czuczwa and R. A. Kites, Environ. Sci. ledmd. 22, 195-200 (1986).




 10. I. Comoni, A. DiMuccio, D. Pontecorvo and L. Vergori, J. Chromatogr. 153. 233-238 (1978).



 11. F. Soxhlet and J. Szombathy. Dingler's Polvtech. Journal 232. 461-165 (1879).




 12. E. W. Dean and D. D. Stark, J, of Ind. and Eni* Chem. 12, 486-496 (1920).



13.  L.L. Lamparski and T.J. Nestrick, Anal. Chfim, 52,2045-2054 (1980).
                                                    tr M  *

                                                    U O 4

-------
Environmental Toxicology and Chemistry Vol. 5, pp. 355-360, 1986
Crimed in the USA. Pergamon Press Ltd.
                                                                           0730-7268/86 $3.00 + .00
           CONTROL OF INTERFERENCES IN  THE  ANALYSIS
                  OF HUMAN ADIPOSE  TISSUE  TO  2,3,7,8-
               TETRACHLORODIBENZO-/7 DIOXIN (TCDD)

           D. G. PATTERSON*,  J.  S. HOLLER, D. F. GROCE,  L.  R. ALEXANDER,
                    C. R. LAPEZA, R.  C. O'CONNOR, and J. A.  LIDDLE
            Toxicology Branch, Clinical Chemistry Division, Center for Environmental Health,
                        Centers for Disease Control,  Public Health Service,
               U.S. Department of Health  and Human Services, Atlanta, Georgia 30333

                        (Received 20 June 1985; Accepted 17 October 1985)   .

       Abstract-While developing a method  to analyze human adipose tissue for 2,3,7,8-tetrachloro-
       dibenzo-p-dioxin (TCDD) at the  1 ppt level, we observed several peaks with all of the analytical
       characteristics of TCDDs in the analyses of wipe tests and blank, quality control, and human adi-
       pose samples at concentrations equal to the low part-per-trillion level in a 10 g sample. The source
       of these contaminants was established to be a cleaning solution used to wash the floors in the lab-
       oratory performing the analysis.

       Keywords—Dioxin   2,3,7,8-TCDD    Interferences    Cleaning solution
               INTRODUCTION
   During the past year, we have been developing
a method  for the analysis of 2,3,7,8-tetrachloro-
dibenzo-p-dioxins (TCDDs) in human adipose tis-
sue down to the 1 ppt level.  In addition, we have
been synthesizing milligram amounts of numerous
dioxin and furan congeners and monitoring possi-
ble laboratory contamination of these compounds
with wipe  tests. Several peaks that have all of the
characteristics  of TCDDs have appeared in  the
analyses of these wipe tests and  human  adipose
samples over a period of time. Because of the very
low level of detection  required for our analyses
and our desire to be able to analyze  for all 22
TCDDs in biological samples, we began to system-
atically search for  the source of these contami-
nants. After establishing the structural identity of
the TCDD isomers  found in the wipe tests,  we
discovered that several of these had never been
worked with nor synthesized in our laboratory. In
addition, the peaks maintained the same relative
ratio to each other in each sample in which they
were detected.  This information suggested a sys-
tematic source  of contamination from some place
   "To whom correspondence may be addressed.
   Use of trade names is for identification only and does
not constitute endorsement by the Public Health Service
or by  the U. S.  Department of  Health and Human
Services.
outside the analytical and synthetic program. An
earlier report [1] suggested a possible link between
the illness of a janitress and her cleaning solution
which contained a mixture of bleach and 2-benzyI-
4-chIorophenol dissolved in acid. In this article, we
describe the results of an investigation of one of
the ingredients used in a cleaning  solution for
washing the floors of the building in which the
dioxin laboratory is located.
           SAMPLE PREPARATION
Enrichment procedure
   The samples (spiked with isotopic marker com:
pounds) are processed in  a two-part procedure
developed by Smith et al. [2] that we have adapted
for use in our laboratory. These authors have pro-
vided a schematic diagram of the cleanup appara-
tus in Figure 2 of their publication [2J. In part I,
the mixture is subjected  to  solvent  extraction
(methylene chloride/hexane, 50:50), and the ex-
tract is, in the same process,  passed through a
series of silica-based absorbents and then through
a carbon/glass fiber adsorbent.  The extract passes
through the absorbents in  the following order:
potassium silicate, silica gel, and finally the carbon
adsorbent.  The residues of interest (polychlori-
nated dizenzo-p-dioxins (PCDDs), polychlorinated
dibenzofurans (PCDFs),  and  non-ortho substi-
tuted polychlorinated biphenyls (PCBs), as well as
other chemical classes such as polychlorinated
                                            355

-------
 356
D. G. PATTERSON ET AL.
 naphthalenes (PCNs), polychlorinated biphenyl-
 enes, and certain pplyaromatic hydrocarbons) are
 retained on the carbon adsorbent and subsequently
 recovered by reverse elution with toluene.
   In part II of the procedure, after a change of
 solvent to hexane, the sample is applied to a sec-
 ond series of absorbents contained in the two tan-
 dem columns.  The first column contains small
 amounts of cesium silicate and sulfuric acid-im-
 pregnated silica gel. The effluent from this column
 flows directly onto the second column containing
 activated alumina on which the final fractionation
 of several classes of residues  is accomplished.
 After the sample volume is reduced, analyses are
 carried out  by  high-resolution gas chromatogra-
 phy/high-resolution mass spectrometry.
   The components of the apparatus used in part I
 of the enrichment procedure are as follows: Col-
 umn  #1  (25 mm  i.d. x 600  mm, Michel-Miller
 chromatography column 5795-48) is connected to
 column #2 (22 mm i.d. X 100 mm, Michel-Miller
 pre-column  5796-34)  and to column #3 (8  mm
 i.d. x 85  mm, Michel-Miller filter column 5813-23,
 all from  Ace Glass, Vineland, NJ), by means of
 Ace Glass #5801 Teflon end fitting adapters and
 standard one-eighth inch o.d. Teflon tubing and
 tube end fittings (available from most chromato-
 graphic supply  companies). The washing solvent
 reservoir  is also  an Ace Glass chromatography col-
 umn (25  mm i.d. x 450 mm,  5795-40).
   The solvent flow switching valves are Hamilton
 miniature inert valves (Hamilton Co., Reno, NV):
 selector valve (#86781) and bypass and reverse
 flow valves (#86779). The valving arrangement is
 designed  to permit the following operations to be
 performed: venting of the solvent line from  col-
 umn #1, venting of the solvent reservoir, bypass of
 column #2, delivery of the effluent from column
 #1 to columns #2 and #3 sequentially, delivery of
 solvent from the reservoir sequentially to columns
 #2 and #3 or to column #3 only, reversal of solvent
 flow in columns #2 and #3, and stoppage of sol-
 vent flow in all lines. The solvent reservoir is rou-
 tinely  pressurized with I to  10  p.s.i.  nitrogen
during column washings. Column #1 is packed in
 the following sequence: one or two discs cut from
glass microfiber filters (GF/F, 4.7-cm diameter,
 Whatman Inc.,  Clifton,-4MJ), a  2-cm depth of
 anhydrous sodium sulfate,  15 g of silica  gel,  15 g
of potassium silicate, another disc of glass micro-
 fiber filter (GF/D), 50 g of a I to 4 (w/w) mixture
of the sample and anhydrous sodium sulfate, and,
lastly, a.2-cm depth of anhydrous sodium sulfate.
Column #2 is packed  with equal volumes, 15 ml
           each, of potassium silicate and silica gel bracketed
           by plugs of glass wool or preferably discs of glass
           fiber filters (3 urn retention GF/D, Whatman Inc.,
           Clifton, NJ). Column #3 is packed with a mixture
           of activated carbon (Amoco PX-21, Amoco  Re-
           search  Corp., Chicago, IL) and glass fibers. The
           apparatus for part II of the enrichment procedure
           consists of two columns arranged in tandem. Col-
           umn #4 is prepared from a disposable Pasteur
           pipet and is packed first with a plug of glass wool,
           then with 2  cm (0.50 g) of sulfuric acid-impreg-
           nated silica gel, then with 3 cm (0.54 g) of cesium
           silicate, and, finally, with 0.5 cm of anhydrous
           sodium sulfate. Column #5 is constructed from a
           225-mm length of 5 mm i.d.  heavy-walled glass
           tubing  fitted with a 50-ml reservoir  and a 24/40
           ground glass joint.  Column #5 is  packed with a
           plug of glass wool, followed by 3.50 ml (3.65 g) of
           activated alumina, and then 0.5 cm of anhydrous
           sodium sulfate. The alumina is packed firmly by
           sharply tapping the supporting clamp. A flow
           chart of the  two part cleanup procedure is shown
           in Figure 1 along with the function of each step in
           the procedure.

           Materials
              All  solvents are glass-distilled grades (Burdick
           & Jackson, Muskegon, MI). Silica Gel  60, 70 to
           230 mesh (EM Reagent, MC/B, Cincinnati, OH),
           acid alumina (AG4, Bio Rad  Labs,  Richmond,
           CA), and sodium sulfate (Mallinckrodt AR) were
           used.
              All  glassware is washed with acetone, toluene,
           and, finally, with 50/50 hexane/methylene chlo-
           ride (in that order)  before use.
              Alumina,  sodium sulfate,  potassium and ce-
           sium silicates, sulfuric acid-impregnated  silica  gel,
           and carbon  on glass  fibers are all  prepared as
           described  by Smith  et al. [2].  The cleaning solu-
           tion was Vesphene  II (for descriptive  purposes
           only) manufactured by Vestal Labs a Division of
           Chemed Corp., St.  Louis, Missouri.

           Sample Preparation
              Before samples were  prepared, a clearing  run
           through the apparatus was conducted that con-
           sisted  of  100 ml of hexane/methylene chloride
           50/50 (solvent A), followed by two samples that
           were carried through the sample cleanup process
           described  in the preceding section.
              Sample I. Consisted of 100 ml of hexane/
           methylene chloride  50/50  (solvent A),  "spiked"
           with an internal standard  solution consisting of
           240 pg of l3C,2-Iabeled 2,3,7,8-TCDD. The stan-

-------
                             Analysis of human adipose for dioxins
                                                       357
    SAMPLE PREPARATION
    COLUMN EXTRACTION/
    PRIMARY CLEAN-UP
    ADSORPTION  OF
    PLANAR AROMATICS
    SECONDARY CLEAN-UP
    FRACTIONATION OF
    PLANAR AROMATICS
    ANALYSIS
     HOMOGENIZATION/DRYING
ADDITION  OF INTERNAL STANDARD
                                      POTASSIUM SILICATE
                                           SILICA  GEL
                                      POTASSIUM SILICATE
                                           SILICA  GEL
      CARBON/GLASS FIBERS
                                         PART I
                                         (AUTOMATED)
         CESIUM SILICATE
         H2SO4/SILICA GEL
              ALUMINA
                                                                     f-  PART
       CAPILLARY  GC/HRMS
                               Fig. 1. Dioxin analysis flow chart.
dard is accurately measured, with an SMI "micro/
pettor." The disposable pipet is primed by dipping
the glass capillary into the standard and operating
the plunger rapidly 10 times before the standard is
dispensed. This sample is then loaded onto column
#1, by using washing of solvent A, and carried
through the entire five-column cleanup procedure
described  in the preceding section. This sample
serves as a blank for the cleanup apparatus.
   Sample 2. Consisted of 20 g  of the cleaning
solution extracted five times with 20-ml portions
of solvent  A; the combined extracts were "spiked"
with 240 pg of 13C|2-2,3,7,8-TCDD and carried
through the cleanup procedure with the same col-
umns and same absorbents used  to process  the
blank (Sample I).

         INSTRUMENTAL ANALYSIS

   The instrument system consists of a Vg ZAB-
2F high-resolution mass spectrometer with a Vg
2250 data system and a Hewlett Packard 5840 gas
chromatograph. The analyses are conducted in an
isomer-specific mode, with a 60M SP2330 capillary
column. The chromatographic conditions are in-
jection in the splitless mode while the temperature
is maintained at 100°C for 2 min, programming to
180°C  at 20°C/min, programming to 220°C at
               3°C/min, and holding at the final temperature for
               20 min. The mass spectrometer is operated in the
               high-resolution (static RP =10000 at 10% valley)
               selected-ion recording mode, with perfluorotri-
               butylamine providing the lock mass at 313.9839 or
               254.9854. The most sensitive mode of detection and
               quantitation uses a lock mass at 313.9839 and mon-
               itors ions in the molecular  ion cluster (319.8964,
               321.8933) and the same corresponding ions in the
               internal  standard (331.9864, 333.9335). The less
               sensitive confirmation mode monitors  additional
               ions due to loss of chlorine from the molecular ion
               (284.9274, 286.9244) and loss of COC1 from the
               molecular ion (256.9324, 258.9296). In this mode
               only the most intense ion of the internal standard
               (333.9335) is monitored. In  the quantitation mode
               1-2 pg of 2,3,7,8,-TCDD can readily be observed
               at a signal-to-noise greater than 3/1. Generally, 20
               pg of 2,3,7,8-TCDD  is  the  minimum  amount
               required to yield a 3/1 signal-to-noise for the ions
               257 and 259. Analytical standards containing 1.5,
               2, 10, 20, and 50 pg/ML of 2,3,7,8-TCDD and 48
               pg/AiL of I3C12-TCDD are used to establish  rela-
               tive response factors for quantitative calculations.
               Recovery of the internal standard after sample
               cleanup is estimated on the basis of absolute ion
               counts of the internal standard  in the sample
               versus those in the standard.
                                               r O '-;

-------
358
                                      D. G. PATTERSON ET AL.
          RESULTS AND DISCUSSION
   Two peaks with all the analytical characteristics
of 1,3,6,8- and 1,3,7,9-TCDD (see experimental)
have appeared in samples of human adipose, qual-
ity control material, blanks, and wipe tests ana-
lyzed in our laboratory. The levels of these two
peaks were always in the very low part-per-trillion
in adipose, quality control, and blank samples.
These  peaks have not been  observed in runs of
analytical standards,  indicating a  source of the
peaks external to the analytical instrument system.
The levels in wipe tests varied considerably, de-
pending on the area in which the test was taken,
as well as when it was taken. The results of the
analysis on the samples described  in the experi-
mental section are given in Table  1. The recon-
structed ion  chromatograms for a  system blank,
an extract from  20 g of the cleaning  solution,
and  a  standard of the 22 TCDDs are shown in
Figure 2. The isomer specificity for the 1,3,6,8-
and  1,3,7,9-isomer is demonstrated in the figure by
the baseline separation of these two  isomers in our
analytical method. The criteria for a positive iden-
tification of  a TCDD are outlined  in Table 2.
   We have confirmed elevated levels of two com-
pounds that  have the  analytical characteristics of
1,3,6,8- and 1,3,7,9-TCDD in the cleaning solution
used to clean the floors in the building containing
    SYSTEM
     BLANK
    CLEANING
    SOLUTION
    EXTRACT
              |X15
                       1379
    CHROMATOGRAPHIC
      SEPARATION OF
      TCOO  ISOMERS
   0 h«S. 24MNS. I6SECS
                             0 HflS. 47MINS. SOSECS.
Fig. 2. Mass chromatograms (m/z  322) for a system
blank, cleaning solution extract, and  a standard mixture
of TCDD isomers.
                       Table 1. Summary of TCDDs' observed in Samples 1 and 2.
Retention
time
33:51
36:08
38:39

40:24

43:54
44:29
Relative
retention
timeb
0.772
0.824
0.882 •

0.922

1.001
1.015
TCDD isomer
(RRT of
standard)
1,3,6,8(0.773)
1,3,7,9(0.825)
1,3,7,8(0.882)
1,2,4,8
1,2,4,7(0.922)
1,3,6,9
2.3,7,8(1.002)
Various'
Concentration'
Blank Sample 1
ND «.l)
ND
ND

ND

NQ (0.2)'
ND
Concentration0
Sample 2
1.7d
NQ (0.7)
NQ (0.2)

NQ

NQ (0.2)'
NQ
Signal
to
noise
42
17
3.0

2.1

3.5
1.0
Isotope
ratio
(320/322)
0.73
0.86
0.78

0.69

0.84
0.70
'All concentrations are calculated assuming a response factor of 1 relative to 2,3,7,8-TCDD and are in part-per-trillion
 (ppt). The recovery based upon absolute areas of the internal standard is at least 55%.
"The relative retention time (RRT) is calculated on the basis of 1.00 for 2,3,7,8-13C,2-TCDD internal standard.
'When calculated concentration is below 1.0 ppt,  the determination is not quantifiable with the calculated amount
 given in parentheses,  for example, NQ (ppt); ND is not detected, with the limit of detection for the analysis given
 in parentheses.
dThis represents a minimum amount present since the I3C,2-TCDD internal standard was spiked  after the extrac-
 tion of the sample. In a separate experiment, a 20-g sample was spiked with internal standard before extraction and
 the overall recovery was 30%, with a calculated concentration of 2.6 ppt.
'The observed signal for 2,3,7,8-TCDD is indistinguishable from the amount present in the 2,3,7,8-l3C,2-TCDD in-
 ternal standard.
rl,2,3,8-, 1,2,3,7-, 1,2,4,9-, 1,2,4,6-, and 1,2,3,4-TCDD all elute at this retention time.

-------
                               Analysis of human adipose for dioxins
                                                                          359
     Table 2. Criteria for TCDD identification

Retention Time of  Each group of ions must max-
Monitored Ions
Ratio of Isotopic
Ions Monitored

Ratio of Signal
to Noise
imize within 1 s of each other.
The relative retention time (to
13C,2-TCDD) must be within 2
pans per thousand of the analyti-
cal standard.

For TCDDs, 0.77 ±0.1
(m/z 320/322)

Greater than or equal to 3.0
the dioxin laboratory. The three active ingredients
in this germicidal detergent are the sodium salts of
ortho-hydroxy biphenyl (5.5%), para-/-amylphenol
(1.1%), and ortho-benzyl-p-chlorophenol (3.3%).
This latter compound probably is prepared from
the sodium  or  potassium phenoxide and benzyl
chloride, followed by chlorination of the resulting
o-benzylphenol (US Paten* 1,967,825)  [3].
   The self condensation, along with the accom-
panying "Smiles" rearrangement [4] of either of
two different trichlorinated phenols (TCPs), could
account for  the possible presence  of  these two
TCDDs in the cleaning solution (see Fig. 3). One
of these TCPs  (2,4,6-TCP) could be formed  by
chlorination  of unreacted phenol in the last step
of the synthesis of ortho-benzyl-p-chlorophenol
(chlorophene).
   A mixture of compounds containing, among
other  substances,  2-hydroxy-2',3,4,4',5-penta-
chlorodiphenylether at the part-per-million level
was put through the cleanup procedure,  and no
                      Ci
                                                           1,3,6,8,-TCDD
                                                        -  1.3.7.9-TCDD


                          Fig. 3. Formation of 1,3,6,8- and 1,3,7,9-TCDDs.
                                                           1.2.3,8-TCDD

                                                           1,2,3,7-TCDD

                          Fig. 4. Formation of 1,2,3,8- and 1,2,3,7-TCDDs.
                                                  .
                                                    J

-------
 360
                                       D. G. PATTERSON  ET AL.
 TCDDs were detected.  If TCDDs  were formed
 during the cleanup procedure by interacting with
 acidic or  basic chromatographic  materials, then
 this ortho-hydroxydiphenyl ether should have pro-
 duced measurable amounts of l-,2,3,8- and 1,2,3,7-
 TCDDs (see Fig. 4).
    A review of the on-line toxicological informa-
. tion data  bases suggests that the 1,3,6,8-isomer is
 less  active in one biological test by -10~7  com-
 pared with the 2,3,7,8-isomer. Due to  the low
 concentration and the apparent  low toxicity of
 1,2,3,8-TCDD, we feel that this cleaning solution
 does not  present  a health  risk to  workers. The
 observation of this isomer in general laboratory
 contamination is analytically important in labora-
tories where ultra-trace dioxin analyses are being
conducted.

                 REFERENCES

1.  Lynch, R.E., G.R. Lee, and J.P. Kushner. 1975. Por-
   phyria cutanea tarda associated with disinfectant mis-
   use. Arch. Intern. Med. 135:549-552.
2.  Smith, L.M., D.L. Stalling, and J.L. Johnson. 1984.
   Determination of part-per-trillion levels of polychlo-
   rinated dibenzofurans and dioxins in environmental
   samples. Anal. Cliem. 56:1830-1842.
3.  Windholz, M. and S. Budauari. 1983. The Merck In-
   dex and Encyclopedia of Chemicals, Drugs, and Bio-
   logicals. Merck and Co., Inc., Rahway, NJ, p. 2364.
4.  Kende, A.S. and M.R. Decamp. 1975. Smiles rear-
   rangement in the synthesis of hexachlorodibenzo-p-
   dioxins. Tetrahedron Letters 33:2877-2880.
                                             r .1  r

-------
United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
P.O. Box 15027
Us Vegas NV 89114-5027
EPA 600/4-86-004
January, 1986
Research and Development
Protocol for the
Analysis of 2, 3, 7, 8-
Tetrachlorodibenzo-p-Dioxin
by High Resolution Gas
Chromatography/
High-Resolution
Mass Spectrometry

-------
PROTOCOL FOR THE ANALYSIS OF 2,3,7,8-TETRACHLORODIBENZO-£-DIOXIN BY
HIGH-RESOLUTION GAS CHROMATOGRAPHY/HIGH-RESOLUTION MASS SPECTROMETRY
                                 by
                 John S.  Stanley and Thomas M.  Sack
                     Midwest Research Institute
                    Kansas City, Missouri  64110
                      Contract Number SAS 1576X
                           Project Officer

                          Werner F. Beckert
                     Quality Assurance Division
             Environmental Monitoring Systems Laboratory
                      Las Vegas, Nevada  89114
             ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
                 Office of Research and Development
                U.S.  Environmental Protection Agency
                      Las Vegas,  Nevada  89114
                                    f 4 >~)
                                    U '-1 (L

-------
                                   NOTICE
     The information  in this document has been funded wholly  or in part by
the  United  States  Environmental  Protection Agency  under Contract  Number
SAS 1576X  to  the  Midwest Research Institute, Kansas City, Missouri.  It has
been subject to the Agency's peer and administrative review, and it has been
approved  for publication  as  an  Environmental Protection  Agency  document.
Mention of trade  names or commercial products does  not  constitute endorse-
ment or recommendation  for use.
                                     11

-------
                                   PREFACE
     This report  describes  the activities  completed as  part  of a  single-
laboratory   evaluation   of   a   high-resolution  gas   chromatography/high-
resolution mass  spectrometry  method for  the determination  of  tetrachloro-
dibenzo-£-dioxins in water, soil,  and  sediment samples.   The work described
in this  report was  completed  at  the Midwest Research Institute  under  con-
tract to  Viar and  Company  (Special Analytical Services SAS 1576X)  for the
U.S.   Environmental  Protection  Agency,   Environmental  Monitoring  Systems
Laboratory,  Quality Assurance Division, Las  Vegas,  Nevada.   The revision of
the protocol  to  allow  for lower quantitation limits for tetrachlorodibenzo-
£-dioxins was carried out at the Environmental Monitoring Systems Laboratory-
Las Vegas.

     This report was prepared  with assistance from M.  McGrath.   The authors
acknowledge  the technical project  monitor,  W. F.  Beckert, as well  as  R. K.
Mitchum and  S. Billets  of the Environmental  Monitoring  Systems Laboratory-
Las Vegas and, especially, Y. Tondeur  of  the Environmental  Research Center,
University of Nevada,  Las Vegas  for guidance  provided  during  this  study.
                                     111
                                            r

-------
                                  ABSTRACT
     This report provides the results of the single-laboratory evaluation of
a  high-resolution   gas  chromatography/high-resolution  mass  spectrometry
method  for  the  determination  of 2,3,7,8-tetrachlorodibenzo-£-dioxin  and
total  tetrachlorodibenzo-£-dioxins  at  concentrations  ranging  from  10  to
200 pg/g  (ppt)  in  soils  and 100 to 2,000 pg/L  (ppq)  in  water.   The report
summarizes  the  data  for  the precision  and  accuracy  of triplicate measure-
ments of five solid and five aqueous samples.  The results indicate that the
method is capable of  generating accurate and precise data within the concen-
tration limits  specified  above and within absolute recoveries  of 40 to 120
percent with 50 percent precision.  An attempt to reach a quantitation limit
for TCDD  of 2 ppt (or  less) for  soil  and 20 ppq (or less) for aqueous sam-
ples was  not  successful.   Based on the data generated during this study and
based  on  discussions at  the  Environmental Monitoring  Systems  Laboratory-
Las Vegas, the Environmental Monitoring Systems Laboratory-Las Vegas revised
certain parts  of the protocol to  lower the quantitation  limit for tetra-
chlorodibenzo-£-dioxins to 2 ppt in soil and 20 ppq in water samples.
                                     IV
                                       rr

-------
                                  CONTENTS
Preface	   iii
Abstract	    iv
Figures	    vi
Tables	   vii

   1.  Introduction 	     1
   2.  Conclusions	     3
   3.  Recommendations	     5
   4.  Experimental Procedures	     7
            Sample description.	     7
            Sample preparation	     7
            Reagents	     9
            HRGC/HRMS instrumentation 	     9
            Mass measurement accuracy 	    11
            Chromatographic resolution	    13
            Injection technique 	    13
   5.  Results and Discussion	    14
            Approach to cleanup column evaluation 	    14
            Final method evaluation 	    22

References	    ^2
Appendices	    43
   A.  Validated Analytical Protocol
   B.  Proposed Analytical Protocol
                                           rr

-------
                                   FIGURES



Number                                                                  Page



  1    Column cleanup procedures specified in the protocol	     15



  2    Column cleanup procedures proposed by the EMSL-LV	     16



  3    Background levels of  1,3,6,8- and 1,3,7,9-TCDD observed over

         the single-laboratory evaluation study 	     41
                                     vx
                                            ' -I
                                            /

-------
                                   TABLES

Table                                                                   Page

  1    Solid Samples Used for HRGC/HRMS Method Evaluation ......      8

  2    Aqueous Samples Used for HRGC/HRMS Method Evaluation .....      8

  3    TCDD Isomers Used for HRGC/HRMS Method Evaluation .......     10

  4    Composition of Concentration Calibration Solutions (pg/pL) .   .     10

  5    HRGC/HRMS Operating Conditions ................     12

  6    Recovery (%) of Several TCDD Isomers from Cleanup Option A .   .     16

  7    Recovery (%) of Several TCDD Isomers from Cleanup Option B ..   .     19

  8    Recovery (%) of Several TCDD Isomers from Cleanup Option C .   .     20

  9    Recovery (%) of Several TCDD Isomers from Cleanup Option D .   .     21

 10    Initial Calibration Summary ..................     23

 11    HRGC and Mass Resolution Check Summary ..... .......     24

 12    TCDD Data Report Form .....................     ?.6

 13    Accuracy and Precision of the HRGC/HRMS Analysis for
         2,3,7,8-TCDD from Laboratory Aqueous Matrix Spikes .....     33
 14    Precision of the HRGC/HRMS Analysis for 2,3,7,8-TCDD of
         Soil and Fly Ash Samples ..................     34

 15    Accuracy of the HRGC/HRMS Method for the Determination of
         TCDD Isomers Spiked into Aqueous Matrices ..........     35

 16    Accuracy of the HRGC/HRMS Method for the Determination of
         TCDD Isomers Spiked into Soil Matrices ...........     36

 17    Fortified Field Blank Results .................     37
                                     VII
                                               rr /? o
                                               
-------
                                  SECTION 1

                                INTRODUCTION

     The U.S.  Environmental Protection Agency's  (EPA)  strategy  for dealing
with dioxin  requires  the  development and validation of an analytical method
capable of  achieving detection of  the  tetrachlorodibenzo-£-dioxins (TCDD),
specifically 2,3,7,8-TCDD, at the parts-per-trillion (ppt) level in soil and
sediment and parts-per-quadrillion  (ppq)  level  in water.1   This validated
method will be used by qualified contract laboratories to extend the analyt-
ical capabilities for such analyses to all EPA regional and program offices.

     This report deals specifically with the single-laboratory evaluation of
a  high-resolution  gas   chromatography/high-resolution  mass  spectrometry
(HRGC/HRMS)  analysis  method for  TCDDs  in  soil,  sediment,  and  water.   The
method  (Appendix A)  is  intended to provide quantitative  determination of
TCDD at levels  of 10 to 200 pg/g  (soil  and sediment)  and 100 to 2,000 pg/L
(water) at  a mass resolution of  10,000.   This  single-laboratory evaluation
has been  completed as  part of the  validation process  recommended by EPA.2

     The proposed method  was prepared after  several  candidate  methods were
reviewed and their best features were selected.   After peer review, the pro-
posed method was  refined  for completeness, technical accuracy,  clarity, and
regulatory applicability.   The  single-laboratory evaluation of the proposed
analytical method has been accomplished through three tasks.  The first task
involved  preliminary   performance  testing  of  the   method  using  TCiJD-
contaminated  soils  and TCDD-spiked aqueous  samples.   The results  o* this
study  indicated  that the  proposed  method  required modification to achieve
the target method detection limits and the accuracy and precision criteria.
The second task focused on ruggedness testing of the chromatographic cleanup
procedures.   The results  of this  study were used  to modify  the proposed
method.  This  report is  focused  on the results  of  the triplicate analysis
of five  solid and five  aqueous  samples  completed  under the third task of
the evaluation, using the modified method.

     Section  2  of  this  report  summarizes  the conclusions based  on  the
single-laboratory  evaluation of  this  method using TCDD-contaminated soils
and TCDD-spiked  aqueous samples.   Section 3 presents  recommendations that
should  be  considered  for  inclusion in the  method before proceeding with
collaborative testing.   Section 4 presents  some  specific experimental con-
ditions, and  Section 5 summarizes  the  analytical  data  for  the triplicate
analysis of  four soil, one  fly ash, and five aqueous  samples  completed in
the third task  of the single-laboratory evaluation.  Triplicate analyses of
a l-pg/(jL calibration  solution  did not give satisfactory results.  In order
to achieve a quantitation limit of 2 ppt for soil (using a 10-g sample) and
                                        rr /?

-------
20 ppq  for water  (using  a  2.0-L sample),  the  protocol evaluated  in  this
study was  modified.  The  rationale for  the modifications  and  the  revised
protocol are included as Appendix B.

-------
                                  SECTION 2

                                 CONCLUSIONS

     The  single-laboratory  evaluation  of  the  analytical  method for  the
determination of 2,3,7,8-TCDD  in  soil and aqueous samples demonstrates that
the method as  described  is capable of achieving the target detection limits
of 10 pg/g (ppt) for soils and 100 pg/L (ppq) for water.

     The relative  response  factors  (RRF)  determined for native 2,3,7,8-TCDD
versus the internal standard 13C12-2,3,7,8-TCDD, and the RRF of the internal
standard versus the recovery standard 13C12-1,2,3,4-TCDD over the five-point
concentration calibration  curve demonstrate that the HRGC/HRMS method main-
tains a  linear response  for  2,3,7,8-TCDD from 10 to 200 ppt  for soils  and
100 to 2,000 ppq for water.

     The results of  the  analysis  of spiked aqueous samples demonstrate that
internal standard  (isotope  dilution)  quantitation provides an accurate mea-
surement of  2,3,7,8-TCDD.   The accuracy of the 2,3,7,8-TCDD measurement for
triplicate analysis  of  four water samples  spiked  at various concentrations
was quite good.  The accuracy of measurement  for  2,3,7,8-TCDD averaged  104
percent  for  three  aqueous matrices prepared  as laboratory matrix spikes.
The absolute recovery of  the  internal standard  13C12-2,3,7,8-TCDD did  not
significantly affect  the accuracy  of the  2,3,7,8-TCDD determination.   The
precision of the analyses for 2,3,7,8-TCDD ranged from 3.6 to 16 percent for
replicate analyses of the five aqueous samples.  The precision of the crip-
licate analyses of  the  soil samples was somewhat higher than determined -"or
aqueous  samples.   The  precision  of  triplicate analyses  of the  four  soil
samples ranged from 19 to 50 percent.   The difference in precision from that
of the aqueous samples may be attributable to the potential for TCDD aC£,urp-
tion on the soil samples.

     The results from the analyses  of soil  and aqueous samples spiked with
additional TCDD isomers  demonstrate  that the internal standard quantitatiun
gives good estimates  of  total TCDD values.  The accuracy of the analyses of
fortified distilled  water  and influent  and effluent  wastewaters averaged
101 ± 14 percent for five TCDD isomers.   The accuracy of the measurements of
these isomers for  the  four fortified soil samples averaged 87 ± 24 percent.

     The results of the  analyses demonstrate that the requirements for abso-
lute recovery of the internal standard (40 to 120 percent) and precision of
replicate analyses  (RPD < 50 percent) can be  achieved  for  relatively clean
samples.
                                         rr ^ \
                                         i \ 11 i.

-------
     The sample matrix can  severely impact the performance of the analytical
method.  This  is  evidenced by the consistent low  recovery  of the internal
standard from the  fifth aqueous sample, an industrial wastewater, and from a
fly ash sample.  The low recovery from the industrial wastewater is possibly
due  to the effect of coextractants  on  the elution  sequence from alumina.
The  low  recoveries observed for the  fly  ash sample,  on the other hand, may
be attributed to adsorption by the sample matrix.

     One of the most critical variables in the analytical method is the com-
pleteness  of  removal  of the benzene  from the extract before proceeding with
the  acidic alumina column fractionation.   The  cleanup  column ruggedness
testing experiments demonstrated that the recoveries of 2,3,7,8-TCDD and the
other  TCDDs  are affected  by the presence of benzene  in the alumina column
fractionation step.

     The analyst  must be  aware of  the  potential  problem  of interferences
arising  from  background   contamination.   For  example,  the 1,3,6,8-  and
1,3,7,9-TCDD  isomers  were  present  in the  fortified  field  blanks  in this
work.  From other  referenced activities it becomes clear that these isomers
may  present  problems in  other laboratories  as  well.    The  fortified field
blanks are important tools in  assessing  the background contamination prob-
lems over  time.

     Although  the  l.O-pg/pL standard did not yield satisfactory results in
this study, due  to unacceptable ion  ratios, the response factors are within -
the established curve.  The data for  the triplicate analyses  of  the 1.0-|jg/|jL
standard demonstrate that the characteristic ions for TCDD were  greater than
20:1 for the  m/z 322  S/N and approximately  10:1  for  m/z 259 S/N.   Thus, it
should be  possible to extend the detection limit to 1 pg/|jL  if  an allowance
for abundance ratios based  on ion statistical errors is  incorporated.

     Based on  the  column performance  and bleed  characteristics, the column
of cfiolrs  for the  analysis  for TCDD at ppt (for soils and sediments) and ppq
(for. water)  levels  appears  to be  the  50-m CP-Sil  88 with  a  0.2-pm film
thickners.   To preserve the performance characteristics  of the HRGC columns,
an injection technique that excludes  any air is highly  recommended.
                                              r-- c;
                                              O o <

-------
                                 SECTION 3

                              RECOMMENDATIONS

1.  Mass measurement accuracy should properly be determined relative to the
    lock mass  (if  any),  rather than m/z 254.9856, because it is that rela-
    tionship which  will determine  how accurately the  masses of  the  TCDD
    ions will be measured.

2.  It  is  recommended  that  the chromatographic  resolution  check  be  per-
    formed on  the  summed ion chromatograms of m/z 259 + m/z 320 + m/z 322.
    This yields a  chromatogram which is less noisy and more representative
    of the true column performance.

3.  The 5  percent peak  width criterion for mass  resolution  should be the
    selected  mass/1,000 mmu  rather  than  31.9  mmu  because  the  protocol
    allows peaks other  than  m/z 319 to be  used  for  resolution measurement
    (e.g., 31.7 mmu if m/z 317 is used).

4.  It  is  recommended that  the mass measurement  accuracy  be recorded and
    reported along with the resolution check summary table.

5.  The  addition  of  the  recovery standard  13C12-1,2,3,4-TCDD should  be
    achieved by using a spike volume  of  25 to 50 |JL  rather than 5 pL to
    minimize errors resulting from volume measurement.

6.  The recommended  temperature program  settings  in the method shou.M be
    converted  to  those presented  in  the experimental  section of  this re-
    port.   These conditions were established for analysis with tridecane r.s
    the solvent.

7.  Lower  limits  of detection  can be achieved by allowing  the analyse to
    concentrate the final  extract  to  as low as 10 pL.  It may be necessary
    to use the smaller  final volume with other HRMS instruments to achieve
    the same levels of detection.

8.  The method should recommend several  techniques  to break up emulsions
    resulting  from  extraction of aqueous samples.   In  this  evaluation the
    emulsion phase was  put through a column packed  with glass wool, which
    was  then  rinsed  with additional  methylene  chloride.    Other  options
    might include  stirring or centrifugation of the emulsion phase.

9.  The method should specify  the procedure to deal with  aqueous samples
    containing  high  levels  of  suspended  solids.   In  this  study  it  was
                                            i- c. '\
                                            •J uO

-------
     necessary to centrifuge  the  soil extract sample before proceeding with
     the extraction.

10.   It is highly recommended that the method be modified such that the ben-
     zene extract is  completely exchanged to hexane prior to cleanup on the
     silica column since this is apparently one of the most critical factors
     leading to successful sample analysis.

11.   It may be worthwhile to evaluate a cleanup procedure in which the char-
     coal column  precedes the  alumina  column as a means  to  improve method
     recovery.

-------
                                  SECTION 4

                           EXPERIMENTAL PROCEDURES
SAMPLE DESCRIPTION

     Five  solid  samples  were  provided  by  the  Environmental  Monitoring
Systems  Laboratory-Las  Vegas   (EMSL-LV)  to  the  Midwest  Research Institute
(MRI)  for analysis  for 2,3,7,8-TCDD  and total  TCDD using  the  analytical
method in Appendix A.  A description of the five solid samples and the esti-
mated  2,3,7,8-TCDD  concentrations from previous analyses  by an independent
laboratory are provided in Table 1.   Each sample was analyzed in triplicate
as specified  in  the protocol.   One of the triplicate samples for each soil
sample was spiked with the seven TCDD isomers  (1,3,6,8-; 1,3,7,9-; 1,2,3,7-;
1,2,3,8-;  1,2,3,4-;  1,2,7,8-;  and  1,2,8,9-TCDD)  at  approximately  10 times
the estimated level of 2,3,7,8-TCDD specified in Table 1.

     Five aqueous samples  were generated for  the evaluation of the analyt-
ical method at  the  ppq detection level.  Table 2 presents a description of
each water type and lists the fortification levels of 2,3,7,8-TCDD and seven
additional  TCDD  isomers  (1,3,6,8-;  1,3,7,9;  1,2,3,7-;  1,2,3,8-;  1,2,3,4-;
1,2,7,8-; and 1,2,8,9-TCDD) in each sample.

     The  influent  and  effluent  wastewater  samples  were collected  from a
sewage treatment facility in metropolitan Kansas City, Missouri.  The indus-
trial wastewater was  obtained  from a holding  pond  within a hazardous waste
area that was  known to be highly contaminated with PCBs and possibly other
chlorinated aromatic compounds  (chlorobenzenes).   This aqueous  sample  was
very acidic (pH < 1) and was dark in color.

     The soil extract was prepared from 30 g of a soil sample, Hyde Park 002
(H2), and  1 gallon  of distilled water.  The mixture  was stirred constantly
(at least 24 hrs) until just prior to subsampling of 1.0-L aliquots.

SAMPLE PREPARATION

     All  samples  listed in Tables  1  and  2  were extracted  and analyzed in
triplicate according  to the protocol provided  in Appendix A.   As indicated
in Tables 1  and 2,  one aliquot  of  each sample  matrix was  fortified  with
additional  TCDD  isomers,  which  represent  the  compounds  that  elute first
(1,3,6,8-TCDD),  last (1,2,8,9-TCDD),  and  within the  approximate retention
window of 2,3,7,8-TCDD  (1,2,3,7-;  1,2,3,8-;  and 1,2,3,4-TCDD) from the HRGC
columns used for sample analysis.

-------
        TABLE 1.  SOLID SAMPLES USED FOR HRGC/HRMS METHOD EVALUATION


EPA sample no.
B25-Piazza Road (B5)
Hyde Park 001 (HI)
B52-Shenandoah (Bl)
Hyde Park 003 (H3)
RRAI-5,7,8 (FA)

Matrix
Soil
Soil
Soil
Soil
Fly ash

Approximate
sample
size
10 g
10 g
1 8
1 8
10 g
Estimated
2,3,7,8-TCDD
concentration
(ppt)B
50
70
360
1,700
NRd

Spike level
(ppt.) of
TCDD isomers
100
140
720
1,700
e

.Approximate sample size of each replicate sample.
 Estimated level of endogenous 2,3,7,8-TCDD reported to MRI by
 W. Beckert in letters dated April 19, 1985 and August 30, 1985.
 .Approximate fortification level of each of seven additional TCDD isomers.
 No estimate of 2,3,7,8-TCDD concentration was reported.
 Additional TCDD isomers were not spiked into this matrix.
      TABLE 2.  AQUEOUS SAMPLES USED FOR HRGC/HRMS METHOD EVALUATION	

                                                Fortification   Fortification
                                  Approximate     level of        level of
                                     sample     2,3,7,8-TCDD    TCDD isomers
         Sample type                 size           (ppq)           (ppq)
Distilled water (DW)
POTW influent (IWW)
POTW effluent (EWW)
Industrial wastewater (IND)
Hyde Park 002; soil extract (H2W)
1
1
1
1
1
.0 L
.0 L
.0 L
.0 L
.0 L
250
500
1,000
500
c
500
1,000
2,000
1,000
c

.Approximate sample size of each replicate sample.
 Approximate fortification level of each of seven additional TCDD isomers.
 This aqueous sample was not fortified with TCDD isomers.

                                      8
                                       u

-------
     All  samples  were  fortified with  500 pg 13C12-2,3,7,8-TCDD  in 1.5 mL
acetone.   The solid  samples were  extracted  continuously for  24 hr  in  a
Soxhlet  apparatus  with benzene  and  the  1.0-L  aqueous samples  were batch-
extracted using 2.0-L separatory funnels and three 60-mL portions of methyl-
ene chloride.  The extractions of the influent wastewater (IWW) and effluent
wastewater  (EWW) and  the  soil extract  (H2W) were  complicated by the forma-
tion  of emulsions.    In each case,  the emulsion was  removed  by passing the
methylene  chloride  and emulsion  layer  through a  column  of  glass  wool pre-
rinsed  with methylene chloride.  The extract and resulting aqueous layers
were collected in a sample bottle and the glass wool plug was rinsed with an
additional  10 mL  methylene chloride.  Following the  complete extraction of
the aqueous  sample,  the  contents of the  bottle  were  transferred to a clean
250-mL  separatory  funnel  and the methylene chloride  was removed  from the
aqueous  phase that was transferred  with  the  emulsion.  All  extracts were
concentrated  with  Kuderna-Danish  evaporators and nitrogen  evaporation to
approximately  1.0  mL.  Each  extract was  taken  through  the  entire cleanup
procedure  including   the  acidic silica,  acidic  alumina,  and  Carbopak C as
specified  in  the protocol (Appendix A).  The HRGC/HRMS analysis of each ex-
tract was completed as specified below.

REAGENTS

     All solvents  for extraction and cleanup were  obtained  as "Burdick and
Jackson  distilled-in-glass"  quality.   The tridecane (99 percent purity) was
obtained  from Aldrich (TS,  740-1).   The  chromatographic materials, acidic
alumina  (100-200  mesh AG-4,  Biorad  Laboratories  132-1340),  silica (70-230
mesh Kieselgel 60,  EM Reagent, American Scientific Products C5475-2), sodium
sulfate, potassium  carbonate, Celite 545®  (Fisher Scientific Company), and
the silanized glass   wool  and Carbopak C  (80-100  mesh Supelco 1-1025) were
prepared  for  use  as  specified  in Section 7 of the  protocol (Appendix A).

     Table  3  provides the sources of standards used to prepare the calibra-
tion  solutions,  sample fortification  solutions,  recovery  standard spiking
solution,  internal  standard  spiking solutions,  field  fortification solu-
tions, and TCDD isomer fortification solutions.

     Table  4  is  a summary of the concentration calibration standards pre-
pared for the HRGC/HRMS method evaluation.  These standards were prepared as
specified  in  the protocol (Appendix A) .  The standard HRCC6 was included in
the final  evaluation  of  the HRGC/HRMS method as a means  to demonstrate the
lower limit of detection under optimum instrumental conditions.

HRGC/HRMS INSTRUMENTATION

     Sample extracts  and  calibration standards were  analyzed  using a Carlo
Erba Mega  Series gas  chromatograph  (GC) which was  coupled to a Kratos MS50
TC double-focusing mass spectrometer (MS).  The GC/MS interface was simply a
direct connection of  the  GC column to the ion source via a heated interface
oven.   A Finnigan  2300  Incos data system was  used for data acquisition and
processing.
                                        r - ,.,
                                        U u I

-------
         TABLE 3.  TCDD ISOMERS USED FOR HRGC/HRMS METHOD EVALUATION
       Isomer
      Stock
  concentration
     Source
Standard code
2,3,7,8-TCDD
13C12-2,3,7,8-TCDD
1,2,3,4-TCDD
13C12-1,2,3,4-TCDD
l,3,-6,8-/l,3,7,9-TCDD
1,2,3,7-/1,2,3,8-TCDD
1,2,7,8-TCDD
1,2,8,9-TCDD
Column performance
standard
7.87 ± 0.79

  50 ± 5

    2.7 mg/mL

  50 ± 5 Mg/mL

    0.82 mg/mL

    0.5 mg/mL

    0.39 mg/mL

    1.46 mg/mL

   10 (Jg/mL
EPA QA Reference
Materials

Cambridge Isotope
Laboratories

Cambridge Isotope
Laboratories

Cambridge Isotope
Laboratories

Cambridge Isotope
Laboratories

Cambridge Isotope
Laboratories

Cambridge Isotope
Laboratories

Cambridge Isotope
Laboratories

Cambridge Isotope
Laboratories
20603

R00201 (Lot
AWN-1203-65)
ED-915C (Lot
6578)
R00212 (Lot
AWN-1203-93)
ED-913C (Lot
F2086)
ED-905C (Lot
7371)
ED-915C (Lot
7184)
ED-916C (Lot
MLB-682-26)
ED-908 (Lot
No. R00215)
a...
 Mixture of TCDD isomers including 2,3,7,8-; 1,2,3,4-; 1,2,3,7-/1,2,3,8-;
 1,2,7,8-; and 1,4,7,8-TCDD.
    TABLE 4.  COMPOSITION OF CONCENTRATION CALIBRATION SOLUTIONS  (pg/ML)

HRCC1
HRCC2
HRCC3
HRCC4
HRCC5
HRCC63
Recovery standard
13C12-1,2,3,4-TCDD
2.5
5.0
10.0
20.0
40.0
1.0
Analyte
2,3,7,8-TCDD
2.5
5.0
10.0
20.0
40.0
1.0
Internal standard
13C12-2,3,7,8-TCDD
10.0
10.0
10.0
10.0
10.0
10.0

 This solution is not specified in the analytical method in Appendix A.
                                      10
                                       JoO

-------
     The HRGC/HRMS operating conditions used in the final phase of this work
are summarized in  Table  5.   The GC operating  conditions  recommended in the
protocol were  not used  for  these analyses  for three  reasons.   First,  the
TCDDs have rather long retention times, and the solvent (tridecane) boils at
235°C.  Thus  no  benefit could  be realized with a  low initial temperature.
Second, past  experience  at  MRI  has  indicated  that 200°C  is  an acceptable
starting temperature  for these  types  of analyses when tridecane is used as
a  solvent.   Finally,  since  the  CP-Sil 88  and SP-2330  phases  are both very
polar and thinly  coated, it has been recommended that they not be subjected
to  rapid heating  or  cryogenic  cooling  to  prevent   thermal  shock to  the
column.3

     The MS  was  tuned daily to  yield  a resolution of  at  least  10,000 (10
percent valley) and optimal response  at m/z 254.986.  This step was followed
by calibration of an accelerating voltage scan beginning at m/z 254 (typical
calibration  range  was 255  to  605 amu).  Other voltage scans  from the same
data  file were then  used  to establish and document both  the  resolution at
m/z 316.983  and the mass  measurement  accuracy at m/z 330.979.

MASS MEASUREMENT ACCURACY

     For this  work, mass measurement  accuracy was  measured  relative to PFK
m/z 254.986, as  required by the protocol, by  applying the mass correction,
Am, to  the  entire  spectrum,  which yields  an  error  of 0 ppm at m/z 254.986.
In this way,  it  was possible to meet routinely the 5 ppm accuracy criterion
at m/z  330.979.   However,   if  a  lock  mass other than  254.986 is used,  the
mass  measurement  accuracy  should  be  measured  relative  to that  lock  mass,
since  it  is that peak which is  used  to maintain magnet  alignment and will
ultimately control  the mass  measurements during the selected  ion monitoring
(SIM) experiments.

Mass Resolution

     Mass resolution at m/z 316.983 was documented by an output of the Incos
PROF  program.   However,  the  computer-generated  value  for resolution  was
found  to  be significantly  higher than the value measured manually.  Thus,
the manually determined  resolution, which  was nearly identical to the value
measured by  using the peak  matching unit, is reported.  Closer inspection of
the PROF source  code  revealed that resolution is computed via a statistical
method, not  as m/Am  at  5  percent height.   Incos users  should therefore be
aware  of  this  discrepancy,  because  the computer-generated value  can  be as
much as 20 percent over the proper value.

     Following calibration, the SIM experiment descriptor was updated to re-
flect the new calibration.   Six masses (see Table 5) were monitored by scan-
ning ^ m/10,000 amu over  each mass.  The total cycle time was kept to 1 sec.
The m/z 280.983  ion from PFK was used as a lock mass because it is the most
abundant PFK  ion within  the range of  m/z  255 to 334  and  therefore permits
the use of  low partial pressures of PFK,  which minimizes PFK interferences
at the analytical masses.
                                      11
r c.
                                          q
                                          J

-------
                  TABLE 5.  HRGC/HRMS OPERATING CONDITIONS
Mass spectrometer
          Accelerating voltage:
          Trap current:
          Electron energy:
          Electron multiplier voltage:
          Source temperature:
          Resolution:

          Ions monitored

             258.930
             319.897
             321.894
             331.937
             333.934
             280.9825 (lock mass)
          Overall SIM cycle time

Gas chromatograph

          Column coating:
          Film thickness:
          Column dimensions:

          Helium linear velocity:
          Helium head pressure:

          Injection type:
          Split flow:
          Purge flow:
          Injector temperature:
          Interface temperature:
          Injection size:
          Initial temperature:
          Initial time:
          Temperature program:
= 1 sec
           8,000 V
             500 pA
              70 eV
           2,000 V
             280°C
          10,000 (10% valley definition)

           Nominal dwell times  (sec)

                     0.15
                     0.15
                     0.15
                     0.15
                     0.15
                     0.10
           CP-Sil 88
           0.2 (jm
           50 m x 0.22 mm ID

           ~ 25 cm/sec
           1.75 kg/cm2 (25 psi)

           Splitless,  45 sec
           30 mL/min
           6 mL/min
           270°C
           240°C
           2 |JL
           200°C
           1 min
           200°C to 240°C at 4°C/min
                                      12

-------
CHROMATOGRAPHIC RESOLUTION

     Chromatographic resolution values were measured for the SIM plot of m/z
320.  However,  it may  be  advantageous to measure chromatographic resolution
from a plot  of the  sum of m/z  259,  320,  and 322.   The sum trace has better
signal-to-noise ratio  (S/N)  and peak definition than the  SIM  plots,  which
permits a more accurate measurement of resolution.

Selection of the HRGC Column

     Three different HRGC columns were evaluated in the course of this proj-
ect:  SP-2330  (60 m x  0.24 mm); DBS (60 m x 0.22 mm); and CP-Sil 88 (50 m x
0.22 mm).  By  evaluating  the  mass spectra of the  bleed from each column at
240 to 250°C,  it  became apparent that the column background may be the lim-
iting factor in achieving the desired detection limit for this method.  The
DBS column provided the least  amount of background at 250°C, and the SP-2330
had the worst.   This coincides with the fact that quantitation at the detec-
tion limit (i.e., 2.5  pg/HlO  with the SP-2330 column was difficult at best.
The CP-Sil  88   column  appeared  to  offer less bleed  than  the  SP-2330 column
and indeed does permit more  accurate quantitation due to reduced background
contribution.

     The chromatographic  performance  afforded by these columns is a further
issue, since the  column best  suited for  low detection limits,  DB-5, cannot
meet the 25 percent valley chromatographic resolution criteria in all cases.
Both the SP-2330  and CP-Sil  88 columns can easily resolve the 2,3 , 7 ,8-TCDD.
However, based  on the  bleed  considerations discussed above, the 50-m CP-Sil
88  column  is  recommended for the  best combination  of  low bleed  and good
isomer separation.

     It may  also  be advisable  that other HRGC  columns  (including SP-2340,
Silar IOC,  and SP-2331) that have, been used for 2,3,7,8-TCDD analysis at the
1-ppb soil level be evaluated  for background contribution and their applica-
tion for HRMS analysis at ppt  and ppq concentrations.

INJECTION TECHNIQUE

     The HRGC  column performance can degrade very  quickly  if proper injec-
tion  techniques are  not  used.   Specifically,   the SP-2330  and CP-Sil  88
phases are very sensitive to  02 and  will  decompose rapidly at 200°C if any
trace of 02 is present.  Therefore, the common practice of using 1 |JL of air
to  flush  the  syringe  and effect reproducible injections is  to be avoided,
since even that small  amount  of air  per  injection can cause column perfor-
mance to degrade in less than  one week of continued use.

     The following  injection  technique  is recommended.  First rinse the sy-
ringe copiously with isooctane (or other volatile solvent, such as toluene).
Dry the syringe by  drawing air through it.  Pull  up and expel several vol-
umes of tridecane until all bubbles are gone, and leave 1 pL of tridecane in
the barrel.  Finally,  pull  up 2 pL of the sample solution and inject.  This
technique has  worked very well  and yields injection reproducibility compar-
able to that of the air purge method, without introducing air onto the ana-
lytical GC column.
                                      13
                                         r (^ i
                                         obl

-------
                                  SECTION 5

                           RESULTS AND DISCUSSION

     The primary  purpose of any method validation process is to assure that
the  method  under  consideration  is  adequate to meet  testing and monitoring
requirements.1  The  single-laboratory  evaluation  of the analytical protocol
presented in this report  has been preceded  by several  evaluation  and im-
provement steps.  These  have included the preparation of a written protocol,
technical review  of  the protocol for  completeness,  technical accuracy, and
clarity;  preliminary  testing  to  evaluate  performance  of  the  analytical
method;  and revision  and  refinement  of  the  written protocol  based on the
results of  the preliminary testing.

     Prior  to the assessment of the refined protocol presented in Appendix A,
the  proposed analytical method  had  been  evaluated  for performance  through
the  analysis  of  several duplicate samples.  The results  of the preliminary
evaluation  indicated that problems existed in the design and approach to the
extract  cleanup  steps,  which  greatly  affected the  method detection limit,
accuracy, and precision.

     This section presents a  summary  of  the  studies  that  have  led to the
refinement  of the analytical  protocol as provided  in Appendix A and  also
summarizes  the single-laboratory evaluation of this protocol.

APPROACH TO CLEANUP COLUMN EVALUATION

     The initial  method evaluation completed under  the first task resulted
in very low recoveries of the internal standard, 13C12-2,3,7,8-TCDD,  and the
accuracy and precision  of  duplicate  sample analyses were poor.   After re-
viewing the data,  it was apparent that the problems were the result of poor
chromatographic separation in the cleanup columns.  The initial protocol in-
volved reducing sample extract volumes to 1.0 mL in benzene, elution through
the  acidic  silica column  with hexane,  and  collection of  the total eluent
which was then  added to the acidic alumina  column.   The alumina column was
further  eluted  with hexane/20-percent  methylene  chloride.   The  eluate was
concentrated  and  cleaned further using a  Carbopak C/Celite column,  and the
TCDDs were  eluted with 2 mL toluene.

     Column  cleanup  techniques  were  revised and further evaluated following
the  procedures  depicted in  Figures  1  and  2.   The  column evaluations were
completed with triplicate measurements at'three spike levels  (0.10, 1.0, and
10 ng) equivalent  to  10, 100, and 1,000 ppt of TCDD in solids with several
TCDD  isomers (2,3,7,8-;  1,3,6,8-;  1,3,7,9-; 1,2,3,4-;  1,4,7,8-;  1,2,3,7-;
1,2,3,8-; and 1,2,8,9-TCDD).

                                      14

-------
          OPTION A
        1 ml Benzene Extract
         H2SO4 - SiO2
         4.0g
             Si O2
             l.Og
          Acidic AI2O3
          6.0g
                30mL 20% CH2Cl2/Hexane
       Concentrate to 100/it
        Carbopak C/Celite
                6mL Toluene
           HRGC/HRMS
    OPTION B
1 ml Benzene Extract
  H2SO4 - SiO2
  4.0g
      SIO2
      l.Og
  Acidic A1203
  6.0g
        30mL20%
 Carbopak C/Celite
                                                    6mL Toluene
    HRGC/HRMS
Figure  1.    Column   cleanup  procedures   specified  in  the  protocol

                                    15
                                    583

-------
           OPTION  C
       1 mL Benzene Extract
         H2SO4 - SiO2
             Si O2
             l.Og
      Concentrate to O.SmL
              I
         Acidic AI2O3
         o.Og
                30mL 20% CH2Cl2/Hexane
      Concentrate to 100/it.
        Carbopak C/Ceiite
                6mL Toluene
           HRGC/HRMS
    OPTION D
1 mL Benzene Extract
                                              4.0g
         - SiO2
      SiO2
      l.Og
                                           Concentrate to O.SmL
       I
  Acidic AI2O3
  o.Og
                                                     30 ml 20% CH2Cl2/Hexane
 Carbopak C/Celite
                                                     6mL Toluene
                                                HRGC/HRMS
Figure 2.   Column  cleanup procedures  proposed  by the EMSL-LV.

                                    16
                                        564

-------
     The TCDD isomers were  added to 1-mL portions of benzene and were taken
through the four  sample  cleanup sequences depicted in Figures 1 and 2.   One
of the  replicates  for  each  procedure was also spiked with 100 ng of Aroclor
1260.

     The results  of  the  sample analyses are provided in Tables 6 through 9.
As noted  in Tables  6  and  7,  recoveries of  the  TCDD  isomers  were  low and
quite variable  for the  early  eluting isomers 1,3,6,8-  and  1,3,7,9-TCDD as
compared to 1,2,8,9-TCDD.   Recovery of 1,2,8,9-TCDD was still low and vari-
able (approximately 60 percent recovery with an RSD of -v 20 percent).   These
results were generated using the procedures specified in the original proto-
col  (see Figure 1).  The results of the  analyses  following  the cleanup op-
tions A and B demonstrate that accurate quantitation of all TCDD isomers is
not possible using only  the 13C12-2,3,7,8-TCDD surrogate standard.  The low
recoveries measured  for  options A and B are obviously a result of the pres-
ence of benzene in the  eluent  from  the  acid-modified silica column that is
taken directly through the acidic alumina column.

     In contrast,  options C and D (Tables 8 and 9) demonstrate quantitative
recovery of the TCDD isomers.   Some background contamination has been noted
from the  acidic alumina  for the  1,3,6,8-  and 1,3,7,9-TCDD  isomers.   This
material had  previously  been prepared by Soxhlet  extraction with methylene
chloride and activation  at  190°C prior to use.  As noted in Tables 8 and 9,
the  average  recovery of  the other spiked TCDD isomers  was  greater  than 84
percent.

     When the recoveries of the different isomers and the 13C12-2,3,7,8-TCDD
are compared, the  average relative percent difference ranges from 1 percent
for 2,3,7,8-TCDD  (Table  3)  to  24 percent for 1,2,3,4-TCDD (Table 4).   These
results demonstrate  that either of  these cleanup  procedures (options C and
D) will provide good recovery and reliable quantitation of 2,3,7,8-TCDD and
very good estimates  of  the  concentrations of the other TCDD isomers present
in the  samples.   No  interferences were observed  in  the samples spiked with
100 ng  Aroclor  1260.   The lack of PCB interferences was especially noted in
the extracts of samples spiked at 0.10 ng/TCDD isomer.

     In addition to the evaluations of the cleanup procedures presented above,
the acid-modified silica gel/acidic alumina columns and the Carbopak C/Celite
column  were evaluated  separately.  Evaluation of  the silica/alumina  at the
0.10-ng spike level  as  shown in Figure 2 resulted in an average recovery of
120 percent for 1,2,3,4-, 1,2,3,7-,  1,2,3,8-, and 1,4,7,8-TCDD; 114 percent
for  2,3,7,8-TCDD;  118 percent   for  13C12-2,3,7,8-TCDD;  and  118  percent for
1,2,8,9-TCDD.    The results  for the recovery of  1,3,6,8-  and 1,3,7,9-TCDD
indicated that some contamination originated from the acidic alumina.

     Replicate analyses  of the Carbopak C/Celite column at the 0.10-ng spike
level resulted in average recoveries of 97 percent for 1,3,6,8-TCDD; 88 per-
cent  for  1,3,7,9-TCDD;  81  percent  for  1,2,3,4-,  1,2,3,7-,  1,2,3,8-,  and
1,4,7,8-TCDD;  75 percent for 2,3,7,8-TCDD; 96 percent for 13C12-2,3,7,8-TCDD;
and  90  percent  for 1,2,8,9-TCDD.   Elution  of the  Carbopak C/Celite  column
with additional toluene  beyond 6 mL did not improve recoveries even for the
samples spiked at 10 ng/TCDD isomer.

                                      17
                                             r~ f ~~
                                             ubo

-------
                                                       TABLE 6.  KECOVEKY (%) Of SEVERAL TCDO  I.SOUtUS  HiOH CI.EANUi' OPTION A
           00
CT
c:
Recovery (%) ot
Spike
level

InL Bcfutnc Ealrocl
1
H2SO4 - SIO2
4.09
Si O2
I.OR
1
Acidic AljOj
10ml 20%
Concentrate to 100/iL
1
Cwbopok C/C.liu
6mLlolu«n
HRGC/HRMS

1 ng
1 ng
1 ng
10 ng
10 ng
ChjC.^...™, ^ ^
Mean
% RSI)
1,3,6.8 1,3,7,9
4.6 9.6
12 21
3.1 9.3
12 19
6.2 12
6.3 12
51 36
i.'Ols
21
40
28
41
29
31
32
24
1.2.3.7/
1,2,3.8
14
33
17
31
20
20
23
34
TCDU isomer
2.3,7,8
19
39
23
44
31
29
31
31

13C12-2,3,7.8
27
48
31
.38
36
34
36
20

1,2,8,'J
38
64
40
76
61
59
56
26
                                    ''Sample was also spiktMl will] 100 ng ol  Arorlor  1260.

-------
                                              TABLE 7.  RECOVERY (%) Of SEVERAL TCDD 1SOHEKS FROM CLEANUP OPTION  B
cn
cr;
_ _ _ .__ 	 ____ __. 	 ___ 	 , 	 _
Spike
level
l«Lft«n**nt Cjitiocl
1
HjSCU-SIOj
4.0,
SI02
I.Og

Acidic AI}O3
«.0g
1 ng
1 ng
1 ng
10 ng
10 ng
30«L 20% CH2Cl2/H.»on.
10 nga
Corbopol. C/C.IIlt

6nLUIixni
Mean
HRGC/HRMS

% BSD

1,3.6,8
7.6
2.4
6.4
O.'J
1.8
11
5.0
79

1,3.7.9
15
21
25
7.6
11
17
16
40
Recovery
1.2.3.4/ 1
1.4,7,8 1
35
31
34
14
36
44
32
31
(%) of
,2,3,7/
.2,3.8
14
17
22
21
25
31
22
28
TCOD isoiner
2.3,7.8
37
34
40
50
47
59
45
21

»3C12-2,3,7,8
38
40
41
44
48
52
44
12

1,2,8,9
48
52
57
70
59
78
61
19

                            Sample was also  spiked with  100 ng of Aroclor 1260.

-------
                                                      TABLE  8.   RECOVER*  (%)  OF SEVERAL TCDO ISOMEKS FKOH CLEANUP OPTION C
cn
a:
Cc
ro
O

Recovery (%) of TCDO isomer

1 mi ft*fu*m Extract
1
HjS04 - SiOj
4.0g
SIOJ
I.Og
, 1
Concentrate to 0. 1 »L
1
Acidic AljOj
«.08
1 30.1 30%
Canc
-------
                                                   TABLE 9.   RECOVERY (%) OF SEVERAL TCDD ISOMERS FROM CLEANUP  OPTION 0
cn
'


\ ml ft*n>«ni Extract
1
MjSC>4 - SiOj
4.0,
SiOz
1.0,
, 1
Cencinliiili to O.inL
1
Acidic AI2Oj
6.09
1 30ml 30%
Cwfaepol C/C.m.
6 ml Toluc
HRGC/HRMS


Spike
level 1
0.10 ng
0.10 ug
0.10 ngd
1.0 ng
1.0 ng
1.0 ngd
10.0 ng
CHjOzAu.on. 10.0 ng
10.0 ngd
K
Mean
% RSD

,3,6, 8a
147
290
260
90
180
135
81
114
126
1S8
46

l,3,7,9a
197
200
360
93
IBS
158
107
72
138
168
51
Recovery
1,2, 3,4/ 1
1,4,7,8 1
113b
,2l"
110b
106
95
85
50
105
89
97
21
(%) of TCDU
,2,3,7/
,2,3,8 2,3
c
c
c
103
121
79
104
108
95
101
14
isomer
,7,8
72
71
84
85
109
47
101
107
91
85
23

»3Cl2-2,3,7,8
84
90
97
86
90
60
76
82
92
84
13

1,2,8,9
70
71
e
53
108
63
84
118
112
85
29

                                The  1,3,6,8- ;m.1 1,3,7,9-TCOD isomers were also noted  iti reagent blanks from Lhe acidic alumina
                                column.  No such interferences were noted from the acidified silica gel or tlie Carbopak C/Celite
                               .column.
                                Resolution of 1,2,3,4-, 1,2,3,7-/1,2,3,8-, and 1,4,7,8-TCDD was not achieved.  This value
                                represents recovery of the four isomers.
                               ^Recovery reported with 1,2,3,4-/l,4,7,8-TCUU.
                                Sample was also spiked with 100 ng of Aroclor-1260.
                                IIKGC/IIRMS analysis was interrupted prior to  the elution of this isomer.

-------
     Three additional  experiments  were completed to evaluate the efficiency
of reverse elution of the carbon  column.  The  Carbopak C/Celite was placed
in a 5-mL disposable pipette packed at both ends with glass wool plugs.  The
column  was eluted in one  direction  for the hexane,  cyclohexane/methylene
chloride,  and the  methylene  chloride/methanol/benzene  mixture.   The column
was then  turned over and eluted with  6  ml  toluene.   Triplicate analyses at
the 0.10  ng/TCDD isomer  spike  level  demonstrated average  recoveries  of 98
percent  for  1,3,6,8-TCDD;  91 percent  for   1,3,7,9-TCDD;  104 percent  for
1,2,3,4-,  1,2,3,7-,  1,2.3,8-,  and 1,4,7,8-TCDD;  116 percent  for  2,3,7,8-
TCDD;  102  percent for  13C12-2,3,7,8-TCDD; and  93  percent for 1,2,8,9-TCDD.

FINAL METHOD EVALUATION

     Based on the  results  of the  column  evaluation study,  the analytical
method was revised  to specify the  cleanup procedure presented as Option D in
Figure 2.  The  final protocol,  as presented  in  Appendix  A, was then evalu-
ated as described below.

     The data presented in Tables  10 through 17 are summaries of the initial
column calibration,  HRGC  and HRMS resolution checks, and the results of the
sample analysis.

Calibration

     Table  10  summarizes  the  RRF data  for the  concentration calibration
standards from  the  initial calibration and the routine monitoring of the RRF
values over  the time required to  complete the  sample analyses.  The RRF(I)
as specified  in  the protocol is a measure of  the  response of 2,3,7,8-TCDD
versus the internal  standard, l3C12-2,3,7,8-TCDD.  The value for RRF(I) var-
ied ±9.4 percent  over the five concentration levels of 2,3,7,8-TCDD ranging
from 2.5 pg/Ml-  to 40 pg/pL.   The RRF(II) is  used  to calculate the absolute
recovery  of   the internal  standard as  compared to  the  recovery  standard
13C12-1,2,3,4-TCDD.   The  average  RRF(II)  was determined  to  vary by ± 19.3
percent  over the  calibration  curve.   The  variability  of  the  RRF (I)  and
RKF(II)  were  determined  to be  less  than  ± 10 percent  and  ± 18 percent,
respectively, over  all data points required to complete the sample analysis.

     In addition to the analysis  of calibration standards  specified in the
protocol,  solution HRCC6  was analyzed in triplicate  to determine the lower
limit of  sensitivity (1 pg/|jL).   Although the calculated RRF(I) and RRF(II)
values and the  S/N are within the specified criteria, the ion ratio for the
native compound and recovery standard indicate that these measurements fall
outside the acceptable calibration window.
                                      22
                                          r" r^ i\
                                          u f(J

-------
                                                                     TABLE 10.  INITIAL CALIBRATION SUMMARY
ro
CO

Calibration
standard
IIRCC1
HRCCI
HRCC1
1WCC2
HRCC2
1IRCC2
HRCC2
HRCC3
I1RCC3
HRCC3
HRCC4
HRCC4
HRCC4
HRCC5
HRCC5
HRCC5
Date
09/12/8S
09/12/85
09/12/85
09/12/85
09/12/85
09/12/85
09/12/85
09/13/85
09/13/85
09/13/85
09/13/85
09/13/85
09/13/85
09/13/85
09/13/85
09/13/85
Time
09:05
09:44
12:31
13:00
13:27
13:53
15:39
10:31
10:57
11:23
13:02
13:29
13:56
14:22
14:49
15:15
m/z 320/322
0.82
0.84
0.84
0.73
0.80
0.92
0.78
0.78
0.78
0.73
0.77
0.73
0.77
0.78
0.7S
0.76
ra/z 332/334 (IS)
0.77
0.80
0.73
0.77
0.73
0.66
0.77
0.79
0.80
0.83
0.80
0.76
0.78
0.80
0.82
0.78
m/z 332/32
0.80
0.73
0.87
0.83
0.81
0.69
0.73
0.79
0.78
0.78
0.76
0.78
0.76
0.78
0.83
0.79
          IIRCC6
          HRCC6
          HRCC6

          HRCCI
          HRCCI

          HRCC2
          HRCC2
          IIRCC2
          IIRCC2
          HRCC2
          IIRCC2
          HRCC2
          HRCC2
          HRCC2
09/16/85
09/16/85
09/16/R5

09/16/85
09/16/85

09/20/85
09/23/85
09/23/85
09/24/85
09/25/85
09/26/85
09/27/85
09/30/85
10/03/85
10:43
11:19
13:44

12:42
14:40

10:46
08:47
10:46
10:47
08:39
08:56
09:33
09:19
08:53
1.32
1.18
0.86

0.87
0.83

0.86
0.82
0.88
0.89
0.7C
0.77
0.78
0.83
0.68
0.71
0.83
0.80

0.83
0.79
0.73
0.80
0.80
0:69
0.78
0.80
0.84
0.80
0.75
0.84
1.14
1.01

0.75
0.85
0.80
0.70
0.83
0.69
0.89
0.72
0.81
0.85
0.81
(RS) S/N 259 S/N 322 S/N 334(15)
41:1 > 65:1 > 65:1
48:1 > 65:1 > 65:1
24:1 > 65:1 > 65:1


36:1 > 65:1 > 65:1
58:1 > 65:1 > 65:1
78:1 > 65:1 > 65:1
76:1 > 65:1 > 65:1


97:1 > 65:1 > 65:1
111:1 > 65:1 > 65:1
110:1 > 65:1 > 65:1


96:1 > 65:1 > 65:1
> 144:1 > 65:1 > 65:1
> 144:1 > 65:1 > 65:1


> 144:1 > 65:1 > 65:1
> 144:1 > 65:1 > 65:1
> 144.1 > 65:1 > 65:1


Overall Hean

10: 21: > 65:1
9.6: 25: > 65:1
18: 30: > 65:1
36: 63: > 65:1
48: > 63: > 65:1
> 75: > 63: > 63:1
> 75: > 63: > 63:1
42: > 63: > 63:1
26: > 63: > 63:1
58: > 63: > 63:1
73: > 63: > 63:1
49: > 63: > 63:1
73: > 63: > 63:1
29:1 > 63:1 > 63:1



Mean
X RSD




Mean
X RSD



Mean
X RSD



Mean
X RSD



Hean
X RSD
(RRF)
X USD














RRF(l)
0.783
0.794
0.750
0.776
2.9X
0.829
0.853
0.7993
0.861
0.848
2. OX
0.974
0.965
0.972
0.970
0.5X
0.945
0.935
0.967
0.949
1-7X
0.964
1.01
0.989
0.987
2.2X
0.906
9.4X
0.917
0.878
0.935
0.876
0.850
0.835
0.832
0.941
1.01
0.949
0.941
1.04
0.854
0.955
RRF(II)
2.18
2.27
2.28
2.24
2.4X
1.76
1.93
2.033
2.27
1.99
13. OX
1.53
1.57
1.58
1.56
1.9X
1.49
1.52
1.49
1.50
1.1X
1.52
1.42
1.43
1.46
3.9X
1.75
19. 3X
1.44
1.09
1.58
1.86
2.11
1.98
2.65b
1.41
1.68
1.96
1.57
1.68
1.67
1.53
          .Nol  included in mean RRF computation.
           Not  within  allowable limits  for routine calibration.

-------
                   TABLE 11.  HRGC AND MASS RESOLUTION CHECK SUMMARY
Date
9/12/85
9/12/85
9/12/85
9/12/85
9/13/85
9/13/85
9/13/85
9/13/85
9/16/85
9/16/85
9/16/85
9/16/85
9/17/85
9/17/85
9/18/85
9/18/85
9/19/85
9/19/85
9/20/85
9/20/85
9/20/85
9/20/85
9/23/85
9/23/85
9/23/85
9/23/85
9/24/85
9/24/85
9/24/85
9/24/85
last.
ID
MS50
HS50
MSSO
MS50
MS50
MSSO
MS50
MS50
HSSO
MS50
MS50
MS50
MS50
MS50
MS50
MS50
MSSO
MS50
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
Sol.
ID
.
PC
PC
-
-
PC
PC
-
-
PC
PC
-
PC
-
-
PC
-
PC
-
PC
PC
-
- '
PC
PC
-
PC
PC

Time
07:51
08:32
16:05
16:47
08:26
08:45
15:48
16:23
09:25
10:45
15:16
15:57
10:26
10:14
07:58
08:18
11:14
12:56
08:00
08:16
15:44
16:14
07:55
08:15
16:01
16:45
09:51
10:15
16:01
16:33
TCDD isomer
File resolution
name (% valley)
MID254I12X1
8367I12XQ1 5.9
8367I12XQ9 2.9
MID254I12X2
MID254I13X1
8367I13XQ1 6.9
83671 13XQ12 11.4
MID254I13X2
MID254I16X1
8367I16XQ1 11.9
83671 16XQ8 23.0
MID254I16X2
8367I17XQ1 13.3
MID254I17X1
MID254I18X1
83671 18XQ1 20
MID254I19X1
8367I19XQ1 3.5
MID254I20X1
8367I20XQ1 6.7
8367I20XQ5 4.1
MID254I20X3
MID254I23X1
8367I23XQ1 8.8
8367I23XQ6 12.5
Manual check3
HID254I24X1
8367I24XQ1 12.2
8367I24XQ3 13.1
HID254I24X2
Mass
resolution
at 10% valley
10,774
-
-
10,450
10,230
-
-
10,384
10,294
-
-
10,388
-
•10,824
11,019
-
11,679
-
12,068
-
-
10,777
10,096
-
12,500
10,374
-
-
10,567
Mass
measurement
error
5 ppm
-
-
-
0 ppm
-
-
-
4 ppm
-
-
- .
-
2 ppm
1 ppm
-
4 ppm
-
3 ppm
-
-
-
1 ppm
-
~
3 ppm
-
-

A manual resolution check was performed due to data system failure.
                                      (continued)
                                           24

-------
                                TABLE 11.   (continued)
Date
9/25/85
9/25/85
9/25/85
9/25/85
9/26/85
9/26/85
9/26/85
9/26/85
9/27/85
9/27/85
9/27/85
9/27/85
9/30/85
9/30/85
9/30/85
9/30/85
10/3/85
10/3/85
10/3/85
10/3/85
Inst.
ID
MSSO
MS50
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
MSSO
Sol.
ID
-
PC
PC
-
-
PC
PC
-
-
PC
PC
-
-
PC
PC
-
-
PC
PC
'
Time
07:50
08:05
16:13
16:45
08:07
08:21
15:49
16:21
08:19
09:02
16:01
16:29
08:15
08:33
15:10
15:41
07:59
08:20
15:56
16:29
File
name
MID254I25X1
8367I25XQ1
8367I25XQ3
MID254I25X2
MID254I26X1
8367I26XQ1
8367I26XQ3
MID254I26X2
MID254I27X1
8367I27XQ1
8367I27XQ3
MID254I27X2
MID254I30X1
8367I30XQ1
8367I30XQ3
MID254I30X2
MID254J03X1
8367J03XQ1
8367J03XQ3
MID254J03X2
TCDD isoraer
resolution
(% valley)
-
11.1
6.5
-
-
8.3
13.2
-
-
11.9
11.8
-
-
< 25
< 25
-
-
17
12
"
Mass
resolution
at 10% valley
11,165
-
-
11,419
10,989
-
-
10,499
11,564
-
-
10,639
11,149
-
-
11,321
10,567
-
-
10,442
Mass
measurement
error
0 ppm
-
-
-
3 ppm
-
-
-
1 ppm
-'
-
-
5 ppm
-
-
-
0 ppm
-
-
'
A manual resolution check was performed due to data system failure.
                                           25

-------
                                                                                         TABU II.  TCDD DATA RWORT n»l
                                                                                               tactymuia AMI..?
ro
. Aliauot
Settle Air-dry wt. (|)
•o. or Vol. (L)
IM7-83-I576I-DW 1.0 L

1367-82-13761-DV 1.0 L
1367-19-13761-nM 10 L



1167-11-15761-tW 1.0 I

oM7-90-137oI-[IM 1.0 L







1367-05-I576X-IH) 1.0 L
1367-92-I576X-IWW 1.0 L



1367-14-1576X-DWN 1 0 L






?Aqueoue aeaple data reported ae pp<
'Criteria for poaltlve Identlflcatl
laoaer rould not fae identified.


ran
lao*er
2,3,7,1-
1,1,6,1-
1,1.7,9-
2.1.7,1-
1.1.6,1-
1.1.7,9-
c
2.1.7.1-
c
1,1.7.1-
1,3,6,8-
1,3,7,9-
,2,3,7/
.2.3,8-
,1,1,4-
.1.7,1-
,2.1,9-
1.3,7.1-
1,1,7,1-
1,3,6,1-
1,1,7,9-
1.1,7,1-
1,1.7,1-
1.3,6.8-
1, ,7,9-
1. ,3.7/
l! !l|4-
1. ,7,1-
1.1,1.9-
and aoil
o require



Retentio
TCM) "C|
11:15
16:4}
17:54
21:40
16:41
17:51
24:10
21:29
23:37
11:59
17:02
11:14
22:11

22:31
14:30
30:01
21:55
22:01
17:06
11:16
24:31
24:31
17:06
11:11
21:21
22:15
24:12
30:02
aaeople data
that the lo


TCDD (ppt
n tisje or peq) listr.
1-2,3,7,8 Heal. Dl ID ' Date Ti«e
21:14 221 - IG50 09/20/15 11:27
- 117
16 5 -
21:17 2,277 - HS50 09/20/15 14:41
134
212
137
21:27 1.090 - MS50 09/20/15 15:11
75.9 -
21:56 1,010 - IG50 09/23/15 11:17
502
766 -
1,160

1,140
1,430
1.330
11:55 1.290 - IG50 09/23/15 12:51
21:59 501 - 1650 09/23/13 13:29
191
201
55.2 -
'516
512
395
403
616
140
321
presented aa ppt.
l ratio! fall between 0.67 and 0.90.

(continued)

Relative loa Abundance Ralloab
320/321 111/114(1!) 111/114(RS) 1 Rec.
0.71 0.63 0.61 40
O.i9
0.10
0.51 0.69 0.72 14
0.79 0.72 0.71 96
1.02
0.14
0.90
0.77 0.71 0.71 61
0.19
0.74 0.72 0.12 91
0.15
0.13
0.11

0.71
0.71
0.10
0.11 0.90 0.11 23
0.74 0.10 0.71 75
0.15
0.75
0.19
0.71
0.10
0.74
0.72
0.72
0.10
0.79





•It 259
73:1
64:1
41:1
21:1
10:1
11:1
6:1
146:1
7:1
27:1
22:1
30:1
54:1

45:1
71:1
22:1
37:1
49:1
25:1
27:1
1:1
49: 1
21:1
23' 1
73:1
53:1
42:1
61.1
70:1
23:1




5/1
•/t 32} Wl 334(IS) Comenta
>63:l >49:l Ratio unacceptable. Rerun.
50:1
10:1
> 63:1 > 63:1 Saaa>le aplked at twice requested level.
> 3:1 Rerun.
> 11:1
63:1
> 63:1 > 63:1
11:1
34:1 > 63:1
11:1
33:1
> 61:1

60:1
> 63:1
11:1
63:1 10:1 Low recovery.
63:1 > 63:1
19:1
29:1
16:1
63:1 23:1 Low recovery.
31:1
> 63: 1
63:1
50:1
45:1
55:1
63:1
20:1




-------
              ro
01
^,7
Cl


Saajple A
Ho
8367-9I-I576X-IW



8367-66-15761-iHw
8367-93-I576X-IMI







8367-70- 15 J6l-HI»







8367-65- I576K-15

8367-66-I576X-85D

B367-67-lS76X-»S»







'Aqueous aaav>le data


Aliquot
it-dry «t. (|) ICDO
or Vol. (L) IIOK
1.0 I 2.3.7,
1.3,6,
1.3.7,
1,2.7,
1.0 L 2,3,7,
1.0 L 2,3.7,
1.3.6.
1.3,7.
1.2.3,
1.2,3,
l.J.3.
1.2,7,
1,2,1.
10.01 | 2,3,7,
1.3.6.
1.3.'.
1,2.3,
1,2,3,
1.2.3,
1.2,'.
1.2.1,
10.00 | 2,3,7,
1.3,7,
9 15 i 2,3,7,
1.3,7.
10.00 | 2.3,7.
1.3,6.
1.3,7,
1.2.3,
1.2.3.
1,2,3,
1.2.7,
1 .2.8,



Reteo
TCDD '
- 22:00
• 17:03
- 18:14
- 24:10
- 21:50
- 21:39
- 16:47
- 17:57
/ 21:58

- 21:11
- 21:01
- 29:35
- 21:38
- 16:45
- 17:55
/ 21:58

- 22:10
- 24:09
- 29:36
- 2I:*7
- 18:05
- 21-44
- 18:03
- 21:47
- 16:55
- 18:05
/ 22:07

- 22:19
- 24:16
- 29:45

TABLE 12. I
HRGC/IOUIS
tCDO (ppt
tioo lite or fpqj Initr.
ICll-2,3.7.8 Mea>. DL ID Dale Tix
21:57 534 - MSSO 09/23/85 15:01
246
122
54.7 -
21:48 1,430 - HSSO 09/23/85 15:31
Jl:38 530 - I1S50 09/24/85 11:19
582
690
940

1,180
1 . 790
691
21:37 30.3 - ItSSO 09/24/65 12:55
29.0 -
Sl.l -
I2S

118
252
100
21:45 18.2 - IB50 09/24/85 13:27
8.5 -
21-43 15.1 - HS50 09/24/85 13:56
4.2 -
21:46 12.9 - KSSO 09/24/85 14:29
NO 9.2
15.2 -
61.8 -

54.1 -
147
63.9 -
ion ralloa f«ll between 0.67 and 0.90.
continued)
Analyiii
Relative Ion Abundance Ratio!
320/322 332/334(15) 332/334OS) I Kec.
0.87 0.83 0.81 77
0.71
0.78
0.70
0.80 0.76 0.79 29
0.76 0.72 0.73 71
0.82
0.80
0.72

0.79
0.82
0.77
0 85 0.78 0.77 56
0.63
0.87
0.61

0.69
0.61
0.85
0.87 0.78 0.75 73
0.85
0.67 080 0.72 85
0.87
0.66 0.73 0.71 46
0.59
0.67
0.77

0.90
0.75
0.83




•/z 2S9
72: 1
47:1
36:1
7:1
49:1
23:1
42:1
43:1
49:1

50:1
72:1
22:1
12:1
15:1
24:1
46:1

38:1
73:1
20.1
12:1
9.4:1
18:1
6.3:1
9:1
11:1
14:1
4R:1

35: 1
73:1
26:1


- - 	 	
S/«
•/z 322 ml I 334(15) Coraeoli
> 62:1 > 63:1
48:1
32:1
16:1
63:1 > 33:1 Lou recovery.
4:1 > 63:1
4 :l
6 :l
4 :1

6 :l
6 :l
2 :!
15:1 > 63:1
26:1
33:1
63:1

58:1
> 63:1
23:1
42.1 > 63:1
25:1
11:1 > 63 : 1
12:1
16:1 > 63:1
15: 1 Ratio unacceptable.
20:1
63:1

58:1
63:1
22:1

                            :Uo.»er could not be identified

-------
                                                                                                          TMU II.  (continued)
Saiple
He.
1167-6I-IS76»»I

I167-69-IS76I-IID


l]67-7l-IS76l-«l



I167-72-IS76.-IID


IM7-7]-IS76i-ll»








1167-74-IS76I-H1





1367-77-IS76.-FA



Aliquot
Air-dry wt. (() TCDD
or Vol. .9-
c
1.2,7,1-
e
9 94 | i,),7,l-
1.1.6.1-
1.1.7,9-

Rcteotion tiae
TCDO 13C|i-2,3,7
2I:«2 2I-.40
16;4»
11:11 21:11
16:4!
17:52
21:17 2I:1S
16:44
17:54
19:04
21:1} 21:1]
16:42
17:53
21:41 21:41
I6:4«
11:59
19:09
22:01

22: IS
24:12
29:41
21:44 21:42
16:49
11:00
19:10
24:59
26:01
21:16 21:14
16:42
I7:S1
19:02
ICDD (ppt
or PM) ln»tt
1 He.i Ot 10
14.) - HSSO
4.S -
M.6 - »SO
5.1 -
10.0 •
117 - n»o
160
1IJ
SO. 6 -
71} - 1050
201 -
101
1,210 • IBSO
111
435
51
Sll

695
1,170
461
2.020 - IBSO
164
217
70.6 -
11.7 -
27.1 -
1.720 - KSV>
I.MO
1,150
1,250
ReUlivt lor, Abundance Ritioi^
Dit< Tl» 120/122 132/334(13) 112/114(13) X Ice.
Of/24/15 15:02 0.12 0.67 0.75 7]
O.t7
09/24/15 15:12 0.70 0.74 0.11 46
0.69
0.11
09/2S/IS 10:01 0.11 0.11 0.11 95
O.IS
0.14
0.69
09/25/15 10:10 0.71 0.15 0.10 75
0.11
0.12
0,65
09/25/15 11:!7 0.77 0.11 0.11 10
0.77
O.IS
0.11
0.72

0.71
0.12
0.11
09/25/15 11:00 0.11 0.11 0.11 79
0.16
0.11
0.74
0.92
0.61
09/25/15 11:10 0.10 0.12 0.12 4
0.11
0.11
0.10
SIX
i7Tl59 •/! 122 mil liUTsT CoKntl
29:
II:
19:
II:
15:
97:
14:
44:
7.1:
71:
16:
40:
6.S:
97:
47:
79:
6.7:
57:

SI:
17:
21:
> 145:
22:
24:
9:
2.5:
1:
67:
109:
94:
60.
61: > 61:1
14:
31.
5:
10:
61:
14:
24:
4:
63:
22:
21:

21:
15:
4:
27:

29:
61:
29:
> 63:
>
> 1 :
> ;


21:
31:
14:
22:

> 54:1


> 63:1



> 61:1


> 63:1








> 61:1





6: 1 Low recovery.
.


r\>
00
                      oui  tuple d«t» reported •• ppq a


                      er could not be Identified.
                                                          oil aaa^le data  presented  at  ppt.
:>een 0.67 and 0.90



 (continued)

-------
cn
•vl
               ro
               10

Aliquot
Stifle Air-dry « . (|) ICDD

1367-77-1576X-FA c
(concluded) c
c
I.2.3.7/
1,2,3,1-
1.2,3.4-
c
1,2,7.1-
c
c
c
c
c
1367-94-I576X-H2V 1.0 L 2,3,7,8-

1.3.6,8-
1,3,7,9-
c
c
c
1.2.7,1-
c
c
c
c
•367-13- 1576X-OM1 1.0 L 2,3.7, -
1,3,6, -
1,3,7. -
c
1.3,6. -
1.3,7. -
^Aqueoue •••pie d«ta reported •• ppq ind toil

Retention 11

20:00
20.13
21:25
21:5*

22:0>
22:35
24:05
24: 49
25:23
25:55
27:11
28:07
21:42 21:

16:47
17.51
19:01
20:06
20:49
24:13
24:55
26:01
27:25
21:11
21:11 21:
16:25
17:14
21:39
It 22
17:12
•Japlr dill pre

TCDD (ppt
•e or ppq)

1,220
27*
109
675

4,660
179
720
3,460
155
3,430
441
2,920
41 »C

IK
K
1C
K
K
»C
«C
•C
DC
HC
10 265
m 167
125
50.1 -
140
106
looted <| ppt.
TABU 12. (continued)
Instr. ReUtive loo Abundto.ce fUtloi

HS50 09/25/15 13:30 0.14
0.75
0.77
0.10

0.11
0.18
0.11
0.10
0.75
0.71
0.79
0.61
IIS50 09/25/15 15:11 0.16 1.3 0.12 ND

0.11
0.14
0.79
0.98
0.68
0.75
0.80
0.72
0.82
0.81
HS50 09/26/85 09:56 0.79 0.73 0.70 42
0.91
0.71
0.71
0.81
0.96
0.67 >nd 0.90.



44
16
6
32

146
31
23
80
7
16
14
68
> 154:

IB:
12:
20
6
}
3
17
63
9
30
11
11:
14:
1.2:
6.3:
3.»


S/

1 :
:
2. :
1 :

6 :
1 :
1 :
5 :

6: :

4 :
> 65:

55:
40:
61:
20:
11:
3:
35:
63:
8:
25:
31:
24:
21:
7:
12-
11:

















4:1 332/334(13) Ratio unacceptable; no jeount
co«putationa performed.

-






-

47:1
Ratio unacceptable.




                               *r could not be identified.

-------
                    CJ
                    o
en


Aliquot
Saaple Air-dry vt. (|)
•o. or Vol. (t)
I16I-U-IS76X-IW 1.0 I



I167-IS-IS7II-I2H) 1.0 I















I16I-M-IS7U-UVI 1.0 L


















TCDO Retention tieje
leoe*r ICDO "C, ,-2, 1,7.1
.1,1,1- 21:11 21:15
.M.I- 16:11
.1,7.1- I7:M
.1.7,1- 11:44
,1.1,1- 21:11 11:11

.1,6.1- I6:2S
.1,1.1- 17:35
11:41
11:31
20:10
11:11
,1.1,1- 21:11
24:21
25:21
U:4I
21:11
21:14
10:11
11:01
.1.1.1- 11:14 11:11

,3,8,1- 16:16
,1,7,1- 17:15

19:41
10:12
11:11
,1,7,1- 11:40
24:14
15:30
16:51
27:31

10:14
11:11



Tn»
«... Dl
,010
III
111
III
K

•C
K
K
K
K
K
K
K
K
K
K
1C
K
1C
•C

K
K
K
K
K
K
K
K
•C
•C
•c
K
•C
K

TAIL! 11. (continued)
•KGC/MmS Aoalnla
laatr. lelatlvt Inn Abundance latloi*
ID Dale Tle» 110/111 1)2/114(15) 1)1/1)4(13) 1 dec.
IBSO 01/16/IS 12:32 0.76 0.71 O.IS 66
0.11
0.41
0 76
IO50 01/16/15 11:0} O.I) I).) 0.67 1C

0.17
0.11
0.11
0.17
0.71
014
0.70
0.76
0.76
0.11
0.11
0.11
a.n
O.SI
IBSO 01/16/IS 14:21 0.11 1.47 0.17 K

.71
.71
.71
.11
.61
.S|
.11
.71
.75
.76
.70
.77
.14
.11




•/a 2S
11:
11:
II:
6:
> 146:

74:
41:
46:
IS:
1:
S:
S:
S4:
71:
II:
16:
7:
10:
4:
> I4S:

40:
11:
14:
II:
1:
4:
S:
SI:
11:
II:
11:
1:
11:
1:



s
fit 32
> 61:
> 10:
> IS:
11:
> 6):

> 26:
> 14:
> 11:
> S:
> 4:
> 1:
l.S:
19:
6):
1:
IS:
S:
IS:
l.S:
> 61:

11:
IS:
11:
6:
S:
6:
1:
41:
61:
10:
21:
6:
20:
25:



n
•/a 134(11) Coeatenta
> 63:1
.
.
•
11:1 111/1)4(1$) latin unacceptable) nn Mount
conputationa perfortwd.

.
.
-
-
•

•
•
.
-
.
.

1:1 111/1)4(18) Ratio unacceptable; no aojoojt
coatiuletiou perforaN*.
-
-
-
•
•atio unacceptable.
Katie unacceptable.
-

-
•
-
.
-
Ratio unacceptable.

                                     Criteria  for p«alti*e identification require that the  ion  ratioa (all between 0.67 and 0.10.
                                     laoe»r coeild not be ideal 11 led


                                                                                                            (continued)

-------
C1
-v!


Aliquot
Sample Air-drj M. (|) TCDD
Ho. or Vol. (1) lao<*r
8367-75-I576X-H3D 1.16 |



B367-76-I576-X-H3K 1.14 |







8367-78- I576X-FAD 10.04 |















8367-99-1576X-FA1C 9.93 |








,3,7,1-
.3.6.1-
.3,7,9-

.3.7,1-
.3,6.8-
,3.7.9-
,2.3,7/
.2,3,8-
.2,3,4-
,2,7,8-
.2,1,9-
.3,7,1-
,3,6.8-
,3,7,9-



,2.3.'/
,2,1.1-
,2.1.4-







.3.7.8-
.3.6,8-
,3,7,9-



,2.3,7/
,2,3.1-
.2,3.4-
a
Criteria for poaitive identification require
laoKr could not be identified.



Retention time
TCDD "C,,-2,3.7
21:20 21:19
16:31
17:41
11:49
21:39 21:39
16:45
17:56
21:51

22:11
24:10
29:36
21:33 21:30
16:41
17:51
19:01
19:57
20:41
21:52

22:05
22:33
24:01
24:45
25:51
27:14
28:03
30:01
21:39 21:31
16:45
17:56
19:06
20:01
20:47
2I:S8

22:11

that the ion ratios


TABLE
KRG1
TCDD (ppt
" PPQ) ln«tr.
8 Heaa. DL ID Date
2,260 - HS50 09/27/85
116
163

1,800 - HS50 09/30/85
383
367
825

B55
2,310
952
1,020 - HS50 09/30/85
926
747
610
557
146
286

2,260
KD 329
356
3,520
3,680
558
3,120
3.270
1,160 - HSSO 09/30/85
1.390
1.160
881
818
194
423

3.620

fall between 0.67 and 0.90.

1 continued)
12. (continued)
fTJjjjjjS Analyaia 	
Relative Ion Abundance Ratioafa
Tile 120/322 312/334(15) 332/33KHS)
15:31 0.79 0.88 0.80
0.90
0.67

09:57 0.80 0.79 0.80
0.80
O.BI
O.BI

0.13
0.15
0.77
10:29 0.73 0.71 O.M
0.77
0.80
0.17
0.19
0.71
0.67

0.79
0.91
0.74
0.71
0.79
0.11
0.76
0.79
13:00 0.80 0.84 0.88
0.78
0.79
0.85
0.74
0.76
0.80

0.86




- ... ... . 	 . -- . .- ...
,» / 25, ,'tf 1 334..S) C
99 > 145: • 63: > 63:1
14:
20
6
86 > 143
44
41
79

60
145
46
6.7 45
55
42
32
23
8.7
17

75
16
13
92
97
17
80
83
5 56:
94:
80.
60:
35:
13:
29:
131 :
32-
' 5:
> 7:
> 2.5
> 63:
> 15:
> 17:
> 29:

•• 24:
63:
23:
32:
37:
25:
22:
14:
6:
10:

63:
11:
8:
56:
63:
10:
48:
47:
25:
40:
Ji-
ll:
17:
5:
10:
r,3:
10:


1
> 63:1







9:1 Recovery low.








Ratio unaccrptatble.






6: 1 Low recovery.












-------
                                                                                                        TAIU 12.  (concluded)
10
SMVlo Air-dry «*. (|)
•o. or Vol. (it
TCDO (t.l.otioo tin
lico*r TCOO "C, i-l.l, 7
H67-tt-l376I-r«l 22:M
(coocltJed)






6M7-IOO-I376I-OM> 1.0 L


•167-I02-I376I-IB 300 mi
6167-IOI-I376I-DM) 1.0 ml


B167-I01-I376X-IB) 300 ml


I167-I04-I}76I-UH 4N mi







•M7-I03-I376I-IW 610 mi







,1,7.1- 16:01
14:32
13:31
17:10
26:10
10:09
M:67
,1,7,1- 21:04 21:01
,1.6,6- 16: It
. 1.1,1- 17:26
,1,7,6- 21:02 21:01
,1,7,1- 20:34 20:33
,1,6,1- 16:14
,1.7,t- 17:21
. 1,7,6- 20:37 20:36
.1,6.1- 16:17
,l,J,t- 17:21
.1,7,1- 21:00 20:St
,1,6,1- 16:11
,1,7.1- 17:27
11:12
24:06
23:09
27:16
29:6*
,1,7,1- 20:36 20:36
.1,6,6- 16:13
,1.7,9- 17:22
11:10
24:0)
23:07
24:17
2t:43
or ppo.) loitr.
i He... 01 10
561 • IBIO
316
4,110
4.SM
n> 611
1,960
4,010
1,170
246 - IBM
617
tn
604 - IBM
1,030 - IBM
137
1*4
611 - IBM
D 41
17
17,100 - IBM
» 71
164 -
Ml -
ID 417
311 •
S7J •
111 -
21,100 - IB30
214 -
102 -
451 -
364 -
710 -
318
147 -
«eUti» loo Abiudioco lotloi*
0>t< Tl«e 120/121 111/114(18) 111/114(19)
09/10/13 11:00 .11
.62
.73
.71
.1$
.77
.86
•"
10/01/63 11:12 0.66 0.62 0.74
0.77
O.M
10/01/43 12:31 0.71 0.17 0.13
10/01/13 11:21 0.61 0.11 O.N
0.11

10/01/13 14:01 0.71 0.73 0.71
0.51
0.69
10/01/1] 14:11 0.74 0.14 0.74
1 OS
O.M
0.13
0.63
0.70 .
0.11
O.M
10/01/1} 11:11 O.M 0.14 0.71
0.13
O.M
0.71
0.60
0.11
0.10
0.79
S/N
I Ice. •/• 251 mil 122 •/• 114(18) CooBenti
1 12: 10:
20:
117:
146:
10:
14:
121:
26:
61.3 21:
71:
SI:
39.S 16:
6O 71:
17:
26:
37 11:
6:
I:
71 > 143:
9:
16:
»:
It:
12:
11:
II:
96 > 143:
16:
17:
10:
20:
11:
11:
10:
34:
61:
7:
46:
44:
16:
18:
61:
47:
11:
61:
11:
26:
42:
11:
17:
> 61:
II:
11:
S3:
21:
12:
26:
21:
> 61:
19:
21:
H:
14:
11:
14:
11: 10:



Kitio unacceptable.



> 61:1


> 61:1
> 61:1


> 61:1


> 61:1
title wucccptoblt.


•jtlo u»«rcepteble.



> 61:1






	 ___ 	 . 	
               .Aooeove tmmflt mill reported •• ppo,  tod loll •••pie dil« prevented ••  ppt.
                Criteria for polillee lorntiticitloe require tbil the ioo r.Uoi fill  beMea 0.67 tod 0.10.
                luo*r could >ot be Idratllled.

-------
                                                   TABLE 13.
                                                              ACCURACY AMD PRECISION OF THE  HRGC/HRMS ANALYSIS FOR 2,3,7,8-TCDD
                                                                       FROM LABORATORY AQUEOUS MATRIX SPIKES
cn
GO
            co
            CO
2,3,7,8-TCDD 2,3,7,8-TCDD
Sample matrix Spike level (ppq) Detected (ppq)
Distilled water (DW)



Effluent wastewater (EWW)




Influent wastewater (IWW)




Industrial wastewater (IND)




Industrial wastewater (IND)


Soil extract (H2W)



250
250
250

1,000
1,000
1,000


500 .
500
500


500
500
500


.
-

_
.





Average cone.
RPR



Average cone.
RPR



Average cone.
RPR



Average cone.
RPR


Average cone.
RPD


Average cone.
RPD
234
265
246
248
12.5
1,090, 1,030
1,010
1,050
1,050
7.6
534
508
530
524
5.0
1,290
1,520
1,430
1,410
16
604
628
616
3.9
27,100
28,100
27,600
3.6
2,3,7,8-TCDD
Recovery (1)
93.6
106
103
Average rec. 101
RPR 9.3
109, 103
101
105
Average rec. 105
RPR 7.6
107
102
106
Average rec. 105
RPR 4.8
258
304
286
Average rec. 283
RPD 16
-
-

.
-


I3C,2-2
Absolute



Average rec.
RPR



Average rec.
RPR



Average rec.
RPR



Average rec.
RPR


Average rec.
RPD


Average rec.
RPD
,3,7,8-TCDD
recovery (%)
82
42
69
64
63
61, 66
91
80
75
40
77
75
71
74
8.1
23
20
29
24
38
60
57
58
5.2
78
96
87
25
                         "Relative percent range (calculated from the difference of the high  and  low values divided by the average of all  values  and
                         .multiplied by 100 percent).
                          Relative percent difference.

-------
                                                       TABLE 14.  PRECISION OF THE IIRGC/HRMS ANALYSIS FOR 2,3,7,8-TCDD
                                                                         OF SOIL AND FLY ASH SAMPLES
C-7
          U)

Sample matrix
B25-Piazza Road (B5)



Hyde Park 001 (HI)




B52-Shenandoah (Bl)




Hyde Park 003 (H3)




Fly ash - RRAI




Endogenous
2,3,7,8-TCDD 2,3,7,8-TCDD
level (ppt)a Detected (ppt)
50


Average cone.
RPR6
70


Average cone.
RPR
360


Average cone.
RPR
1,700


'Average cone.
RPR
.


Average cone.
RPR
18.2
15.1
12.9
15.4
36
34.3
36.6
30.3
33.7
19
937
785
1,280
1,000
50
2,020
2,260
1,800
2,030
23
1,720
1,020
1,160
1,300
54
13C,2-2,3,7,8-TCDD
Absolute recovery (X)



Average rec.
RPR



Average rec.
RPR



Average rec.
RPR



Average rec.
RPR



Average rec.
RPR
73
85
48
69
54
73
46
56
58
47
95
75
80
83
24
79
99
86
88
23
4
7
5
5.3
57

                                      Estimated level of endogenous 2,3,7,8-TCDD reported to MRI by Dr.  W.  Beckert in letters dated
                                     bApril 19, 1985, and August 30, 1985.
                                      Relative percent range (calculated from the difference of the high and low values, divided by
                                      the average of all values, and multiplied by 100 percent.

-------
                              TABLE 15.  ACCURACY OF THE HRGC/HRMS METHOD FOR THE DETERMINATION OF TCDD 1SOHEKS SPIKED INTO AQUEOUS MATRICES



                                          Effluent wastewater             Distilled vater             Influent  wastewater          Industrial wastewater
TCDD
analyte
1.3.6.8
1.3.7.9
1.2,3,7/1,2.3,8
1,2.3,4
1,2.7,8
1,2,8,9
2,3,7,8
l3CI2-2,3.7,8
Spike
(pg)
1,840
840
1,680
2,440
3,080
1,200
1,000
SOO
Measured
(Pg)
502
766
1,860
1,840
3,430
1,330
1,010
455
Recovery
(X)
27
91
110
75
111
111
101
91
Spike
(pg)
460
210
420
610
770
300
250
SOO
Measured
(Pg)
512
395
403
616
840
328
234
410
Recovery
(X)
111
190
96
101
no
110
94
82
Spike
(PR)
920
420
840
1,220
1,540
600
500
500
Measured
(Pg)
582
690
940
1,180
1,790
691
530
355
Recovery
(t)
63
164
112
97
116
115
106
71
Spike Measured
(pg) (PR)
920
420
840
1,220
1,540
600
500
SOO
HD«
ND
NO
ND
586
ND
904b
100
Recovery
(t)
0
0
0
0
38
0
181
20
en                                                                                         .    —-


GO                 |W detected.

(Vj;                  Measured value corrected for endogenous 2,3,7,8-TCDD content (averaged 616 pg/L).
^^"^      CO
         en

-------
                                     TABLE 16.  ACCURACY OF THE HHCC/HRMS METHOD FOR THE DETERMINATION OF TCDD ISOMERS SPIKED  IHTO SOIL MATRICES
Hyde Park 001 (HI)
TCDD
analyte
1,3,6,8
>,3,7,9
1,2,3,7/1
1.2,3,4
1,2,7,8
1,2.8,9
2.3.7,8
1SC,,-2,3
*HD - oot
to
cn
Spike
(Pi)
130
60
,2,3,8 120
170
220
84
-
,7,8 SOO
detected. The
Measured Recovery
(PR) (X)
29.0
51.1
12S
118
2S2
100
30.3
280
value in
22
86
106
69
117
119
-
56
parentheses
B25-Piazza Road (85)
Spike
(PS)
92
42
84
120
ISO
60
-
SOO
reflect*
Measured
(Pt)
HD (9.2)
IS. 2
61.8
S4.I
147
63.9
12.9
240
the estia
Recovery
(I)
• o
36
74
44
95
107
-
48
BS2-Sheoandoah
Spike
(PR)
660
300
600
880
1,110
430
-
SOO
Measured
(PC)
333
635
S18
69S
1,170
463
1,280
400
(Bl)
Recovery
(X)
SO
210
87
79
106
108
-
80
Hyde Park 003
Spike
(PR)
1,560
710
1,430
2,070
2,620
1,020
-
500
Measured
(PI)
383
367
82S
8SS
2,330
952
1,800
430
(H3)
Recovery
(X)
24
51
58
41
89
93
-
86
ated detection limit.
01
CO

-------
                                                                                            TABU I).  FOBT1F1F.D FIELD BUMK RESULTS
cn
oc
             co
Aliquot
Saiple Air-dry vt. (t)
Do. or Vol. (L)
8367-62- I576X-FFVB 1.0 L
8367-64- I576X-FFSB 10 |
B367-63-1576X-FF5A 10 |
8367-61- I576X-FFWA l.OL
8367-81- I576X-FFB 10 .01 «
8367-80-I576X-FFA 10.01 |
B367-97-1S76X-FFA 10 L
8367-98- 1576X-FTB 1 0 L
Retention limt
Katiire
23:38
23:38
23:40
23:40
22:17
21:37
21:37
21:43
'
23
23
23
23
22
21
21
21
JC
:35
:38
:38
:39
:15
:35
:37
:43
Inter.
ID '
HS50
MS50
MS50
HS50
HS50
MS50
MS50
ns50
Date
09/11/85
09/11/85
09/12/85
09/12/85
09/20/85
09/20/85
09/20/65
09/20/85
Ti-e
14:08
14:42
14:47
15:14
13:33
14:09
09: 10
09:24
Relati
320/322
0.78
0.79
0.77
0.80
0.88
0.86
0.82
0.80
ve Ion Abundance Ratio!
332/334(15) 332/334OS) X >ec.
0 71 65
0.80 66
0.78 - 71
0.76 - 79
0.74 29
0.73 - 83
0.83 - 50
0.82 48

•/z 259
> 145:1
> 145:1
145: 1
145:1
144:1
145:1
23:1
145:1
S/K
*/l 322
> 63:1
> 63:1
> 63:1
> 63:1
> 63:1
> 63:1
> 63:1
> 63:1

mil 334(18) Cootenti
> 63:1
> 63:1
> 63:1
> 63:1
> 63:1 . Low reiovery
> 63:1
> 63:1
> 63:1

-------
HRGC and Mass Resolution

     Table  11 presents a summary of all chromatographic and mass resolution
checks  completed during  the  final method evaluation.  As  per the protocol
requirements the required mass resolution was demonstrated as the first and
last  quality control  requirements  for each  day.   The  column performance
check mixture was also analyzed before the  first  sample  analysis and after
the  final  sample  analysis  each day as a QC measure  to  assure that speci-
ficity  for 2,3,7,8-TCDD  was  maintained.   The mass  measurement accuracy at
m/z  330.979 is  also included  in  this  table, as it was verified  on a daily
basis prior to any sample analyses.

Sample Analysis

     The results from  the analysis of  the aqueous and soil samples are pro-
vided in Table  12.  The data  in Table  12 are presented in the format speci-
fied  as  Form B-l in the protocol reporting requirement.   The data are re-
corded  in   the  chronological  order  that  they  were obtained  by  HRGC/HRMS.

     As  indicated in  Table  12, several samples required  reanalysis due to
low  recovery of  the internal  standard, unacceptable ion ratios for 320/322,
or  the  result of interferences at the internal standard.  Two of the dis-
tilled water  samples demonstrated responses for the  characteristic ions at
m/z  259,  320, and 322  for  2,3,7,8-TCDD.   However, the  ion  ratio  for the
native  2,3,7,8-TCDD in  one  replicate  and the  ion  ratio  for  the internal
standard in another  required that both  samples be reanalyzed.  Although both
samples met all  the qualitative criteria, recoveries were noted to be low
(< 20 percent) for one of the  samples  and complete reanalysis of the repli-
cate was required.

     Significant problems were encountered  with the  aqueous  soil extract,
H2W,  and  the fly ash  sample.  The problems with  the  soil extract resulted
from  an  interference  at m/z 332 that  coeluted  with the  internal standard,
13Ci2~2,3,7,8-TCDD.  No  accurate quantitative measurements could be achieved
for  TCDD responses observed for this sample.  The original sample contained
a  large  amount  of suspended  particulate  in each of  the three  replicates.
Problems with the extraction were noted with the first portion of methylene
chloride.   A large amount of  particulate matter was  noted at the interface
of  the  aqueous  and organic phases.  Hence,  the  interference  at m/z 332 and
TCDD  responses observed in these replicates were probably due to direct ex-
traction of the  suspended  soil particulate  rather  than  the  actual water-
soluble TCDD.

     The remaining aqueous  sample for  H2W was centrifuged for 10 min at ap-
proximately 2,000 rpm, and the aqueous phase was  decanted from  the settled
particulate.  The resulting aqueous sample was divided into duplicate 430-mL
samples  and each was  analyzed according  to  the protocol.   The HRGC/HRMS
analysis was successful for  both replicates  with absolute  recoveries of
78 percent and 96 percent of the internal standard.

     The  triplicate analyses  of  the   fly  ash  sample resulted  in absolute
recoveries  less   than  10 percent  for the  internal  standard in each aliquot

                                      38

-------
analyzed.  These  low  recoveries  may be associated with the total fixed car-
bon content  of  the  fly ash material.  Previous work in this laboratory with
fly ash from coal-fired power plants has demonstrated low recoveries of ana-
lytes from materials with high carbon content.4

     The only other sample for which successful analysis was not achieved as
specified  in the protocol  on first analysis  was  the  industrial wastewater
(IND).  The  triplicate  analysis  of the sample resulted in absolute internal
standard recoveries of  23,  20,  and 29 percent.  The criteria for successful
analysis for TCDD as  discussed  in the protocol require an absolute recovery
of 40 to 120 percent.   In addition to the observed low recoveries, the level
of 2,3,7,8-TCDD detected in the sample averaged 1,410 ppq as compared to the
500-ppq spike level.   Two  500-mL aliquots of the unspiked industrial waste-
water sample were reanalyzed  to  determine the background  level of 2,3,7,8-
TCDD.  The results of the duplicate analysis yielded an average 2,3,7,8-TCDD
concentration  of approximately  620 ppq  and  the  absolute  recoveries  were
noted to be 60 percent and 57 percent.  The increase in absolute recovery of
the  internal standard in  the unspiked sample by  approximately a factor of
two is possibly due to the preparation of samples one half the size of that
used  for the original analysis.   This suggests that the sample matrix has a
considerable impact on the effectiveness of the cleanup procedure.

     Table 13 provides a summary of the accuracy and precision of the analy-
ses of the five aqueous sample types for 2,3,7,8-TCDD.  Only the data points
from  Table 12 that  demonstrate  compliance with all QC criteria (ion ratios,
absolute recovery of  the  internal standard, etc.) are included in Table 13.
These data demonstrate that the isotope dilution method of quantitation pro-
vides accurate  and  precise quantitation of 2,3,7,8-TCDD in the aqueous sam-
ples.  It should be noted that even when the absolute recovery of the 13Ci2~
2,3,7,8-TCDD internal standard varies by as much as 66 percent  (RPR) for the
triplicate distilled  water  samples,  the accuracy  of the  measurement of the
spiked 2,3,7,8-TCDD averaged 101 percent with less than 10 percent variabil-
ity.  Table  13  summarizes  data  for both the spiked and unspiked aliquots of
industrial wastewater.  The high recovery noted  for  the  2,3,7,8-TCDD value
in  the  spiked  samples is a result  of the presence of  this  compound at ap-
proximately 620 ppq in the original matrix.

     Table 14  presents a  similar summary  for the five  solid samples ana-
lyzed.  The  precision of  the measurements is not quite as good as noted for
the  aqueous  samples  and  may reflect the  difference  in adsorption  of the
endogenous 2,3,7,8-TCDD  and the  spiked internal  standard  to the matrices.

     Tables  15  and  16 provide data  dealing with  the  accuracy of the HRGC/
HRMS  methods for the determination  of total TCDD  isomers  in aqueous and
solid samples.   In  general,  the  data support  the  use  of the internal stan-
dard  method   of  quantitation for  all  but the  earliest  eluting isomers,
1,3,6,8- and  1,3,7,9-TCDD.  The  accuracy for the additional isomers is very
good  and more  consistent  than is observed  for the solid samples.  This may
be  partially due to  the  differences  in  adsorption to  the soil particles.
                                      39

                       r~ o ""i
                       DO  i

-------
Fortified Field  Blanks

     As  part of the  overall  quality assurance/quality control (QA/QC) pro-
gram identified  in the HRGC/HRMS protocol,  the  analyst is required to ana-
lyze  fortified  field blanks  to demonstrate  (a)  that the  extraction and
cleanup  procedure  will provide  recovery of  the  2,3,7,8-TCDD within the cri-
teria  of greater than 40 percent specified  in the protocol and (b) that the
reagents are  free  from contamination with TCDD isomers.

     Table  17 provides the results  of  the fortified field blanks run before
proceeding  with  sample  analysis  and  also  those  of  an  additional  set of
blanks prepared  along with the  actual  samples.   The  analyses of the forti-
fied field  blanks  at the outset of  the study demonstrated that the recover-
ies of 2,3,7,8-TCDD and 1,2,3,4-TCDD ranged from 65  to 79 percent.  No de-
tectable  levels  of other TCDD isomers  were  found in this preliminary study.
The field  fortification blanks  analyzed with the actual samples resulted in
recoveries  of 29 percent and  83 percent.  More  importantly,  these analyses
demonstrated  some  interferences  arising  from  1,3,6,8-  and 1,3,7,9-TCDD.
Previous  studies  involving evaluation  of   the  cleanup procedure indicated
that these  isomers  are associated with  the acidic alumina cleanup.

     Figure  4 is  a plot  of the ratio  of response  of  1,3,6,8- and 1,3,7,9-
TCDD and the  response of the recovery standard 13C12-1,2,3,4-TCDD versus the
time elapsed since  the  acidic  alumina was  cleaned and activated at 190°C.
The results of  the analyses  of the fortified field blanks  and  the samples
not spiked  with  the 1,3,6,8-  and 1,3,7,9-TCDD isomers  are presented in Fig-
ure 4.   As  noted from this plot, these TCDD isomers  were not initially de-
tected  in the acidic  alumina immediately following cleanup  by  Soxhlet ex-
traction.   The  first  set  of  fortified  field blanks   was  taken  through the
acidic alumina  column 7 days later.  Although response was  observed at m/z
320 and  322 at  the  retention  time for  these isomers,   the ion ratios did not
indicate presence  of the compounds.  Since  the  detectable  levels were well
below  10 pg/g of alumina,  the sample analyses were initiated.  The data for
the fortified field blanks and  samples taken  through  alumina from 14 to 30
days from activation  indicate that the contamination of the  1,3,6,8- and
1,3,7,9-TCDD  isomers apparently  occurs  over  time using  this particular oven.
The background contamination of 1,3,6,8- and  1,3,7,9-TCDD  isomers  has also
been recently addressed by  the Center for Disease Control.5

Note Added  in Proof

     A  second magnetic sector instrument  (built in  1976)  from  a different
manufacturer  was tested and  found  to  be incapable of achieving sufficient
sensitivity  at  10,000 resolving power  to be  used in experiments  for this
study.
                                      40

-------
      Q
      Q

      U
    1.2




    1.1




    1.0
      CO
        *

      CN
  r  0.9
 u
CO
       2 0.8




       3  0.7
       o  0.6

       U


       Js  0.5
       CO

       --  0.4
       00
       o-  0.3
       o  0.2
       o
       &!.
          0.1
• Fortified Field Blanks



  I Aqueous and Environmental Samples
Samples

1
                            10              20

                    Time (Days) Elapsed from Cleanup and

                        Activation of Acidic Alumina
                                                     30
Figure 3.  Background  levels of 1,3,6,8- and  1,3,7,9-TCDD observed

           over the  single-laboratory evaluation study.



                                  41

-------
                                 REFERENCES
1.  U.S. Environmental Protection Agency, "Dioxin Strategy,"  prepared  by  the
    Office of  Water Regulations  and Standards and the Office of Solid Waste
    and  Emergency  Response  in  conjunction  with  the  Dioxin Strategy Task
    Force, Washington, B.C., November 28, 1983.

2.  L. R. Williams, Validation of Testing/Measurement Methods.
    EPA 600/X-83-060, 1983.

3.  GC Bulletin 793C, Supelco Inc., Beliefonte,  Pennsylvania, 1983.

4.  C.  L.  Haile,  J.  S.  Stanley,  T. Walker,  G.  R. Cobb, and B. A. Boomer,
    "Comprehensive  Assessment  of the Specific Compounds Present in Combus-
    tion  Processes.  Volume 3.   National  Survey  of Organic Emissions from
    Coal-Fired  Utility  Boiler Plants,"  EPA-560/5-83-006,  September  1983.

5.  J.  S.  Heller,  D. G.  Patterson,  L. R.  Alexander, D. F.  Groce,  R.   P.
    O'Connor,  and C.  R.  Lapeza,  "Control of Artifacts and Contamination in
    the Development of  a Dioxin Analytical Program," presented at the 33rd
    Annual  Conference on  Mass Spectrometry and  Allied  Topics, May  26-31,
    1985, San Diego, California.
                                      42
                                       590

-------
     APPENDICES
        43




531

-------
                                   APPENDIX A

                         VALIDATED ANALYTICAL PROTOCOL
                            for the Determination of
          2,3,7,8-Tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD)  and  Total
            TCDDs in Soil/Sediment and Water by High-Resolution Gas
                Chromatography/High-Resolution Mass Spectrometry
                               September 10,  1985

     This analytical protocol  has been written in the format used in the
Superfund program, as "Exhibit D" of a Statement of Work  which  in turn  is  part
of an Invitation-for-Bid package under the Superfund Contract Laboratory Program.
The other exhibits of the Statement of Work,  although cited  in  Exhibit  D,  do
not pertain to this method evaluation study.
                                           c; c  9

-------
                         EXHIBIT D
                     Analytical Methods

2,3,7,8-Tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) and Total
  TCDDs in Soil/Sediment and Water by High-Resolution Gas
      Chromatography/High-Resolution Mass Spectrometry
                                r o, l'\
                                u 
-------
                                   EXHIBIT D









Section                       Subject                              Page






   1               Scope and Application	    D-l




   2               Summary of Method	    D-l




   3               Definitions	    D-2




   4               Interferences	    D-3




   5               Safety	    D-3




   6               Apparatus and Equipment	    D-3




   7               Reagents and Standard Solutions	    D-6




   8               System Performance Criteria	    D-8




   9               Quality Control Procedures 	    D-13




  10               Sample Preservation and Handling 	    D-13




  11               Sample Extraction	    D-14




  12               Analytical Procedures	    D-l 7




  13               Calculations	    D-18

-------
1.   SCOPE AND APPLICATION

     1.1   This method provides  procedures  for  Che  detection  and  quantitative
          measurement of  2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD; CAS
          Registry Number 1746-01-6;  Storet  number 3475) at  concentrations of
          10 pg/g (10 parts per trillion)  to 200 pg/g  (200 parts per  trillion)
          in 10-g portions of soil  and  sediment and at  100 pg/L  (100  parts per
          quadrillion)  to 2000  pg/L (2  parts per trillion) in 1-L samples  of
          water.   The use of  1-g  or 100-mL portions permits  measurements of
          concentrations  up to  2,000  pg/g  (2 parts per  billion)  or 20 ng/L,
          respectively*   This method  also  allows the estimation  of  quantities
          of total TCDD present in  the  sample.  Samples containing concentrations
          of 2,3,7,8-TCDD greater than  2 ppb or 20 ng/L must be  analyzed by a
          protocol designed for such  concentration levels, with  an  appropriate
          instrument calibration range.

     1.2   The-minimum measurable  concentration is  estimated  to be 10  pg/g  (10
          parts per trillion) for soil  and sediment samples  and  100 pg/L for
          water samples,  but  this depends  on kinds and  concentrations of
          interfering compounds in  the  sample  matrix.

     1.3   This method is  designed for use  by analysts who are experienced  in
          the use of high-resolution gas chromatography/high-resolution mass
          spectrometry•

     CAUTION:  TCDDs are  extremely  hazardous.   It  is the laboratory's responsi-
               bility to  ensure that  safe  handling procedures are employed.

2.   SUMMARY OF METHOD

     Five hundred pg of   C,2~2,3,7,8-TCDD (internal standard) are added to a
     10-g portion of a soil/sediment  sample  (weighed to 3 significant figures)
     or a 1-L aqueous sample  and  the  sample  is extracted with 200 to  250 mL
     benzene using a Soxhlet  apparatus  with  a  minimum  of 3 cycles per hour or a
     continuous liquid-liquid extractor for  24 hours.   A separatory funnel and
     3 x 60 mL methylene  chloride may also be  used for  aqueous samples.  After
     appropriate concentration  and  cleanup,  50  uL  of tridecane are  added to the
     extract.  Before HRGC-HRMS analysis,  500  pg of a  recovery standard (  C^"
     1,2,3,4-TCDD) are added  to the extract  which  is then concentrated to  a
     final volume of 50 uL.  A  2-uL aliquot  of the concentrated  extract  is
     injected into a gas  chromatograph  with  a  capillary column interfaced  to a
     high-resolution mass spectrometer  capable of  rapid multiple ion  monitoring
     at resolutions of at least 10,000  (10 percent valley).

     Identification of 2,3,7,8-TCDD is  based on the detection of the  ions  m/z
     319.897 and 321.894  at the same  GC retention  time  and within -1  to +3
     seconds GC retention time  of the internal standard masses of m/z 331.937
     and  333.934.  Confirmation of  2,3,7,8-TCDD (and of other TCDD isomers) is
     based on the ion m/z 258.930 which results from loss of COCL by  the parent
     ion.
                                      D-l
                                        n

-------
3.   DEFINITIONS

     3*1  Concentration calibration solutions -- solutions containing known
          amounts of the analyte (unlabeled 2,3,7,8-TCDD), the internal standard
          13C12-2,3,7,8-TCDD and the recovery standard 13C,2~1,2,3,4-TCDD;
          they are used to determine instrument response or the analyte
          relative to the internal standard and of the internal standard
          relative to the recovery standard.

     3.2  Field blank — a portion of soil/sediment or water uncontaminated with
          2,3,7,8-TCDD and/or other TCDDs.

     3.3  Rinsate — a portion of solvent used to rinse sampling equipment; the
          rinsate is analyzed to demonstrate that samples have not been contami-
          nated during sampling.

     3.A  Internal standard — ^Cj2~2»3,7,8-TCDD, which is added to every
          sample (except the blanks described in Sections 4.2.1 and 4.2.3 of
          Exhibit E) and is present at the same concentration in every labora-
          tory method blank, quality control sample, and concentration calibra-
          tion solution.  It is added to the soil/sediment or aqueous sample
          before extraction and is used to measure the concentration of each
          analyte.  Its concentration is measured in every sample, and percent
          recovery is determined using an internal standard method.

     3.5  Recovery standard —   C12~*,2,3,4-TCDD which is added to every sample
          (except for the blanks discussed in Sections 4.2.1.A.2 and 4.2.3.6,
          Exhibit E) extract just before HRGC-HRMS analysis.

     3.6  Laboratory method blank — this blank is prepared in the laboratory
          through performing all analytical procedures except addition of a
          sample aliquot to the extraction vessel.

     3.7  GC column performance check mixture —• a mixture containing known
          amounts of selected standards; it is used to demonstrate continued
          acceptable performance of the capillary column, i.e., separation
          (jC 25% valley) of 2,3,7,8-TCDD isomer from all other 21 TCDD isomers
          and to define the retention time window.

     3.8  Performance evaluation sample — a soil, sediment or aqueous sample
          containing a known amount of unlabeled 2,3,7,8-TCDD and/or other
          TCDDs.  It is distributed by EPA to potential contractor laboratories
          who must analyze it and obtain acceptable results before being awarded
          a contract for sample analyses (see IFB Pre-Award Bid Confirmations).
          It may also be included as an unspecified ("blind") QC sample in any
          sample batch submitted to a laboratory for analysis.

     3.9  Relative response factor — response of the mass spectrometer to a
          known amount of an analyte relative to a known amount of an internal
          standard.

     3.10 Mass resolution check — standard method used to demonstrate static
          resolution of 10,000 minimum (102 valley definition).

                                      D-2
                                        r: r

-------
4.   INTERFERENCES

     Chemicals which elute from the GC column within ^10  scans  of  the  internal
     and/or recovery standard (ra/z 331.937  and 333.934) and  which  produce  ions
     at any of the masses used to detect  or quantify TCDD are potential  inter-
     ferences.  Most frequently encountered potential  interferences  are  other
     sample components that are extracted along with TCDD, e.g.  PCBs,  methoxy-
     biphenyls, chlorinated hydroxydiphenylethers,  benzylphenylethers, chlori-
     nated naphthalenes,  DDE, DDT, etc.  The actual incidence of interference
     by these chemicals depends also upon relative  concentrations, mass  spectro-
     metric resolution, and chromatographic conditions.   Because very  low
     levels of TCDD must be measured,  the elimination  of  interferences is
     essential.  High-purity reagents  and solvents  must be used  and  all  equip-
     ment must be scrupulously cleaned.  Laboratory reagent  blanks (Exhibit  E,
     Quality Control, Section 4)  must  be  analyzed to demonstrate absence of
     contamination that would interfere with TCDD measurement.   Column chromato-
     graphic procedures are used  to remove  some coextracted  sample components;
     these procedures must be performed carefully to minimize  loss of  TCDD
     during attempts to increase  its concentration  relative  to  other sample
     components.

5.   SAFETY
                          •

     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 laboratory is  responsible for  maintaining a file  of
     current OSHA regulations regarding the safe handling of the chemicals
     specified in this method.  A reference file of material data  handling
     sheets should also be made available to all personnel involved  in the
     chemical analysis.  Additional references to  laboratory safety  are  identi-
     fied d~3) (page D-38).  2,3,7,8-TCDD  has been identified  as  a  suspected
     human or mammalian carcinogen.  The  laboratory is responsible for ensuring
     that safe handling procedures are followed.

6.   APPARATUS AND EQUIPMENT

     6.1  High-Resolution Gas Chrotnatograph/High-Resolution  Mass
          Spectrometer/Data System (HRGC/HRMS/DS)

          6.1.1  The GC must be equipped  for temperature  programming,  and  all
                 required accessories must  be available,  such  as syringes, gases,
                 and a capillary  column.  The GC injection port must be  designed
                 for capillary columns.  The use of splitless  injection  tech-
                 niques is recommended.  On-column  injection techiques can be
                 used but this may severely reduce  column lifetime for non-
                 chemical ly bonded columns.  When using the  method  in  this
                 protocol, a 2-uL injection volume  is  used consistently.  With
                 some GC injection ports, however,  1-uL injections may produce
                 improved precision and chromatographic  separation.  A 1-uL
                                      D-3

                                          597

-------
                 injection volume may be used if adequate sensitivity and
                 precision can be achieved.

          NOTE:  If 1 uL is used at all as injection volume, the injection
                 volumes for all extracts, blanks, calibration solutions and
                 the performance check sample must be 1 uL.

          6.1.2  Gas Chromatograph-Mass Spectrometer Interface

                 The GC-MS interface may include enrichment devices, such as
                 a glass jet separator or a silicone membrane separator, or
                 the gas chromatograph can be directly coupled to the mass
                 spectrometer source.  The interface may include a diverter
                 valve  for shunting the column effluent and isolating the mass
                 spectrometer source.  All components of the interface should
                 be glass or glass-lined stainless steel.  The interface com-
                 ponents should be compatible with 300°C temperatures.  The
                 GC/MS  interface must be appropriately designed so that the
                 separation of 2,3,7,8-TCDD from the other TCDD isomers which
                 is achieved in the gas chromatographic column is not appreci-
                 ably degraded.  Cold spots and/or active surfaces (adsorption
                 sites) in the GC/MS interface can cause peak tailing and peak
                 broadening.  It is recommended that the GC column be fitted
                 directly into the MS source.  Graphite ferrules should be
                 avoided in the GC injection area since they may adsorb TCDD.
                 Vespel* or equivalent ferrules are recommended.

          6.1.3  Mass Spectrometer

                 The static resolution of the instrument must be maintained at
                 a minimum 10,000 (10 percent valley).  The mass spectrometer
                 must be operated in a selected ion monitoring (SIM) mode with
                 total  cycle time (including voltage reset time) of one second
                 or less (Section 8.3.A.I).  At a minimum, the following ions
                 which  occur at these masses must be monitored:  m/z 258.930,
                 319.897, 321.894, 331.937 and 333.934.

          6.1 .4  Data System

                 A dedicated hardware or data system is employed to control the
                 rapid  multiple ion monitoring process and to acquire the data.
                 Quantification data (peak areas or peak heights) and SIM traces
                 (displays of intensities of each m/z being monitored as a
                 function of time) must be acquired during the analyses.
                 Quantifications may be reported based upon computer-generated
                 peak areas or upon measured peak heights (chart recording).

NOTE:  Detector zero setting must allow peak-to-peak measurement of the noise
       on the base line.

     6.2  GC Columns
                                      D-4

-------
     For isoner-spec ific determinations of 2,3,7,8-TCDD,  the  following  two
     fused silica capillary columns  are recommended:   a  60-ra  SP-2330  column
     and a 50-m CP-Sil 88 column.   However,  any capillary column which
     separates 2,3,7,8-TCDD from all other TCDDs may  be  used  for such
     analyses, but this separation must be demonstrated  and documented.
     Minimum acceptance criteria must be determined per  Section 8.1.  At
     the beginning of each 12-hour period (after mass resolution has  been
     demonstrated) during which sample extracts or concentration calibra-
     tion solutions will be analyzed, column operating conditions must  be
     attained for the required  separation on the column  to be used for
     samples.  Operating conditions  known to produce  acceptable results
     with the recommended columns  are shown in  Table  2 at the end of  this
     Exhibit.

6.3  Miscellaneous Equipment

     6.3.1  Nitrogen evaporation apparatus with variable flow rate.

     6.3.2  Balance capable of  accurately weighing to 0.01 g.

     6.3.3  Centrifuge capable  of  operating at  2,000  rpm.

     6.3.4  Water bath — equipped with concentric ring  cover and capable
            of being temperature-controlled within j*2°C.

     6.3.5  Stainless steel spatulas or spoons.

     6.3.6  Stainless steel (or glass) pan large enough  to hold contents
            of 1-pint sample containers.

     6.3.7  Glove box.

     6.3.8  Drying oven.

6.4  Glassware

     6.4.1  Soxhlet apparatus — all-glass, Kontes 6730-02 or equivalent;
            90 mm x 35 mm glass thimble; 500-mL flask; condenser of appro-
            priate size.

     6.4.2  Kuderna-Danish apparatus — 500-mL  evaporating flask, 10-mL
            graduated concentrator tubes with ground-glass stoppers,  and
            3-ball macro Snyder column (Kontes  K-570001-0500, K-503000-
            0121 and K-569001-0219 or equivalent).

     6.4.3  Mini-vials — 1-mL  borosilicate glass with conical-shaped
            reservoir and screw caps lined with Teflon-faced  silicone disks.

     6.4.4  Funnels — glass; appropriate size  to accommodate filter
            paper used to filter jar extract (volume  of  approximately 170 mL).

     6.4.5  Separatory funnel -- 2000 mL with Teflon  stopcock.

                                D-5

-------
          6.4.6  Continuous liquid-liquid extractors equipped with Teflon or
                 glass connecting joints and stopcocks requiring no lubrication
                 (Hershberg-Wolf Extractor - Ace Glass Company Vineland,  NJ,
                 P/N 6841-10 or equivalent).

          6.4.7  Chromatographic columns for the silica and alumina chroma-
                 tography — 1 cm ID x 10 cm long and 1 cm ID x 30 cm long.

          6.4.8  Chromatography column for the Carbopak cleanup — disposable
                 5-mL graduated glass pipets, 7 mm ID.

          6.4.9  Desiccator.

          6.4.10 Glass rods.

          NOTE:  Reuse of glassware should be minimized to avoid the risk of
                 cross contamination.  All glassware that is reused must  be
                 scrupulously cleaned as soon as possible after use, applying
                 the following procedure.

                 Rinse glassware with the last solvent used in it then with
                 high-purity acetone and hexane.  Wash with hot water containing
                 detergent.  Rinse with copious amounts of tap water and  several
                 portions of distilled water.  Drain dry and heat in a muffle
                 furnace at 400°C for 15 to 30 minutes.  Volumetric glassware
                 should not be heated in a muffle furnace, and some thermally
                 stable materials (such as PCBs) may not be removed by heating
                 in a muffle furnace.  In these cases, rinsing with high-purity
                 acetone and hexane may be substituted for muffle furnace
                 heating.  After the glassware is dry and cool, rinse with hexane,
                 and store inverted or capped with solvent-rinsed aluminum foil
                 in a. clean environment.

7.   REAGENTS AND STANDARD SOLUTIONS

     7.1  Column Chromatography Reagents

          7.1.1  Alumina, acidic — Extract the alumina in a Soxhlet with
                 methylene chloride for 6 hours (minimum of 3 cycles per  hour)
                 and activate it by heating in a foil-covered glass container
                 for 24 hours at 190°C.

          7.1.2  Silica gel — high-purity grade, type 60, 70-230 mesh; extract
                 the silica gel in a Soxhlet with methylene chloride for  6 hours
                 (minimum of 3 cycles per hour) and activate it by heating in a
                 foil-covered glass container for 24 hours at 130°C.

          7.1.3  Silica gel impregnated with 40 percent (by weight) sxilfuric
                 acid — add two parts (by weight) concentrated sulfuric  acid
                 to three parts (by weight) silica gel (extracted and activated),
                 mix with a glass rod until free of lumps, and store in a
                 screw-capped glass bottle.

                                      D-6
                                              GOO

-------
     7.1.4  Sulfuric acid,  concentrated  — ACS  grade,  specific  gravity  1.84.

     7.1.5  Graphitized  carbon  black (Carbopack C  or equivalent),  surface
            of approximately 12 m^/g,  80/100  mesh  — mix thoroughly  3.6
            grams Carbopak  C and 16.4  grams Celite 545*  in a 40-mL vial.
            Activate at  130° C  for  six hours.  Store in  a  desiccator.

     7.1.6  Celite 545®, reagent grade,  or equivalent.

7.2  Membrane filters or filter paper  with pore size of  <25 urn;  rinse with
     hexane before use.

7.3  Glass wool,  silanized  — extract  with methylene chloride and  hexane
     and air-dry  before  use.

7.4  Desiccating  Agents

     7.4.1  Sodium sulfate  — granular,  anhydrous; before  use,  extract  it
            with  methylene  chloride for  6  hours (minimum of 3 cycles per
            hour) and dry it for >4 hours  in  a  shallow tray placed in an
            oven  operated at 120°C.  Let it cool in a  desiccator.

     7.4.2  Potassium carbonate—anhydrous, granular;  use  as such.

7.5  Solvents —  high purity,  distilled  in glass:   methylene chloride,
     toluene, benzene,  cyclohexane, methanol, acetone,  hexane;  reagent
     grade:  tridecane.

7.6  Concentration calibration solutions (Table 1) —  five tridecane
     solutions containing unlabeled 2,3,7,8-TCDD and 13C,o-l,2,3,4-TCDD
     (recovery standard) at varying concentrations, and    Ci -y-2 ,3,7,8-TCDD
     (internal standard, CASRN 80494-19-5) at a constant concentration
     must be used to calibrate  the  instrument.   These  concentration  calibra-
     tion solutions must be obtained from  the Quality  Assurance  Division,
     US EPA Environmental Monitoring Systems  Laboratory  (EMSL-LV), Las  Vegas,
     Nevada.  However,  additional secondary standards  may  be obtained from
     commercial sources, and solutions may be prepared in  the contractor
     laboratory.   Traceability  of standards must be verified against EPA-
     supplied standard solutions.  Such  procedures will  be documented by
     laboratory SOPs as  required in IFB  Pre-award  Bid  Confirmations, part
     2.f.(4).  It is the responsibility  of the  laboratory  to ascertain  that
     the calibration solutions  received  are indeed at  the  appropriate
     concentrations before  they are injected  into the  instrument.   Serious
     overloading  of the  instrument  may occur  if the concentration  calibra-
     tion solutions intended for a  low-resolution MS are injected  into  the
     high-resolution MS.
     7.6.1   The five concentration calibration solutions contain unlabeled
            2,3,7,8-TCDD and  labeled ^C.--1»2,3,4-TCDD at  nominal  concen-
            trations of 2.5,  5.0,  10.0,  20.0  and  40.0  pg/uL,  respectively,
            and labeled   Cj^-2 ,3 , 7 ,8-TCDD at  a constant nominal concen-
            tration of 10.0 pg/uL.

                                 D-7
                                601

-------
          7.6.2  Store the concentration calibration  solutions  in  1-mL mini-
                 vials at 4°C.

     7.7  Column performance check mixture — this  solventless  mixture must be
          obtained fvom the Quality Assurance Division,  Environmental Monitoring
          Systems Laboratory, Las Vegas, Nevada,  and  dissolved  by  the Contractor
          in 1 mL tridecane.  This solution will  then contain the  following
          components (including TCDDs (A) eluting closely  to 2,3,7,8-TCDD, and
          the first- (F) and last-eluting (L) TCDDs when using  the columns
          recommended in Section 6.2) at a concentration of 10  pg/uL of each of
          these isomers:

          	Analyte	               Approximate Amount Per Ampule

          Unlabeled 2,3,7,8-TCDD                          10 ng

          13C12-2,3,7,8-TCDD                              10 ng

          1,2,3,4-TCDD (A)                                10 ng

          1,4,7,8-TCDD (A)                                10 ng

          1,2,3,7-TCDD (A)                                10 ng

          1,2,3,8-TCDD (A)                                10 ng

          1,2,7,8-TCDD                                    10 ng

          1,3,6,8-TCDD (F)                                10 ng

          1,2,8,9-TCDD (L)                                10 ng


     7.8  Sample fortification solution — an isooctane  solution containing
          the internal standard at a nominal concentration of 5 pg/uL.

     7.9  Recovery standard spiking solution — an  isooctane solution contain-
          ing the recovery standard at a nominal concentration  of  100 pg/uL.
          Five uL of this solution will be spiked into the extract before
          HRGC/HRMS analysis.

     7.10 Internal standard spiking solution — an isooctane  solution containing
          the internal standard at a nominal concentration of 100  pg/uL.   Five
          uL of this solution will be added to a fortified field blank extract
          (Section 4.2.1.A.2, Exhibit E).

8.   SYSTEM PERFORMANCE CRITERIA

     System performance criteria are presented in two sections. One section
     deals with GC column performance criteria while  the other  section consists
     of initial calibration criteria.  The laboratory may use either of  the
     recommended columns described in Section 6.2.   It must be  documented  that

                                      D-8

-------
all applicable system performance criteria  specified  in  Sections  8.1, 8.2
and 8.3 have been met before analysis  of  any  sample is performed.  Table 2
provides recommended conditions that  can  be used  to satisfy  the required
criteria.  Table 3 provides a typical  12-hour analysis sequence.

8.1  GC Column Performance

     8.1.1   Inject 2 uL (Section 6.1.1) of  the column performance check
            solution (Section 7.7)  and acquire selected  ion  monitoring
            (SIM) data for m/z 258.930, 319.897,  321.894,  331.937 and
            333.934 within a total  cycle  time of  ^1 second (Section
            8.3.4.1).

     8.1.2   The chromatographic peak  separation between  2,3,7,8-TCDD and
            the peaks representing  any other  TCDD isomers  must be resolved
            with a valley of <25 percent, where

                           Valley Percent  »   (x/y)(100)

            x  =  measured as in Figures  1  and 2

            y  «  the peak height of  2,3,7,8-TCDD.

            It is the responsibility  of the laboratory to  verify  the con-
            ditions suitable for the  appropriate  resolution of 2,3,7,8-TCDD
            from all other TCDD isomers.  The column  performance  check
            solution also contains  the TCDD isomers eluting first and  last
            under the analytical conditions specified in this protocol
            thus defining the retention time  window  for  total TCDD determi-
            nation.  The peaks representing 2,3,7,8-TCDD,  the first and
            the last eluting TCDD isomers must be labeled  and  identified
            as such on the chromatograms.

8.2  Mass Spectrometer Performance

     8.2.1   The mass spectrometer must be operated in the  electron (impact)
            ionization mode.  Static  mass resolution  of  at least  10,000
            (10 percent valley) must  be demonstrated  before any analysis
            of a set of samples is  performed  (Section 8.2.2).  Static
            resolution checks must  be performed at the beginning  and  at
            the end of each 12-hour period  of operation.  However, it  is
            recommended that a visual check (i.e.,  not documented) of  the
            static resolution be made  using the peak  matching unit before
            and after each analysis.

     8.2.2   Chromatography time for TCDD  may  exceed  the  long-term mass
            stability of the mass spectrometer and  thus  mass drift correc-
            tion is mandatory.  A reference compound  (high boiling PFK  is
            recommended) is introduced into the mass  spectrometer.  An
            acceptable lock mass ion  at any mass  between m/z 250  and m/z
            334 (m/z 318.979 from PFK is  recommended) must be used to
            monitor and correct mass  drifts.

                                 D-9


                                      603

-------
     NOTE:  Excessive PFK may cause background noise problems and contami-
            nation of the source resulting in an increase in "downtime"
            for source cleaning.

            Using a PFK molecular leak, tune the instrument to meet the
            minimum required mass resolution of 10,000 (10Z valley) at
            m/z 254.986 (or any other mass reasonably close to m/z 259).
            Calibrate the voltage sweep at least across the mass range m/z
            259 to m/z 334 and verify that m/z 330.979 from PFK (or any
            other mass close to m/z 334) is measured within _+5 ppm (i.e.,
            1.7 mmu) using m/z 254.986 as a reference.  Documentation of  the
            mass resolution must then be accomplished by recording the
            peak profile of the PFK reference peak m/z 318.979 (or any
            other reference peak at a mass close to m/z 320/322).  The
            format of the peak profile representation must allow manual
            determination of the resolution, i.e., the horizontal axis
            must be a calibrated mass scale (amu or ppm per division).
            The result of the peak width measurement (performed at 5
            percent of the maximum) must appear on the hard copy and
            cannot exceed 31.9 mmu or 100 ppm.

8.3  Initial Calibration

     Initial calibration is required before any samples are analyzed for
     2,3,7,8-TCDD.  Initial calibration is also required if any routine
     calibration does not meet the required criteria listed in Section 8.6.

     8.3.1  All concentration calibration solutions listed in Table 1 must
            be utilized for the initial calibration.

     8.3.2  Tune the instrument with PFK as described in Section 8.2.2.

     8.3.3  Inject 2 uL of the column performance check solution (Section
            7.7) and acquire SIM mass spectral data for m/z 258.930,
            319.897, 321.894, 331.937 and 333.934 using a total cycle time
            of £ 1 second (Section 8.3.4.1).  The laboratory must not
            perform any further analysis until it has been demonstrated
            and documented that the criterion listed in Section 8.1.2 has
            been met.

     8.3.4  Using the same GC (Section 8.1) and MS (Section 8.2) conditions
            that produced acceptable results with the column performance
            check solution, analyze a 2-uL aliquot of each of the 5 concen-
            tration calibration solutions in triplicate with the following
            MS operating parameters.

            8.3.4.1  Total cycle time for data acquisition must be £ 1
                     second.  Total cycle time includes the sum of all  the
                     dwell times and voltage reset times.

            8.3.4.2  Acquire SIM data for the following selected
                     characteristic ions:
                                 D-10

-------
                            m/z          Compound

                          258.930         TCDD  -  COC1

                          319.897         Unlabeled  TCDD

                          321.894         Unlabeled  TCDD

                          331.937         13C12-2,3,7,8-TCDD, 13C12~1 ,2,3 ,4-TCDD

                          333. 934         13C12-2,3,7,8-TCDD, 13C12-1 ,2 ,3 ,4-TCDD
                 8.3.4.3  The ratio of integrated  ion  current  for m/z 319.897 to
                          m/z 321.894 for 2,3,7,8-TCDD must  be  between 0.67 and
                          0.90.

                 8.3.4.4  The ratio of integrated  ion  current  for m/z 331.937 to
                          m/z 333.934 for l JC12-2,3, 7, 8-TCDD and 13C12-1 ,2 ,3 ,4-
                          TCDD must be between  0.67  and 0.90.

                 8.3.4.5  Calculate the relative response  factors for unlabeled
                          2,3,7,8-TCDD [RRF(I) ]  relative to  13C12~2 ,3 , 7 ,8-TCDD
                          and for  labeled   C12-2 ,3 ,7, 8-TCDD [RRF(II)] relative
                          to 13C-1,2,3,4-TCDD as  follows:
                                       RRF(I)   =   —
                                                  QX  '  AIS


                                                  Ais * QRS
                                      RRF(II)   =  	
                                                  QIS " ^S
where

     AX  ™  sum of the integrated ion abundances of m/z 319.897  and  m/z 321.894
            for unlabeled 2,3,7,8-TCDD.

     AIS  »  sum of the integrated ion abundances of  m/z 331.937 and m/z 333.934
            for 13C12-2,3,7,8-TCDD.

    ARs  B  sura of the integrated ion abundances for  m/z 331.937 and m/z
            333.934 for 13C12~1,2,3,4-TCDD.

    QIS  *  Quantity of 13C^2~2,3,7,8-TCDD injected (pg).

    QRS  -  quantity of 13Cj2~1,2,3,4-TCDD injected (pg).

     Qx  *  quantity of unlabeled 2,3,7,8-TCDD injected (pg).


                                      D-ll

                                              605

-------
    RRF is a dimensionless quantity; the units used to express QIS,  QRS and Qx
must be the same*

     8.4  Criteria for Acceptable Calibration

          The criteria listed below for acceptable calibration must  be net
          before analysis of any sample is performed.

          8.4.1  The percent relative standard deviation (RSD) for the response
                 factors from each of the triplicate analyses for both unlabeled
                 and   Cj2~2»3,7,8-TCDD must be less than ^20 percent.

          8.4.2  The variation of the 5 mean RRFs for unlabeled 2,3,7,8-TCDD
                 obtained from the triplicate analyses must be less  than _+20
                 percent RSD.

          8.4.3  SIM traces for 2,3,7,8-TCDD must present a signal-to-noise
                 ratio of >2.5 for m/z 258.930 and MO for m/z 321.894.

          8.4.4  SIM traces for   Cj2~2,3,7,8-TCDD must present a signal-to-
                 noise ratio 2.10 for 333.934.

          8.4.5  Isotopic ratios (Sections 8.3.4.3 and 8.3.4.4) must be within
                 the allowed range.
                                               f * W » ^ •• W WV»**VfcBWAWtl WWIIWCvllk^a
                                                5 triplicate determinations for
                                                13C12-2,3,7,8-TCDD will be used
     NOTE:  If the criteria for acceptable calibration listed in Sections
            8.4.1 and 8.4.2 have been met, the RRF can be considered  inde-
            pendent of the analyte quantity for the calibration concentra-
            tion range.  The mean RRF from
            unlabeled 2,3,7,8-TCDD and for
            for all calculations until routine'calibration criteria (Section
            8.6). are no longer met.  At such time, new mean RRFs will be
            calculated from a new set of five triplicate determinations.

8.5  Routine Calibrations

     Routine calibrations must be performed at the beginning of a 12-hour
     period after successful mass resolution and GC column performance
     check runs.

     8.5.1  Inject 2 uL of the concentration calibration solution which
            contains 5.0 pg/uL of unlabeled 2,3,7.8-TCDD, 10.0 pg/uL
            of 13C12-2,3,7,8-TCDD and 5.0 pg/uL   C12-l,2,3,4-TCDD.
            Using the same GC/MS/DS conditions as used in Sections 8.1,
            8.2 and 8.3, determine and document acceptable calibration as
            provided in Section 8.6.

8.6  Criteria for Acceptable Routine Calibration

     The following criteria must be met before further analysis is per-
     formed.  If these criteria are not met, corrective action must be
     taken and the instrument must be recalibrated.

                                 D-12

-------
          8.6.1   The measured RRF for unlabeled  2,3,7,8-TCDD  must  be  within  +20
                 percent  of  the  mean  values  established  (Section 8.3.4.6)  by
                 triplicate  analyses  of concentration  calibration  solutions.

          8.6.2   The measured RRF for   C,2~2,3,7,8-TCDD must  be within  +20  per-
                 cent of  the mean value established  by triplicate  analysis
                 of the concentration calibration  solutions  (Section  8.3.4.6).

          8.6.3   Isotopic ratios (Sections 8.3.4.3 and 8.3.4.4) must  be  within
                 the allowed range.

          8.6.4   If one of the above  criteria  is not satisfied, a  second attempt
                 can be made before  repeating  the  entire initialization  process
                 (Section 8.3).

          NOTE:   An initial  calibration must be  carried out  whenever  any HRCC
                 solution is replaced.

9.   QUALITY CONTROL PROCEDURES

     See Exhibit E for QA/QC requirements.

10.  SAMPLE PRESERVATION  AND HANDLING

     10.1 Chain-of-custody procedures — see Exhibit G.

     10.2 Sample Preservation

          10.2.1 When received,  each  soil or sediment  sample.will  be  contained
                 in a 1-pint glass jar surrounded  by vermiculite in a sealed
                 metal paint can.  Until a portion is  to be  removed  for  analysis,
                 store the sealed paint cans in  a  locked limited-access  area
                 where the temperature is maintained between 25° and  35°C.
                 After a  portion of a sample has been  removed for  analysis,
                 return the  remainder of the sample  to its original  container
                 and store as stated  above.

          10.2.2 Each aqueous sample  will be contained in a  1-liter  glass
                 bottle.   The bottles with the samples are stored  at  4°C in  a
                 refrigerator located in a locked  limited-access  area.

          10.2.3 To avoid photodecomposition,  protect  samples from light.

     10.3 Sample Handling

     CAUTION:  Finely divided soils  contaminated with 2,3,7,8-TCDD are hazardous
              because of the potential for inhalation or ingestion  of particles
               containing 2,3,7,8-TCDD.  Such  samples  should be handled  in a
              confined environment  (i.e., a closed  hood or  a glove box).

          10.3.1 Pre-extraction sample treatment


                                      D-13
                                         (-"! n';'
                                         vJ U (

-------
                 10.3.1.1 Homogenization — Although sampling personnel will
                          attempt to collect homogeneous samples, the contrac-
                          tor  shall examine each sample and judge if it needs
                          further mixing.

                 NOTE:  Contractor personnel have the responsibility to take a
                        representative sample portion; this responsibility
                        entails efforts to make the sample as homogeneous as
                        possible.  Stirring is recommended when possible.

                 10.3.1.2 Centrifugation — When a soil or sediment sample
                          contains an obvious liquid phase, it must be
                          centrifuged to separate the liquid from the solid
                          phase.  Place the entire sample in a suitable centri-
                          fuge bottle and centrifuge for 10 minutes at 2000 rpm.
                          Remove the bottle from the centrifuge.  With a dis-
                          posable pipet, remove the liquid phase and discard
                          it.  Mix the solid phase with a stainless steel
                          spatula and remove a portion to be air-dried, weighed
                          and  analyzed.  Return the remaining solid portion to
                          the  original sample bottle and store it as described
                          in 10.2.1.

                 CAUTION:  The removed liquid may contain TCDD and should be
                           disposed as a liquid waste.

                 10.3.1.3 Weigh between 9.5 and 10.5 g of the air-dried soil
                          sample (^0.5 g) to 3 significant figures.  Dry it to
                          constant weight at 100°C.  Allow the sample to cool
                          in a desiccator.  Weigh the dried soil to 3 signifi-
                          cant figures.  Calculate and report percent moisture
                          on Form B-l.
11.  SAMPLE EXTRACTION
     11.1 Soil Extraction
          11.1.1 Immediately before use, the Soxhlet apparatus is charged
                 with 200 to 250 mL benzene which is then refluxed for 2 hours.
                 The apparatus  is allowed to cool, disassembled and the benzene
                 removed and retained as a blank for later analysis if required.

          11.1.2 Accurately weigh to 3 significant figures a 10-g (9.50 g to
                 10.50 g) portion of the wet soil or sediment sample.  Mix 100
                 uL of the sample fortification solution (Section 7.8) with
                 1.5 mL of acetone (500 pg of   C,2~2,3,7,8-TCDD) and deposit
                 the entire mixture in small portions on several sites on the
                 surface of the soil or sediment.

          11.1.3 Add 10 g anhydrous sodium sulfate and mix thoroughly using a
                 stainless steel spoon spatula.
                                      D-14

-------
     11.1.4 After breaking up any  lumps,  place  the  soil-sodium  sulfate
            mixture in the Soxhlet  apparatus  using  a glass wool plug  (the
            use of an extraction thimble  is  optional).   Add  200 to 250 mL
            benzene to the Soxhlet  apparatus  and  reflux  for  24  hours.  The
            solvent must cycle completely through the  system at least 3
            times per hour.

     11.1.5 Transfer the extract to a Kuderna-Danish apparatus  and
            concentrate to 2  to 3  uL.  Rinse  the  column  and  flask with 5 mL
            benzene and collect the rinsate  in  the  concentrator tube.
            Reduce the volume in the  concentrator tube to  2  to  3 uL.
            Repeat this rinsing and concentrating operation  twice more.
            Remove the concentrator tube  from the K-D  apparatus and care-
            fully reduce the  extract  volume  to  approximately 1  mL with a
            stream of nitrogen using  a flow  rate  and distance such that
            gentle solution surface rippling is observed.

      NOTE:  Glassware used  for more  than one sample must  be carefully
             cleaned between  uses  to  prevent cross-contamination (Note on
             page D-10).

11.2  Extraction of Aqueous Samples

      11.2.1  Mark the water  meniscus on  the side of the 1-L sample bottle
              for later determination of  the  exact  sample  volume.  Pour
              the entire sample (approximately  1  L) into a 2-L  separatory
              funnel.

      11.2.2  Mix 100 uL of the sample fortification solution with 1.5 mL
              of acetone (500 pg of 13C12-2,3,7,8-TCDD)  and  add the mixture
              to the sample in the  separatory funnel.

      NOTE:  A continuous liquid-liquid extractor may  be used in place of
             a separatory funnel.

      11.2.3  Add 60 mL methylene  chloride to the sample bottle, seal and
              shake 30 seconds to  rinse the  inner surface.   Transfer  the
              solvent to the  separatory funnel  and  extract the  sample by
              shaking the funnel for  2 minutes  with periodic venting.
              Allow the organic layer to  separate from the water phase  for
              a minimum of 10 minutes.  If the  emulsion  interface between
              layers is more  than  one-third  the volume of the solvent
              layer, the analyst must employ mechanical  techniques to
              complete the phase separation.  Collect  the methylene
              chloride (3 x 60 mL)  directly  into  a  500 mL Kuderna-Danish
              concentrator (mounted with  a 10 mL  concentrator tube) by
              passing the sample extracts through a filter funnel packed
              with a glass wool plug  and  5 g of anhydrous sodium sulfate.
              After the third extraction, rinse the sodium sulfate with an
              additional 30 mL of  methylene  chloride to  ensure  quantitiative
              transfer.
                                 D-15
                                      609

-------
     11.2.4   Attach a Snyder column and concentrate the extract until
              the apparent volume of the liquid reaches 1 mL.   Remove  the
              K-D apparatus and allow it to drain and cool for at least
              10 minutes.  Remove the Snyder column, add 50 mL benzene,
              reattach the Snyder column and concentrate to approximately
              1 mL.  Rinse the flask and the lover joint with  1 to 2 mL
              benzene.  Concentrate the extract to 1.0 mL under a gentle
              stream of nitrogen.

     11.2.5 Determine the original sample volume by refilling  the sample
            bottle to the mark and transferring the liquid to  a 1000-ml
            graduated cylinder.  Record the sample volume to the nearest
            5 mL.

11.3 Cleanup Procedures - Column Cleanup

     11.3.1 Prepare an acidic silica column as follows:  Pack  a 1 cm x 10
            cm chromatographic column with a glass wool plug,  a layer  (1
            cm) of Na2804/K2CC^d :1) , 1.0 g silica gel (Section 7.1.2) and
            4.0 g of 40-percent w/w sulfuric acid-impregnated  silica gel
            (Section 7.1.3).  Pack a second chromatographic column (1  cm x
            30 cm) with a glass wool plug, 6.0 g acidic alumina (Section
            7.1.1) and top with a 1-ctn layer of sodium sulfate (Section
            7.4).  Add hexane to the columns until they are free of
            channels and air bubbles.

     11.3.2 Quantitatively transfer the benzene extract (1 mL') from the
            concentrator tube to the top of the silica gel column.  Rinse
            the concentrator tube with two 0.5-mL portions of  hexane.
            Transfer the rinses to the top of the silica gel column.

     11.3.3 Elute the extract from the silica gel column with  90 mL hexane
            directly into a Kuderna-Danish concentrator.  Concentrate  the
            eluate to 0.5 mL, using nitrogen blow-down as necessary.

     11.3.4 Transfer the concentrate (0.5 mL) to the top of the alumina
            column.  Rinse the K-D assembly with two 0.5-mL portions of
            hexane and transfer the rinses to the top of the alumina
            columns.  Elute the alumina column with 18 mL hexane until the
            hexane level is just below the top of the sodium sulfate.
            Discard the eluate.  Columns must not be allowed to reach
            dryness (i.e., a solvent "head" must be maintained.)

     11.3.5 Place 30 mL of 20-f-percent (v/v) methylene chloride in hexane
            on top of the alumina and elute the TCDDs from the column.
            Collect this fraction in a 50-mL Erlenmeyer flask.

     11.3.6 Certain extracts, even after cleanup by column chromatography,
            contain interferences which preclude determination of TCDD
            at low parts-per-trillion levels.  Therefore, a cleanup step
            is included using activated carbon which selectively retains
            planar molecules such as TCDD.  The TCDDs are then removed

                                 D-16
                                           GIG

-------
                 from the  carbon  by  elution  with  toluene.  Proceed as  follows:
                 Prepare a 18-percent  Carbopak  C/Celite 545® mixture by
                 thoroughly mixing 3.6 grams Carbopak C (80/100 mesh)  and 16.4
                 grams Celite  545® in  a 40-mL vial.  Activate at 130°C for 6
                 hours.  Store in a  desiccator.   Cut off  a clean 5-mL  disposable
                 glass pipet at the  4-mL mark.  Insert a  plug of glass wool
                 (Section  7.3)  and push to  the  2-mL mark.  Add 340 mg  of  the
                 activated Carbopak/Celite mixture  followed by another glass
                 wool plug. Using two glass rods, push both glass wool plugs
                 simultaneously towards the  Carbopak/Celite mixture and gently
                 compress  the  Carbopak/Celite plug  to a length of 2 to 2.5 cm.
                 Preelute  the  column with 2  mL  toluene followed by 1 mL of
                 75:20:5 methylene chloride/raethanol/benzene, 1 mL of  1:1
                 cyclohexane in methylene chloride, and 2 mL hexane.   The flow
                 rate should be less than 0.5 mL  min."*.  While the column is
                 still wet with hexane, add  the entire eluate (30 mL)  from the
                 alumina column (Section  11.3.5)  to the top of the column.
                 Rinse the Erlenmeyer  flask  which contained the extract twice
                 with 1 mL hexane and  add the rinsates  to the top of the  column.
                 Elute the column sequentially  with two 1-mL aliquots  hexane, 1
                 mL of 1:1 cyclohexane in methylene chloride, and 1 mL of
                 75:20:5 methylene chloride/ methanol/benzene.  Turn the  column
                 upside down and  elute the TCDD fraction  with 6 mL toluene into
                 a concentrator tube.   Warm  the tube to approximately  60°C and
                 reduce the toluene  volume  to approximately 1 mL using a  stream
                 of nitrogen.   Carefully  transfer the residue into a 1-mL
                 mini-vial and again at elevated  temperature, reduce the volume
                 to about  100  uL  using a  stream of  nitrogen.  Rinse the concen-
                 trator tube with 3  washings using 200 uL of 12 toluene in
                 CH2C12-   Add  50  uL  tridecane and store the sample in  a refrig-
                 erator until  GC/MS  analysis is performed.

12.  ANALYTICAL PROCEDURES

     12.1 Remove the sample extract  or blank from storage, allow  it to warm  to
          ambient laboratory temperature and add  5  uL recovery standard solution.
          With a stream of dry, purified  nitrogen,  reduce the extract/blank
          volume to 50 uL.

     12.2 Inject a 2-uL aliquot of the extract  into the GC, operated under the
          conditions previously used (Section 8.1)  to produce acceptable  results
          with the performance check solution.

     12.3 Acquire SIN data according to 12.3.1.   Use the  same acquisition time
          and MS operating conditions  previously  used (Section 8.3.4)  to  deter-
          mine the relative response factors.

     12.3.1  Acquire SIM data  for  the following  selected characteristic ions:
                                      D-17
                              611

-------
                                       Compound

                                       TCDD - COC1

                                       Unlabeled TCDD

                                       Unlabeled TCDD

                                       13C12-2,3,7,8-TCDD,  13C12-1,2,3,4-TCDD

                                       13C12-2,3,7,8-TCDD,  13C12-1,2,3,4-TCDD


     12.4 Identification Criteria

          12.4.1 The retention time (RT) (at maximum peak height)  of the  sample
                 component m/z 319.897 must be within -1  to «3 seconds  of the
                 retention time of the peak for the isotopically  labeled  internal
                 standard at m/z 331.937 to attain a positive identification of
                 2,3,7,8-TCDD.  Retention times of other  tentatively identified
                 TCDDs must fall within the RT window established by analyzing
                 the column performance check solution (Section 8.1).   Retention
                 times are required for all chromatograms.

          12.4.2 The ion current responses for m/z 258.930, 319.897 and 321.894
                 must reach maximum simultaneously (+_ 1 scan), and  all  ion
                 current intensities must be 2. 2.5 times  noise level for
                 positive identification of a TCDD.

          12.4.3 The integrated ion current at m/z 319.897  must be between 67
                 and 90 percent of the ion current response at m/z 321.894.

          12.4.4 The integrated ion current at m/z 331.937  must be between 67
                 and 90 percent of the ion current response at m/z 333.934.

          12.4.5 The integrated ion currents for m/z 331.937 and  333.934  must
                 reach their maxima within +_ 1 scan.

          12.4.6 The recovery of the internal standard 13C12~2,3,7,8-TCDD must
                 be between 40 and 120 percent.

13.   CALCULATIONS

     13.1 Calculate the concentration of 2,3,7,8-TCDD (or any other TCDD  isoraer)
          using the formula:

                                           Ax '
                                  cx »
                                       AIS • W • RRF(I)
                                      D-18
                                               812

-------
where:
     GX  *  unlabeled 2,3,7,8-TCDD (or  any  other  unlabeled TCDD  isomer) concen-
            tration in pg/g for  soil/sediment  and pg/L  for aqueous  samples.

     AX  =  sum of the integrated  ion abundances  determined  for  m/z 319.897
            and 321.894.

     Ajg =  sum of the integrated  ion abundances  determined  for  m/z 331.937
            and 333.934 of   C12~2,3,7,8-TCDD  (IS »  internal standard).

     QIS -  quantity (in picograms)  of  13Cj2~2,3,7,8-TCDD added  to  the
            sample before extraction (Qxs ° 500 pg).

      W  =  weight (in grams)  of dry soil or sediment sample or  volume of
            aqueous sample (in liters).
 RRF(I)  »  calculated mean relative response  factor  for  unlabeled  2,3,7,8TCDD
            relative to 13C,2-2,3,7,8-TCDD.  This  represents  the grand
            mean of the RRF(I)'s  obtained  in Section  8.3.4.5.

     13.2  Calculate the recovery  of the internal  standard   3Cj2~2,  3,7,8-TCDD,
           measured in the sample  extract, using the  formula:

           Internal standard           A,g  * Q__
           percent recovery    *   ———^——————  x  100
                                  ARS '  RRF(II)  '  Q1S.

where Ajg and Q^g have the same definitions  as above  (Section  13.1)
         =  sum of the integrated ion abundances  determined  for m/z  331.937
            and 333.934 of   C12~l,2,3,4-TCDD (RS =  recovery standard).

    QRS  =  quantity (in picograms)  of  3Cj2~1,2,3,4-TCDD added to  the  sample
            residue before HRGC-HRMS  analysis.

    (QRS " 500 pg).

RRF(II)  =  calculated mean relative  response factor for  labeled 3Cj2-2,3,7,8-
            TCDD relative to 13C12-1,2,3,4-TCDD.   This  represents the grand
            mean of the RRF(II)'s calculated  in Section 8.3.4.5.

     13.3  If the calculated concentration  of unlabeled 2,3,7,8-TCDD exceeds
           200 pg/g for soils or sediments, or 2000  pg/L  for aqueous samples,
           the linear range of response  vs. concentration may  have been  exceeded
           and a smaller portion of  that sample must be analyzed.  Accurately
           weigh to three significant figures a 1-g  portion  of the wet  soil/
           sediment.  Add the sample  fortification solution  (Section 11.1.2),
           extract and analyze as discussed for the  10-g  sample. Similarly,
           add the sample fortification  solution  (Section 11.2.2) to 100 mL of
           the aqueous sample, extract and  analyze.

                                      D-19
                                   G13

-------
     13.4  Total TCDD concentration — all positively identified isoraers of TCDD
           must be within the RT window and meet all identification criteria
           listed in Sections 12.4.2, 12.4.3 and 12.4.4.  Use the expression
           in Section 13.1 to calculate the concentrations of the other TCDD
           isomers, with C^ becoming the concentration of any unlabeled TCDD
           i s one r.

c Total TCDD  -  Sum of the concentrations of the individual TCDDs.

     13.5  Estimated Detection Limit — For samples in which no unlabeled
           2,3,7,8-TCDD was detected, calculate the estimated minimum detectable
           concentration.  The background area is determined by integrating the
           ion abundances for m/z 319.897 and 321.894 in the appropriate region
           of the selected ion monitoring trace, multiplying that area by 2.5,
           and relating the product area to an estimated concentration that
           would produce that product area.

           Use the formula:

                                          (2.5) • (Ax) ' (QIS)
where
                                         (AIS) • (RRF(I)) • (W)
         *  estimated concentration of unlabeled 2,3,7,8-TCDD required to
            produce Ax.

         *  sum of integrated  ion abundance for m/z 319.897 and 321.894 in the
            same group of >5 scans used to measure
            8um °f integrated ion abundance for the appropriate ion character-
            istic of the  internal standard, m/z 331.937 and m/z 333.934.

QjS» RRF(I), and W retain the definitions previously stated in Section 13.1.
Alternatively, if peak height measurements are used for quantification, measure
the estimated detection limit by the peak height of the noise in the TCDD RT
window.

     13.6  The relative percent difference (RPD) is calculated as follows:

                        I  Sj - S2 |           | Si - S2 |
             RPD  -  — — — — — — —  -  - x  100
                     Mean Concentration       (Si + 82) 12

          Si and $2 represent sample and duplicate sample results.

References

1.  "Carcinogens - Working with Carcinogens", Department  of Health, Education
     and Welfare, Public  Health Service, Center for Disease Control, National
     Institute for Occupational Safety and Health, Publication No. 77-206, Aug.
     1977.
                                      D-20
                                              G14

-------
2.   "OSHA Safety and Health  Standards, General Industry" (29 CFR1910),
     Occupational Safety  and  Health Administration, OSHA 2206 (Revised January
     1976).

3.   "Safety in Academic  Che'nistry Laboratories", American Chemical Society
     Publication, Committee on  Chemical Safety, 3rd Edition 1979.
                                      D-21
                          Giro

-------
TABLE 1.  COMPOSITION OF CONCENTRATION CALIBRATION SOLUTIONS


HRCC1
HRCC2
HRCC3
HRCC4
HRCC5
Recovery Standard
13C12-1,2,3,4-TCDD
2.5 pg/uL
5.0 pg/uL
10.0 pg/uL
20.0 pg/uL
40.0 pg/uL
Analyte
2,3,7,8-TCDD
2.5 pg/uL
5.0 pg/uL
10.0 pg/uL
20.0 pg/uL
40.0 pg/uL
Internal Standard
13C12-2,3,7,8-TCDD
10.0 pg/uL
10.0 pg/uL
10.0 pg/uL
10.0 pg/uL
10.0 pg/uL
               Sample Fortification  Solution

              5.0 pg/uL of 13C12-2,3,7,8-TCDD
             Recovery Standard  Spiking Solution

                100 pg/uL 13C12-1,2,3,4-TCDD



            Field Blank Fortification Solutions

          A)  5.0 pg/uL of unlabeled 2,3,7,8-TCDD

          B)  5.0 pg/uL of unlabeled 1,2,3,4-TCDD



             Internal Standard  Spiking Solution

              100 pg/uL of 13C1?-2,3,7,8-TCDD
             (Used only in Section 4.2.1.A.2, Exhibit E)
                             D-22
                                   GiG

-------
                 TABLE 2.   RECOMMENDED GC  OPERATING CONDITIONS
Column coating

Film thickness

Column dimensions

Helium linear velocity


Initial temperature

Initial time

Temperature program
2,3,7,8-TCDD retention
time
SP-2330

0.2 urn

60 m x 0.24 mm

28-29 cm/sec
at 240eC

70°C

4 min

Rapid increase to 200°C
200°C to 250°C
at 4°C/min

24 min
CP-SIL 88

0.22 urn

50 m x 0.22 mm

28-29 cm/sec
at 240°C

45°C

3 min

Rapid increase to 190°C
190°C to 240°C
at 5°C/min

26 min
                                      D-23
                            6 1'V

-------
TABLE 3.  TYPICAL 12-HOUR SEQUENCE FOR 2,3,7,8-TCDD ANALYSIS
1.  Static mass  resolution check




2.  Column performance  check




3.  HRCC2




4.  Sample 1 through Sample "N"




5.  Column performance  check




6.  Static mass  resolution check
                                      10/20/84




                                      10/20/84




                                      10/20/84




                                      10/20/84




                                      10/20/84




                                      10/20/84
0700 hrs.




0730 hrs.




0800 hrs.




0830 hrs.




1800 hrs.




1830 hrs.
                            D-24


-------
                        IM
r-*
                        I*
                                                         2J7I
                                                                                                 CT-SIL N
                                                                          11)4
                                                                                           127t
                                                                                                          I2CI
                                        IN
                                                            IM
                                                                                                   IM
                 Figure 1.  Selected  ion  current profile for ra/r 320 and 322 produced  by  MS analysis for
                            performance check  solution using a 50-m CP Si 1-88 fused  silica  capillary
                            column  and conditions listed in Table 2.

-------
             IM
J
o>
              u


              u
                                            2371
                                                           I2M
                                                                           IIM
                      IM
» I ' '
•M
                                          tit
                                                     440
                                                                                             •M
     Figure 2.  Selected  ion  current  profile  for m/z  320  and  322  produced by MS analysis of performance

                check solution using  a 60-m SP-2330 fused  silica  capillary column  and  conditions

                listed  in Table  2.

-------
                                   APPENDIX  B

                          PROPOSED  ANALYTICAL  PROTOCOL

                            for  the Determination  of
          2,3,7,8-Tetrachlorodibenzo-p-dioxin  (2,3,7,8-TCDD) and Total
              TCDDs in Soil/Sediment  and Water by  High-Resolution
              Gas Chromatography/High-Resolution Mass Spectrometry
                                December  1,  1985


     This analytical  protocol  has  been  written  in the  format  used  in the
Superfund program, as "Exhibit D"  of  a  Statement of Work which in turn is part
of an Invitation-for-Bid package under  the Superfund Contract Laboratory Program.
Also included are other exhibits listed below for the  Statement of Work which
have been tailored to meet the specific requirements of this  protocol:

EXHIBIT B:   Reporting Requirements and  Deliverables
EXHIBIT C:   Sample Rerun Requirements
EXHIBIT D:   Analytical  Method
EXHIBIT E:   Quality Assurance/Quality Control Requirements
                                        821

-------
     This  protocol  (Protocol  B)  is a modification rf the protocol presented as
Appendix A  (Protocol A).  Examination of the results from the single-laboratory
evaluation  of Protocol A had  shown that the minimum amount of 2,3,7,8-TCDD that
could be quantified under the conditions specified in Protocol A was 5 pg.
However, a  requirement existed to lower the quantitat ion limits to 2 ppt for
soil and sediment samples and to 20 ppq for aqueous samples.  The sample size
should stay at 10 g for soil  and sediments and at 1 L for aqueous samples,
since the effect of larger sample sizes on the extract cleanup efficiencies is
not known.  Also, the range of the method should overlap with the 1-ppb lower
limit of the low-resolution analytical method for TCDD used in the Superfund
Contract Laboratory Program without necessitating second extractions for samples
containing higher levels of TCDDs.

     After careful evaluation by EMSL-LV of the requirements and the options,
the following protocol changes were made:

     *  In Protocol B, the following calibration solutions will be used:
          HRCC1:    2 pg/uL   2,3,7,8-TCDD and 13C12-1,2,3,4-TCDD
                   10 pg/uL   13Cl2-2,3,7,8-TCDD
          HRCC2:   10 pg/wL   2,3,7,8-TCDD and 13C12-1,2,3,4-TCDD
                   10 pg/uL   13Cl2-2,3,7,8-TCDD
          HRCC3:   50 pg/pL   2,3,7,8-TCDD and 13C12-1,2,3,4-TCDD
                   10 pg/uL   13Cl2-2,3,7,8-TCDD

-------
     HRCC4:   100 pg/yl_   2,3,7,8-TCDD  and  13C12-1,2,3,4-TCDD
              10 pg/yL   13C12-2,3,7,8-TCDD

o  In Protocol  B, the final  extract  volume will be  10 yL.  The decision
   to select a final  volume  of 10 yL was necessary  in order to comply
   with the  above requirements.   It  is  realized that such a small volume
   may pose  technical difficulties for the analyst.

0  In Protocol  B, the fortification  level  of the  internal standard
   13C12-2,3,7,8-TCDD was raised  from  500  pg/sample to  1,000 pg/sample.
   This allows analysis of soil  and  sediment samples containing  between
   100 ppt and 1.2 ppb of any TCDD isomer  and of  water  samples containing
   between 1 ppt and  12 ppt  of any TCDD isomer by diluting a 2-yL aliquot
   of the remaining extract  concentrate by a factor of  12 with a solution
   of the recovery standard  (100  pg/yL of  13C12-1,2,3,4-TCDD in  tridecane)
   Recoveries will be reported using the data generated from the first
   injection.  Thus,  the decision to dilute  an aliquot  of the 10-yL final
   extract will  not be based on the  concentration of 2,3,7,8-TCDD or total
   TCDD in the sample, but on the concentration of  the  most abundant TCDD
   isomer in the 10-yL final  extract volume.  This  will  eliminate un-
   necessary dilutions of the sample extract and  analyses for samples
   containing between 100 ppt and 250  ppt  for soil  and  sediment  and 1 ppt
   and 2.5 ppt for water samples  of  a  TCDD isomer,  but  for which the
   recoveries were low.
                                    623

-------
              EXHIBIT B
Reporting Requirements and Deliverables
                   G24

-------
1.    SCOPE AND APPLICATION

     The Contractor shall provide  reports  and  other  deliverables as specified
     in the Contract  Reporting  Schedule.   These reports are described below.
     All reports shall  be submitted  in  legible form  or resubmission shall be
     required.  All reports  and  documentation  required, including selected ion
     current profiles (also  called selected  ion monitoring traces), shall be
     clearly labeled  with the Sample  Management Office Case number and associated
     Sample/Traffic Report number(s).   If  documentation is submitted without
     the required identification,  as  specified above, resubmission shall be
     required.

     The Contract Reporting  Schedule  (Section  2)  specifies the  numbers of
     copies required, the delivery schedule  and the  distribution of all required
     deliverables.

     1.1  Sample data package — Hard copy analytical data and  documentation
          are required  as described below.

     NOTE:  This analytical  protocol  is designed  for the  receipt and analysis
            of samples  by batches. Therefore, it is desired  that sample data
            from samples  in  the same  batch be  reported together, i.e., on the
            same reporting form.  However, contract  accounting  and billing are
            based on the  sample unit.

          1.1.1   Case narrative:  Contains the Case  number, Dioxin Shipment
                 Record numbers, Contract  number  and detailed documentation of
                 any  quality control, sample,  shipment and/or analytical pro-
                 blems  encountered in a specific  Case.  Also  included should be
                 documentation  of  any internal decision tree  process used along
                 with a summary of corrective  actions taken.  The Case narrative
                 must be  signed in original signature by  the  Laboratory Manager
                 or his designate.

          1.1.2   Results  of  initial  triplicate analyses of four (4) concentration
                 calibration solutions  (Form H-2), routine calibration solutions,
                 (Form  H-3), including  all selected  ion current profiles or
                 selected ion monitoring (SIM) traces, calculated relative
                 response  factors  (RRF),  and computer-generated quantification
                 reports  (or manual calculations).

          1.1.3   Completed data reporting  sheets  (Forms H-l,  H-4, and H-5,  H-8
                 and  H-9) with  appropriate SIM traces (including the lock mass
                 SIM traces).   Data  results for  levels less  than 10 ppt but
                 above  the quantitation limit  (Section 1.1, Exhibit D) attained
                 for  that sample shall  be  reported to two (2) significant
                 figures; results  greater  than 10 ppt shall be  reported to  three
                 (3)  significant figures.   Apply  the rounding rules  found  in
                 Section  7.2.2,  "Handbook  for  Analytical  Quality Control in Water
                 and  Wastewater Laboratories," EPA-600/4-79-019.  Each SIM  trace
                 shall  include  computer-generated header  information indicating
                 instrumental (GC  and MS)  operating  parameters  during data
                                      B-l
                                           G25

-------
            acquisition.  When samples are analyzed more Chan once,  all
            sample data shall be reported.  Rejected sample runs must be
            separated and attached to the back of the data package and
            marked on the SIM trace as "Rejected," with an explanation of
            the reasons for the rejection.

     1.1.A  SIM traces generated during each GC column performance check
            analysis; peak profile outputs of the reference signal used
            to document the mass resolution.

     1.1.5  Documentation of acceptable MS calibration (Section 8, Exhibit
            D, and Exhibit E) for each confirmatory analysis.  As
            applicable, submit peak matching box settings and calculations
            for accurate mass assignments and any other related printouts.
            State, in ppm, the level of mass accuracy achieved (Section
            8.2.2, Exhibit D).

     1.1.6  A chronological list of all analyses performed (Form H-6).  If
            more than one GC/MS system is used, a chronological list is
            required for each system.  The list must provide the Data
            System File name, the EPA sample number, and (if appropriate)
            the contractor laboratory sample number for each sample,
            blank, concentration calibration solution, performance check
            solution, or other pertinent analytical data.  This list shall
            specify date and time of beginning of analysis.  All sample/
            blank analyses performed during a 12-hour period, must be
            accompanied by two GC column performance check solution ana-
            lyses, one preceding and one following the sample/blank ana-
            lyses.  If multiple shifts are used, the ending GC column
            performance check sample analysis from one 12-hour period
            shall serve as the beginning analysis for the next 12-hour
            period; see Exhibit D, Section 8, for system performance
            criteria.  The same schedule applies to the mass resolution
            check analysis.  See Section 8.2.2, Exhibit D.

     1.1.7  Verification of recovery of TCDDs from cleanup columns
            (Section 11.3, Exhibit D, and Section 4.2.1.2.2, Exhibit E).

1.2  Sample extracts and unused sample portions — Unused portions of
     samples and sample extracts shall be retained by the Contractor for
     a period of six months after receipt.  When directed in writing by
     the Project Officer (PO) or Sample Management Office (SMO), the
     Contractor shall ship (not at Contractor's expense but in accordance
     with Department of Transportation Regulations) specific samples
     and/or extracts to specified locations and persons.  After six months,
     upon obtaining PO or SMO clearance, remaining samples and extracts
     shall be disposed of by the Contractor at Contractor's expense, in
     accordance with applicable regulations concerning the disposal of
     such materials.

1.3  Document Control and Chain-of-Custody Package — The Document Control
     and Chain-of-Custody Package includes all laboratory records received


                                 B-2

-------
          or generated for a specific  case,  that  have  not  been previously
          submitted  to EPA as  a  deliverable.   These  items  include  but  are not
          limited to:   sample  tags,  custody  records,  sample  tracking records,
          analysts logbook pages,  bench  sheets, chromatographic charts,  computer
          printouts,  raw data  summaries,  instrument  logbook  pages, corre-
          spondence,  and the document  inventory (Exhibit  G).

     NOTE:  Pages from logbooks  or bench sheets kept  exclusively in a  high-
            hazard area (containment facility) need  not be copied.

     1.4  Monthly Sample Status  Report — The Monthly  Sample  Status Report
          shall provide the status of  all samples the  Contractor has received
          or has had in-house  during the calendar month.   Required status
          information includes:   samples received, samples extracted,  samples
          analyzed,  samples rerun, and samples which required special  cleanup.
          All samples  shall be identified by  the  appropriate  EPA sample, case
          and batch/shipment numbers.

     1.5  Daily Sample Status  Report — In response  to a  verbal request  from the
          Sample Management Office or  the Project Officer, the Contractor must
          verbally provide sample status information on a same-day basis.
          Should written confirmation  be requested,  the  Contractor must  send the
          daily sample status  information in a written form  that same  day using
          first-class  mail service.   The required Daily  Sample Status  informa-
          tion shall include the items noted  for  the Monthly  Sample Status
          Report and,  in addition, shall require  information  on sample analysis
          reports in progress  and analysis reports submitted/mailed.

2.   In accordance with applicable delivery  requirements,  the Contractor shall
     deliver specified items per the following Contract  Reporting  Schedule
     (Section 2.1).   Recipients  include  the  CLP Sample Management  Office, the
     EMSL/LV QA Division, the  appropriate Regional Technical  Officer and NEIC.

     2.1  Contract Reporting Schedule

                          CONTRACT REPORTING SCHEDULE

Item                             Delivery            	Report Distribution	
 No.  Report	No.  Copies	Schedule	SMO   EMSL/LV  Region   NEIC

 1  Sample Data     3      30 days after  validated     XX        X
     Package                sample receipt date
                                    -OR-
                            10 days after initial     XXX
                               data due  date

2   Sample Extracts         Within 180  days after             As directed
                           analysis, 7 days  after
                          request by Project  Officer
                                  or SMO

                                                                  (Continued)

                                      B-3
                         627

-------
                    CONTRACT REPORTING SCHEDULE (Continued)
Item
No.
Report
No. Copies
Delivery
Schedule
Report Distribution
SMO
EMSL/LV
Region
NEIC
    Document       1
     Control &    Pkg
     Chain-of-
     Custody Package

    Monthly        2
     Sample Status
     Report
                           7 days after request by
                               Project Officer
                                  or SMO
                           5 days following end of
                             each calendar month
    Daily Sample
     Status Report
                          Verbal and/or written
                        upon request by SMO or PO;
                        maximum frequency is daily.
As directed
NOTE:  All results shall be  reported total and complete.
     2.2  Addresses for distribution
          SMO
                                           EMSL-LV
        NEIC
CLP Sample Management Office
P. 0. Box 818
Alexandria, VA 22313
For overnight deliveries, use
 street address:
300 N. Lee St., Suite 200
Alexandria, VA 22314
                                  US EPA EMSL-LV QA Division
                                  Box 15027
                                  Las Vegas, NV 89114
                                   Attn: Data Audit Staff
   US EPA NEIC
   Bldg. 53
   Box 25227
   Denver Fed. Center
   Denver,  CO 80225
                                  For overnight deliveries, use
                                   street address:
                                  944 E. Harmon Ave.
                                  Executive Center
                                  Las Vegas, NV 89109
3.
          Regional Technical Officer — Following contract award and prior to
          Contractor's receipt of the first batch of samples, the Sample Manage-
          ment Office will provide the Contractor with the list of Technical
          Officers for the ten EPA Regions.  SMO will provide the Contractor
          with updated Regional address/name lists as necessary throughout the
          period of the contract.

     FORM INSTRUCTION GUIDE

     This section includes specific instructions for the completion of all
     required forms.  These include instructions on header information as
     well as specific details to the bodies of individual forms.  Instructions
     are arranged in the following order:
                                      B-4

-------
Data Summary (Form H-l)
Initial Calibration Summary (Form H-2 ;  2  pages)
Routine Calibration Summary (Form H-3)
GC and Mass Resolution Check Summary (Form H-4)
Quality Control Summary (Form H-5)
Chronological List of All Analyses Performed  (Form H-6)
GC Operating Conditions (Form H-7)
HRMS TCDD Calibration Report Form (Form H-8)
High-Resolution MS TCDD Data Report Form  (Form H-9)

3.1  Data Summary (Form H-l) — This form  is used  for  summarizing  the
     results from all samples in the batch.  The  detailed  results  are
     available on Form H-8 for each sample.

     Complete the header information at the top  of the page,  including
     laboratory name, case number and  batch/shipment  number (from the
     dioxin shipment record), and matrix  (soil,  sediment,  water).

     Complete the form using one horizontal row  for each sample.

     The SMO sample number should be suffixed with the appropriate letter
     code as needed.

     The TCDD retention time should be  reported  in minutes and  seconds.

     TCDD levels are reported as parts  per trillion (ppt)  regardless  of
     the matrix.  Total TCDD concentration (in ppt) is the sum  of  the
     concentrations of all TCDDs reported on  Form H-9;

     The S/N criteria apply to m/z 259, 320,  322  (for unlabeled TCDD)
     and m/z 322 and 334 (internal and  recovery  standards).  The  symbols
     used are:  (+) all S/N ratios are  2.5 or greater including all TCDDs
     present, (-) S/N ratio for native  2 ,3,7 ,8-TCDD,  the internal  or  the
     recovery standard are less than 2.5, (0) other suspected TCDDs are
     present but did not meet the S/N criteria.

     The file name is the HRGC/HRMS file  name and is  used  for tracking
     results and raw data.

     The comments column should be used for any  remarks  specific  to a
     particular sample.

3.2  Initial Calibration Summary (Form  H-2):   Page 1  —  The header infor-
     mation should be filled in.  The  column  headings are  similar  to  those
     on Form H-l.
                Ax '
     RRF(I)  =

                Q  '  A
                      IS
                                 B-5
                 G29

-------
     RRF(II)  »  	       (Section 8.3.4.5, Exhibit D)
                 QlS  ' ARS

     Page 2 — The header information should be filled in.  For each RRF,
     the mean, percent relative standard deviation (%RSD) and number of
     runs (N) are reported; N must be at least three (3) for each HRCC
     solution.  The grand means (RRFs) are the mean of the individual
     means and are reported with their XRSD and N.  The routine calibra-
     tion relative response factor permissible ranges are also reported
     (Section 8.3.4.8, Exhibit D).

3.3  Routine Calibration Summary (Form H-3) — The header information
     Includes case and batch numbers in addition to the laboratory and
     instrument identification.

     The columns are  the same as on Page 1 of Form H-2.  The results
     reported are for the routine calibration runs rather than the initial
     calibration.  The calculated RRF(I) and RRF(II) must be within the
     routine calibration relative response factor permissible ranges
     (Section 8.3.4.8, Exhibit D) and other criteria listed in Section
     8.6, Exhibit D must be met before further analysis is performed.

3.4  GC and Mass Resolution Check Summary (Form H-4) — The header informa-
     tion should be filled in.  The TCDD isoner resolution (% valley) is
     measured from the column performance check solution (Section 8.1.2,
     Exhibit D).  The resolving power and mass measurement error are measured
     using PFK (or equivalent) (Section 8.2, Exhibit D).

3.5  Quality Control  Summary (Form H-5) — The items should be completed
     as indicated.  The "other Interferences" should be included even if
     they only occur  at one mass.

     Form H-5 in conjunction with Form H-9 is used to report results
     relative to the  fortified field blank pair and rinsate analyses.

     The total TCDD retention time window is a window that includes all of
     the TCDD isomers and is based on the first and last eluting isomers
     in the GC column performance check solution using the conditions sum-
     marized in Form  H-7.  All materials used should be recorded in the
     standard/reagent QC table.  Standards provided by EPA should be
     listed, however, the QC columns may be left blank as these are refer-
     ence materials.

3.6  Chronological List of All Analyses Performed (Form H-6) — The
     header information should be filled in.  If more than one instrument
     is used, use one form per instrument.

     The "Analysis Identification" column should contain enough information
     for the data user to clearly identify the analysis, i.e., HRCC 2
     Routine Calibration, Fortified Field Blank A, Fortified Field Blank B,


                                 B-6

-------
          Reanalysis of Sanple //I, 2, 3, 4, etc.   The "SMO //" column should be
          used only for samples etc. which have an assigned SMO sample number.

     3.7  GC Operating Conditions (Form H-7) — This form must be filled out to
          describe the GC operating conditions used to analyze a batch of
          samples and to analyze the GC performance evaluation check solution.

     3.8  HRMS TCDD Calibration Report (Form H-8) — This form is to be filled
          in for each initial and routine calibration analysis made.  It will be
          the first page of the chrociatograms and calculations for that analysis.
          It is suggested that this form be used as a worksheet for completing
          Forms H-2 and H-3.  S/N ratios greater than five (5) may be reported
          with a (+); S/N ratios of five or less must have a numerical value
          reported with accompanying chromatograms scaled so that the measure-
          ments may be checked by the data user.

     3.9  High-Resolution MS TCDD Data Report (Form H-9) — This form contains
          the details of the data reported in summary on Form H-l.  It will be
          the first page of the chrociatograms and calculations for each sample
          including the fortified field blank pair samples.  All data presented
          (retention times, areas, and S/N ratios) must also be available on
          the accompanying chromatograms.  The chromatograms must be scaled
          so that the data user may check any S/N ratios that are near or below
          five (5).  It is suggested that this form be used as a worksheet for
          completing Form H-l.

4.   REPORTING REQUIREMENTS SUMMARY:

     Items that must be included with the data package:

     4.1  Complete identification of the samples analyzed (sample numbers and
          type).

     4.2  The dates and times at which all analyses were accomplished.  This
          information should also appear on each selected ion current profile
          included with the report.

     4.3  Raw mass chromatographic data which consist of the absolute peak
          heights or peak areas of the signals observed for the ion masses
          monitored.

     4.4  The calculated ratios of the intensities of the M+° to (M+2)+°
          molecular ions for all TCDD isoroers detected.

     4.5  The calculated concentrations of native 2,3,7,8-TCDD and other TCDD
          isomers for each sample analyzed, expressed in picograms TCDD per gram
          of sample (that is, parts per trillion), as determined from the raw
          data.  If no TCDDs are detected, the notation "Not Detected" or
          "N.D." is used, and the minimum detectable concentrations (or detection
          limits) are reported.
                                      B-7
                                                   G 31

-------
                                                             HIGH  RESOLUTION
                                                         FORM  H-l  DATA SUMMARY
                                                       HRGC/HRMS  DIOXIN  ANALYSIS
          Lib:
 Case*
         Bitcn/Shipnwiit It.
                            Matrix:
8MO
Number















TCOD
2.3.7.8 (IS)






























PPt
2.3.7.8-TCDD
Mea*. DL






























Relative Ion
opt Abundance Ratio*
Total 32fl 332 J2i %Rac. 8/N Inat. Ane»y*t* Ha
TCOD 322 334(18) 334(RS) (IS) Criteria ID Data Time Name Comment*





































































































































































w
 I
00
           RB - H>«g«nl Blank
            N- Untobetod TCDD Spfta
            D - Duplicate
           FB- FWd Blank
           8R • Sampte Reran
           ER - Extract Reanahr*!*
NO- NotDatectwl
 DL- Detection Umit
MB- Method Blank
Roc • Recovery
Matrix:  S - Soil  '
       W- Water
       O- Other
S/N Criteria: report (») rf aN S/N  > 2.6
           report (-) If 2.3,7.8 TCDO.
                   "C,f 2,3.7,8 TCDO or
                   "C1f-1.2,3.4-TCOOS/N < 2.5
           report (0) if other TCOD* are auepected
                   not to meet criteria

-------
                                                                  HIGH RESOLUTION
                                                     FORM  H-2  INITIAL CALIBRATION SUMMARY
                                                                                                                             pa«alo<2
                 Lab:
               Contract *:
Instrument ID:
       i
       VO
 o
Co
Calibration File m/i m/t m/i S/N
Standard Name Data Tlma 320/322* 332/334(18)* 332/334(RS)* Criteria RRF(l|b RRF|II)C Cornnenta






































































































































































































                  * Ion ratios mud be in the range of 0.67 to 0.90
                  b 2.3.7.8-TCOD veraut "C,j-2.3.7.8 TCDD
                       S/NCritaria: report |+| if greater than 2.6
                                 report (•) rl toM than 2.6
                    "C,, 2.3.7.8 TCDD wriu* '
'C,2 1.2.3.4 TCOD

-------
                                                                   HIGH RESOLUTION
                                                      FORM  H-2  INITIAL CALIBRATION SUMMARY
                                                                                                                paoa2o<2
                                lab:
                                Date of Inthial Calibration:
                      Contract #:
           Instrument ff:
                                                RRF (I) Mean    %RSD
                                               RRF (II) Maan      % RSO
                                  HRCC1

                                  HRCC2

                                  HRCC3

                                  HRCC4
        eo
         i
RRF (I) Grand Maan:

%RSD: 	

N: 	
RRF (II) Grand MOOT:

HR80: 	

N: 	
                                Routine Calibration Pa
CT;
CO
                                    RRF (l| = 2.3.7.8 TCDD v«
                                    "€,,-2.3.7.81000
            aiblaRanga:
                                         RoutinaCattN
                    Ma Rang*:
                                            RRF (111 = "C  -2.3.7.B-TCDD wt
                                            »C.,-1.2.3.4-TCOD

-------
                                                                  HIGH RESOLUTION
                                                     FORM H-3 ROUTINE CALIBRATION SUMMARY
                Lab:
      Case ft:
Batch ff:
Instrument ID:
C73
CO
Ol
      w
       i
Calibration File m/t m/t m/i S/N
Standard Him* Date Time 320/322* 332/334(15)* 332/334C,, 1.2.3.4 TCOO

-------
                                 HIGH RESOLUTION
               FORM  H-4  GC AND MASS RESOLUTION CHECK SUMMARY
Ub:
                   C«M*
                                         Batch #
  Oat*
bi«t.
 ID
Sol.
 IO
Ton*
          TCDO Iwrrwr    RMorving         *MUw
 File        RMolution      Power       Measurement
Nam*       (%VaU*y)   atlOSValtoy     Error (PPM)
•Me** ueed for HIM* menurement i
                                       B-12

-------
                                        HIGH  RESOLUTION
                            FORM H-5  QUALITY  CONTROL SUMMARY
Lab:
           Case*
                                                                     Batch #
Number of sample* in batch: _

Mean S of recovery for the I.S.:
                                   0 of data points:
Fortified field blank A. S recovery ("C,2-2.3.7.8-TCDD): 	
                                          No   Yes
                                          D   D
                                                                       SMO Sample »:
         Contamination by 1.2.3.4-TCOO
                   "C12-1.2.3.4-TCDD
                                                             Estimated
                                                             Concentration (ppt)
         Other interferences:
                          Retention times:
                          Estimated concentrations (ppt):


Fortified field blank B. * recovery (taC12-1.2.3.4-TCDD): -
                                          SMO Sample «:
                   "C12-2.3.7.8-TCDD
                                                BYes
                                                [j
                                                             Estimated
                                                             Concentration (ppt)
         Other interferences:
                          Retention times:
                          Estimated concentrations (ppt):
Rinsate. % recovery:
                          SMO sample «:
         Contamination by 2.3.7.8-TCDD

                          Other TCOO


Duplicate anarysis. SMO sample «:
                                                             Estimated
                                                             Concentration
                                                             (pg/mL)
          	  "C12-2.3.7.8-TCDD Mean Recovery:

Percent Relative difference "C12-2.3.7.8-TCDD (Recovery) 	

Percent relative difference t*C12-2.3.7.8-TCDD (Concentration) _

Percent relative difference Total TCOO (Concentration)	

HRMS Lab.
Standard /Reagent Number or Origin Date of O.C File Result* of
Type Mfg. « QC Name QC






















































                                               B-13


                                                   63 7

-------
                              HIGH RESOLUTION


        FORM  H-6 CHRONOLOGICAL LIST  OF ALL ANALYSES  PERFORMED
Lab:
Instrument ID:
   Case*
Batch
     Nam*
  Analyci*

Identification
                                               SMO

                                              Number
       O*tt
Time
                                    B-14
                                         /••> ••> o
                                         boo

-------
                             HIGH RESOLUTION
                    FORM H-7  GC OPERATING CONDITIONS
Lab:	 Instrument ID:
GC Column: 	
Film Thickness:
Column Dimensions:
Initial Column Temperature:
Temperature Program:  	
Injector Temperature: 	
Interface Temperature:  	
Injection Mode: 	
Injection Volume:
SplMess Valve Closed Time:
Septum Purge Flow: 	
Injector Sweep Flow:	
Carrier Gas Flow Rate (ml/min or cm/sec):
                        -QQ      B-15

-------
                           HIGH RESOLUTION
            FORM H-8  HRMS TCDD CALIBRATION REPORT FORM
Lab:
Case*:
Batch/Shipment #:
Instrument ID:  	
    Calibration:
       Initial
       Routine
2.3.7.8-TCDD
   m/z 258.930
       319.897
       321.894
'»C,,-2.3.7.8-TCDD
   m/z 331.937
       333.934
"C12-1.2.3.4-TCDD
   m/z 331.937
   m/z 333.934
                Calibration Solution:
               	 GC Column: 	
                    Date of Initial Calibration:
         Analysis Date:
                 Time:
                       File Name
Retention
  Time
Area
Ratios
                                322
                                332,
                                334
                                332
                                334
(•I If S/N i« flraatar than 6. antar ( +); if Ian than S. anar th« maaaurad ratio
                                   B-16
                                              G-iO

-------
                                   HIGH  RESOLUTION
             FORM  H-9 HIGH RESOLUTION  MS TCDD DATA REPORT FORM
l.h
fmm* a

l^ttrunwt ID'
Aliquot
MM «,W«I
SMO Surnpta 0
FiU M.rrut-
M*tri»- lA/ffflr

Circle
One Percent
g L Moittiir*


So" <>•»»•*
Tim.-
GC Co'timp . ., ,
P.trx^inn O»f«
Muturod ppt 2.3.7.8-TCDD
Estimatad Total TCDD (ppt):
2.3.7.8-TCDD
     m/z 258.930
         319.897
         321.894
                                              Detection Umrt 2.3.7.8-TCDD:
                                              	 Report Date: 	
                       Retention
                         Time
                                             Area
                                                               Ratios
S/N'
"C12-2.3.7.8-TCDD
     m/z 331.937
         333.934

"C12-1.2,3.4-TCDD
     m/z 331.937
     m/z 333.934
Percent Recovery "C12-2.3.7.8-TCDD:
                                                          322
                                                          332
                                                          334
                                                          332
                                                          334
                                         Other TCDD*
Estimated
Retention 320 S/N* 259 S/N* 320 S/N* 322 Concentration
Time 322 (ppt)












































































































*H S/N is greater than 5. enter (•»); rf leu than 6 enter the measured ratio
                                           B-17
                                    G41

-------
4.6  The same raw and calculated data which are provided for the actual
     samples will also be reported for the duplicate analyses, the method
     blank analyses, the fortified field blank pair and rinsate analyses,
     and any other QA or performance sample analyzed in conjunction with
     the actual sample set(s).

4.7  The recoveries of the internal standard (  Cj2-2,3,7,8-TCDD) in percent.

4.8  The calibration data, including relative response factors calculated from
     the calibration procedure described in Section 8.3, Exhibit D.  Data
     showing that these factors have been verified at least once during each
     12-hour period of operation must be included (Section 8.5, Exhibit D).
     Exact mass measurement error.  Include peak matching box settings
     and calculations as appropriate.

4.9  The calculated dry weight of the original soil or sediment sample portion
     based on the dry weight determination of another sample portion of approxi-
     mately equal wet weight.  The exact volumes of the water and rinsate
     samples analyzed.

4.10 Documentation of the source of all TCDD standards used and available
     specifications on purity.

4.11 In addition, each report of analyses will include the following- selcted
     ion current profiles:  1) those obtained from all samples analyzed, 2)
     those from each GC colunn performance check, and 3) those from the
     calibration solutions.  The peak profile from each mass resolution
     check must also be part of the data package.

4.12 Identify which HRGC/HRMS system was used for the analyses (manufacturer
     and laboratory identification number of system -01, 02, 03, etc.).

4.13 GC operating conditions such as type of GC column, film thickness, column
     dimensions, initial column temperature, temperature program, injector
     temperature, interface temperature, injection mode and volume, valve time
     (valve flush), septum purge flow, flow rate, and total injector flow
     should be provided (Form H-7).
                                 B-18

-------
       EXHIBIT C
Sample Rerun Requirements
              G43

-------
1.   SCOPE AND APPLICATION

     The Contractor shall be required to reextract and/or perform additional
     cleanup and reanalyze certain samples or batches of samples in a variety
     of situations that may occur in the process of contract performance.
     (For purposes of this contract, the term "sample rerun" shall indicate
     sample extraction of a fresh 10-g soil or sediment portion or 1-L aqueous
     sample, followed by cleanup and analysis, and the term "extract reanalysis"
     shall indicate analysis of another aliquot of the final extract.

     In situations where the sample rerun is required due to matrix effects,
     interferences or other problems encountered because of very complex samples,
     the Government will pay the Contractor for the sample reruns.  Such sample
     reruns shall be billable and accountable under the specified contract
     allotment of automatic reruns.

     In situations where the sample rerun or the extract reanalysis is required
     due to Contractor materials, equipment or instrumentation problems, or
     lack of contractor's adherence to specified contract procedures, the
     sample rerun or extract reanalysis shall not be billable under the terms
     of the contract.

     Contractor's failure to perform any of the sample reruns or extract re-
     analyses specified herein, either billable or non-billable, shall be
     construed as Contractor nonperf ormance and may result in termination of
     the contract for default by the Contractor.

2.   Required Sample Reruns and Extract Reanalyses

     2.1  Automatic sample reruns and extract reanalyses that may be billable
          as such under the contract.

          2.1.1  If the percent recovery for the internal standard   C~2, 3,7,8-
                 TCDD was outside of the acceptance limits of >bO percent and
                 ^120 percent,  the Contractor shall reextract and reanalyze the
                 sample.   If the percent recovery for the sample rerun is still
                 outisde  the acceptance limits, then both analyses can be billed
                 if the recoveries from both analyses are either 120%.
                 If, however, the percent recovery for the sample rerun is
                 within the acceptance limits, or if it is still outside the
                 acceptance limits but the percent recoveries from the original
                 analysis and the sample rerun are not both either <40% or
                 >120%, then the sample rerun may not be billed.

          2.1.2  If the internal standard was not found upon monitoring m/z
                 331.937  and 333.934,  the Contractor shall reextract and
                 reanalyze the  sample.  If the internal standard is not
                 found in the sample rerun, the' sample rerun is billable.  If
                 the internal standard is found in the sample rerun, then the
                 sample rerun is not billable.
                                      C-l

-------
     2.1.3  If either one of the isotope abundance ratios for m/z 319.897/
            321.894 or for 331.937/333.934 is less than 0.67 or greater
            than 0.90 and all other criteria contained in Section 12.4 of
            Exhibit D are met, then the extract shall be reanalyzed.  If
            both ion abundance ratios now meet the criterion, these values
            shall be reported as the isotope abundance ratios, and the
            Contractor shall not bill the Government for the extract
            reanalysis.  If the ratio in question is still outside the
            criterion, the Contractor shall rerun the sample (Section 7.2,
            Exhibit E).  If either one of the ratios determined from the
            sample rerun is still outside the acceptance limits, then
            both runs and the extract reanalysis can be billed if the
            corresponding isotope abundance ratios from both runs are
            either <0.67 or >0.90.  If, however, both isotope abundance
            ratios from the sample rerun meet the criteria, or if both
            corresponding isotope abundance ratios from the original run
            and the sample rerun are not both either <0.67 or >0.90,
            then the extract reanalysis and the sample rerun may not be
            billed.

     2.1.4  If the recoveries of 2,3,7,8-TCDD (Section 4.2.1.1.3.1,
            Exhibit E) and/or 1,2,3,4-TCDD (Section 4.2.1.2, Exhibit E)
            in the fortified field blank pair are <40% or >120%, the
            Contractor shall reextract and reanalyze a second portion of
            the field blank sample (Section 4.2, Exhibit E).  If the
            percent recoveries for the sample rerun are still outside the
            acceptance limits, then both analyses can be billed as long
            as the recoveries from both analyses are either <40% or >120%.
            If, however, the percent recoveries for the sample rerun
            are within the acceptance limits, or if they are still outside
            the acceptance limits but the percent recoveries from the
            original run and the sample rerun are not both either <40%
            or >120%, then the sample rerun may not be billed.

            NOTE:  Fortified field blanks as described in Sections
                   4.2.1.1.4 and 4.2.1.2.2, Exhibit E, can never be billed.

2.2  Automatic sample extract dilution and HRGC/HRMS analysis, billable as
     such under the Contract.

     If any individual or group of coeluting TCDD isomer concentrations in
     the 10-uL final extract exceeds 100 pg/uL, the analyst will perforn a
     dilution as specified in Section 13.3, Exhibit D, and reanalyze the
     diluted portion using HRGC/HRMS.


2.3  Sample reruns and/or extract reanalyses to be performed at Contractor's
     expense (i.e., not billable under the terns of the contract).

     2.3.1  If the method blank contains any signal in the TCDD retention
            time window at or above the method quantitation limit (2 ppt


                                 C-2


                                            G45

-------
       for soil and sediment  and  20 ppq  for aqueous  samples),  the
       Contractor shall rerun all positive  samples  in the  batch  of
       samples (Section A.1.2,  Exhibit E).

2.3.2  If the system performance  using the  GC column performance
       check (PC) solution  does not meet specified  criteria,  the
       Contractor shall take  corrective  action,  demonstrate acceptable
       GC column performance, and reanalyze the  extracts from all
       positive samples run during the time period  between the last
       acceptable PC run and  the  unacceptable PC run (Section 2.4,
       Exhibit E).

2.3.3  If a false positive  is reported for  an uncontaminated soil
       (blind QC) sample, upon notification by the  Sample  Management
       Office the Contractor  shall reextract and reanalyze all samples
       reported as positive in the associated batch  of samples
       (Section 8.1.1,  Exhibit E).

2.3.4  If the analysis  results for a performance evaluation blind QC
       sample fall outside  of EPA-established acceptance windows, upon
       notification of  the  Sample Management Office  the Contractor
       shall reextract  and  reanalyze the entire  associated batch
       of samples (Section  8.4.1, Exhibit E).

2.3.5  If the isotope abundance ratio for m/z 319.897/321.894 or for
       331.937/333.934  is  less than 0.67 or greater  than 0.90, and
       all other criteria  contained in Section 12.4  of Exhibit D are
       met, then the extract  shall be reanalyzed'.  If the ion abundance
       ratio in question now meets the criterion, this value shall  be
       reported as the  isotope abundance ratio,  and  the Contractor
       shall not bill the Government for the extract reanalysis.

2.3.6  If the system performance  mass resolution check does not  meet
       the specified criterion, the Contractor shall take corrective
       action, demonstrate  acceptable mass  resolution and reanalyze
       the extract from all positive samples analyzed during the time
       period between the  last acceptable mass resolution check  and
       the unacceptable mass  resolution  check (Section 2.4,
       Exhibit E).
                            C-3

-------
                         EXHIBIT D
                     Analytical Method

2,3,7,8-Tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD)  and Total
  TCDDs in Soil/Sediment and Water by High-Resolution Gas
      Chrotnatography/High-Resolution Mass Spectrometry
                               P 4 7
                               ^ i (

-------
                                   EXHIBIT D
Section                       Subject                              Page
   1                Scope and Application	     D-l




   2                Summary of Method	     D-l




   3                Definitions	     D-2




   4                Interferences	     D-3




   5                Safety	     D-3




   6                Apparatus and Equipment	     D-4




   7                Reagents and Standard Solutions.  ......     D-6




   8                System Performance Criteria	     D-9




   9                Quality Control Procedures 	     D-1A




  10                Sample Preservation and  Handling  	     D-1A




  11                Sample Extraction	,     D-15




  12                Analytical Procedures	     D-18




  13                Calculations	     D-19

-------
1.   SCOPE AND APPLICATION

     1.1  This method provides procedures for the detection and  quantitative
          measurement of 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD; CAS
          Registry Number 1746-01-6;  Storet  number 3475)  at concentrations  of
          2 pg/g (2 parts per trillion) to 100 pg/g (100  parts per trillion)
          in 10-g portions of soil and sediment and at  20 pg/L (20 parts  per
          quadrillion) to 1000 pg/L (1 part  per trillion) in 1-L samples  of
          water.  Dilution of an aliquot of  the final extract permits  measure-
          ment of concentrations up to 1.2 ng/g (1.2 parts per billion)  or  12
          ng/L (12 parts per trillion), respectively.  This method also  allows
          the estimation of quantities of total TCDD present in  the  sample.
          Samples containing concentrations  of any individual TCDD isomer or
          group of coeluting TCDD isomers greater than  1.2 ng/g  or 12  ng/L  must
          be analyzed by a protocol designed for such concentration  levels,
          with an appropriate instrument calibration range.

     1.2  The minimum measurable concentration is estimated to be 2  pg/g  (2
          parts per trillion) for soil and sediment samples and  20 pg/L  (20
          parts per quadrillion) for water samples, but this depends on  kinds
          and concentrations of interfering compounds in  the sample  matrix.

     1.3  This method is designed for use by analysts who are experienced in
          the use of high-resolution gas chromatography/high-resolution  mass
          spectrometry.

     CAUTION:  TCDDs are assumed to be extremely hazardous.  It  is the labora-
               tory's responsibility to ensure that safe.handling procedures are
               employed.

2.   SUMMARY OF METHOD

     One thousand pg of    C,2-2,3,7,8-TCDD (internal standard) are added  to a
     10-g portion of a soil/sediment sample  (weighed to 3 significant  figures)
     or a 1-L aqueous sample, and the sample is extracted with 200 to  250 mL
     benzene using a Soxhlet apparatus for soils and sediments with  a  minimum
     of 3 cycles per hour, or with methylene chloride using a continuous  liquid-
     liquid extractor for aqueous samples for 24 hours.  A separatory  funnel
     and 3 x 60 mL methylene chloride may also be used  for aqueous samples.
     After appropriate cleanup, 10 uL of a tridecane solution of the recovery
     standard (  C12~l»2,3,4-TCDD) are added to the extract which is then
     concentrated to a final volume of 10 uL.  One to three uL of the  concen-
     trated extract is injected into a gas chromatograph  with a  capillary
     column interfaced to a high-resolution  mass spectrometer capable  of  rapid
     multiple ion monitoring at resolutions  of at least 10,000 (10 percent
     valley).

     Identification of 2,3,7,8-TCDD is based on the detection of the ions n/z
     319.897 and 321.894 at the same GC retention time  and within -1 to  +3
     seconds GC retention time of the internal standard masses of m/z  331.937
     and 333.934.  Confirmation of 2,3,7,8-TCDD (and of other TCDD isomers) is
                                      D-l
                                          '
                                          '1
                                          •->

-------
     based on the ion m/z 258.930 which results from loss of COCL by the parent
     molecular ion.

3.   DEFINITIONS
                    i
     3.1  Concentration calibration solutions — solutions containing known
          amounts of the analyte (unlabeled 2,3,7,8-TCDD), the internal standard
          I3C12-2,3,7,8-TCDD and the recovery standard 13C,2~1.2,3,4-TCDD;
          they are used to determine instrument response of the analyte
          relative to the internal standard and of the internal standard
          relative to the recovery standard.

     3.2  Field blank — a portion of soil/sediment or water uncontaminated with
          2,3,7,8-TCDD and/or  other TCDDs.

     3.3  Rinsate — a portion  of solvent used to rinse sampling equipment; the
          rinsate is analyzed  to demonstrate that samples have not been contami-
          nated during sampling.

     3.4  Internal standard —  13C12-2,3,7,8-TCDD, which is added to every
          sample (except the blank described in Sections 4.2.1 of Exhibit E)
          and is present at the same concentration in every method blank and
          quality control sample.  It is added to the soil/sediment or aqueous
          sample before extraction and is used to measure the concentration of
          each analyte.  Its concentration is measured in every sample, and
          percent recovery is  determined using an internal standard method.

     3.5  Recovery standard —  ^ Cl2~l»2»3,4-TCDD which is added to every sample
          extract (except for  the blank discussed in Sections 4.2.1, Exhibit E)
          just before the final concentration step and HRGC-HRMS analysis.

     3.6  Laboratory method blank — this blank is prepared in the laboratory
          through performing all analytical procedures except addition of a
          sample aliquot to the extraction vessel.

     3.7  GC cdlumn performance check mixture — a mixture containing known
          amounts of selected  standards; it is used to demonstrate continued
          acceptable performance of the capillary column, i.e., separation
          (_< 25% valley) of 2,3,7,8-TCDD isomer from all other 21 TCDD isomers,
          and to define the TCDD retention time window.

     3.8  Performance evaluation sample — a soil, sediment or aqueous sample
          containing a known amount of unlabeled 2,3,7,8-TCDD and/or other
          TCDDs.  It is distributed by the EMSL-LV to potential contractor lab-
          oratories who must analyze it and obtain acceptable results before
          being awarded a contract for sample analyses (see IFB Pre-Award Bid
          Confirmations).  It  may also be included as an unspecified ("blind")
          QC sample in any sample batch submitted to a laboratory for analysis.
     3.9
Relative response factor — response of the mass spectrometer to a
known amount of an analyte relative to a known amount of an internal
standard.
                                      D-2

-------
     3.10 Mass resolution check — standard method used to demonstrate static
          resolution of 10,000 minimum (10% valley definition).

     3.11 Positive response for a blank — defined as a signal in the TCDD
          retention time window, at any of the masses monitored, which is
          equivalent to or above the method quantitation limit (2 ppt for soil
          and sediment, and 20 ppq for aqueous samples).

     3.12 Sample rerun — extraction of another 10-g soil or sediment sample
          portion or 1-L aqueous sample, followed by extract cleanup and
          extract analysis.

     3.13 Extract reanalysis — analysis of another aliquot of th final extract.

4.   INTERFERENCES

     Chemicals which elute from the GC column within _+10 scans of the internal
     and/or recovery standard (m/z 331.937 and 333.934) and which produce within
     the TCDD retention time window ions at any of the masses used to detect or
     quantify TCDD are potential interferences.  Most frequently encountered
     potential interferences are other sample components that are extracted
     along with TCDD, e.g. PCBs, chlorinated tnethoxybiphenyls, chlorinated
     hydroxydiphenylethers, chlorinated benzylphenylethers, chlorinated naphtha-
     lenes, DDE, DDT, etc.  The actual incidence of interference by these
     chemicals depends also upon relative concentrations, mass spectrometric
     resolution, and chromatographic conditions.  Because very low levels of
     TCDDs must be measured, the elimination of interferences is essential.
     High-purity reagents and solvents must be used and all equipment must be
     scrupulously cleaned.  Blanks (Exhibit E, Quality Control,  Section 4) must
     be analyzed to demonstrate absence of contamination that would interfere
     with TCDD measurement.  Column chromatographic procedures are used to
     remove some coextracted sample components; these procedures must be
     performed carefully to minimize loss of TCDDs during attempts to increase
     their concentration relative to other sample components.

5.   SAFETY

     The toxicity or carcinogen!city 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 laboratory is responsible for maintaining a file of
     current OSHA regulations regarding the safe handling of the chemicals
     specified in this method.  A reference file of material data handling
     sheets should also be made available to all personnel involved in the
     chemical analysis.  Additional references to laboratory safety are identi-
     fied O-3) (page D-21).  2,3,7,8-TCDD has been identified as a suspected
     hunan or mammalian carcinogen.  The laboratory is responsible for ensuring
     that safe handling procedures are followed.
                                      D-3
                                  P r~  1
                                  b u 1

-------
6.   APPARATUS AND EQUIPMENT

     6.1  High-Resolution Gas Chromatograph/High-Resolution Mass
          Spectrometer/Data System (HRGC/HRMS/DS)

          6.1.1  The GC must be equipped for temperature programming, and all
                 required accessories must be available, such as syringes, gases,
                 and a capillary column.  The GC injection port must be designed
                 for capillary columns.  The use of splitless injection tech-
                 niques is recommended.  On-column injection techiques can be
                 used but this may severely reduce column lifetime for
                 nonchemically bonded columns.  When using the method in this
                 protocol, a 2-uL injection volume is used consistently.  With
                 some GC injection ports, however, 1-uL injections may produce
                 improved precision and chromatographic separation.  A 1- to 3-uL
                 injection volume may be used if adequate sensitivity and
                 precision can be achieved.

          NOTE:  If 1 uL or 3 uL is used at all as injection volume, the injec-
                 tion volumes for all extracts, blanks, calibration solutions
                 and the performance check sample must be 1 uL or 3 uL.

          6.1.2  Gas Chromatograph-Mass Spectrometer Interface

                 The GC-MS interface may include enrichment devices, such as a
                 glass jet separator or a silicone membrane separator, or the
                 gas chromatograph can be directly coupled to the mass spectrome-
                 ter ion source.  The interface may include a diverter valve
                 for shunting the column effluent and isolating the mass spec-
                 trometer ion source.  All components of the interface should
                 be glass or glass-lined stainless steel.  The interface com-
                 ponents should be compatible with 300°C temperatures.  The
                 GC/MS interface must be appropriately designed so that the
                 separation of 2,3,7,8-TCDD from the other TCDD isomers which
                 is achieved in the gas chromatographic column is not appreci-
                 ably degraded.  Cold spots and/or active surfaces (adsorption
                 sites) in the GC/MS interface can cause peak tailing and peak
                 broadening.  It is recommended that the GC column be fitted
                 directly into the MS ion source.  Graphite ferrules should be
                 avoided in the GC injection port since they may adsorb TCDD.
                 Vespel™ or equivalent ferrules are recommended.

          6.1.3  Mass Spectrometer

                 The static resolution of the instrument must be maintained at
                 a minimum 10,000 (10 percent valley).  The mass spectrometer
                 must be operated in a selected ion monitoring (SIM) mode with
                 total cycle time (including voltage reset time) of one second
                 or less (Section 8.3.4.1).  At a minimum, the following  ions
                 which occur at these masses must be monitored:  m/z 258.930,
                 319.897, 321.894, 331.937 and 333.934.


                                      D-4

-------
          6.1.4  Data System

                 A dedicated hardware or data system is employed to control the
                 rapid multiple ion monitoring process  and to acquire the data.
                 Quantification data (peak areas or peak heights) and SIM traces
                 (displays of intensities of each m/z being monitored as a
                 function of time) must be acquired during the analyses.
                 Quantifications may be reported based  upon computer-generated
                 peak areas or upon measured peak heights (chart recording).

NOTE:  Detector zero setting must allow peak-to-peak measurement of the noise
       on the base line.

     6.2  GC Columns

          For isomer-specific determinations of 2,3,7,8-TCDD, the following
          fused silica capillary columns are recommended:  a 60-m SP-2330 (SP-
          2331) column and a 50-m CP-Sil 88 column.  However, any capillary
          column which separates 2,3,7,8-TCDD from all  other TCDDs may be used
          for such analyses, but this separation must be demonstrated and
          documented.  Minimum acceptance criteria must be determined per
          Section 8.1.  At the beginning of each 12-hour period (after mass
          resolution has been demonstrated) during which sample extracts or
          concentration calibration solutions will be analyzed, column operating
          conditions must be attained for the required  separation on the column
          to be used for samples.  Operating conditions known to produce accept-
          able results with the recommended columns are shown in Table 2 at the
          end of this Exhibit.

     6.3 ' Miscellaneous Equipment

          6.3.1  Nitrogen -evaporation apparatus with variable flow rate.

          6.3.2  Balance capable of accurately weighing to +0.01 g.

          6.3.3  Centrifuge capable of operating at 2,000 rpm.

          6.3.4  Water bath — equipped with concentric ring cover and capable
                 of being temperature-controlled within ^2°C.

          6.3.5  Stainless steel spatulas or spoons.

          6.3.6  Stainless steel (or glass) pan large enough to hold contents
                 of 1-pint sample containers.

          6.3.7  Glove box.

          6.3.8  Drying oven.

     6.4  Glassware

          6.4.1  Soxhlet apparatus — all-glass, Kontes 6730-02 or equivalent;

                                      D-5


                           G53

-------
                 90 mm x 35 ram glass thimble; 500-mL flask; condenser of appro-
                 priate size*

          6.4.2  Kuderna-Danish apparatus — 500-mL evaporating flask, 10-mL
                 graduated concentrator tubes with ground-glass stoppers,  and
                 3-ball macro Snyder column (Kontes K-570001-0500, K-503000-
                 0121 and K-569001-0219 or equivalent).

          6.4.3  Mini-vials — 1-mL borosilicate glass with conical-shaped
                 reservoir and screw caps lined with Teflon-faced silicone disks.

          6.4.4  Funnels — glass; appropriate size to accommodate filter
                 paper used to filter jar extract (volume of approximately 170 mL).

          6.4.5  Separator/, funnel — 2000 mL with Teflon stopcock.

          6.4.6  Continuous liquid-liquid extractors equipped with Teflon or
                 glass connecting joints and stopcocks requiring no lubrication
                 (Hershberg-Wolf Extractor - Ace Glass Company, Vineland,  NJ;
                 P/N 6841-10 or equivalent).

          6.4.7  Chromatographic columns for the silica and alumina chroma'
                 tography — 1 cm ID x 10 cm long and 1 cm ID x 30 cm long.

          6.4.8  Chromatographic column for the Carbopak cleanup — disposable
                 5-raL graduated glass pipets, 6 to 7 mm ID.

          6.4.9  Desiccator.

          6.4.10 Glass rods.

          NOTE:  -Reuse of glassware should be minimized to avoid the risk of
                 cross contamination.  All glassware that is reused must be
                 scrupulously cleaned as soon as possible after use, applying
                 the following procedure.

                 Rinse glassware with the last solvent used in it then with
                 high-purity acetone and hexane.  Wash with hot water containing
                 detergent.  Rinse with copious amounts of tap water and several
                 portions of distilled water.  Drain, dry and heat in a muffle
                 furnace at 400°C for 15 to 30 minutes.  Volumetric glassware
                 must not be heated in a muffle furnace, and some thermally
                 stable materials (such as PCBs) may not be removed by heating
                 in a muffle furnace.  In these two cases, rinsing with high-
                 purity acetone and hexane may be substituted  for muffle-furnace
                 heating.  After the glassware is dry and cool, rinse with hexane,
                 and store  inverted or capped with solvent-rinsed aluminum foil
                 in a clean environment.

7.   REAGENTS AND STANDARD  SOLUTIONS

     7.1  Column Chromatography Reagents

                                      D-6

-------
     7.1.1  Alumina,  acidic — extract  the alumina in a  Soxhlet  with
            methylene chloride for 6  hours (minimum of 3  cycles  per hour)
            and activate it by heating  in a foil-covered glass container
            for 24 hours at 190°C.

     7.1.2  Silica gel — high-purity grade,  type 60, 70-230  mesh;  extract
            the silica gel in a Soxhlet  with  methylene chloride  for 6  hours
            (minimum  of 3 cycles per  hour) and activate  it  by heating  in  a
            foil-covered glass container for  24  hours at  130°C.

     7.1.3  Silica gel impregnated with  40 percent (by weight) sulfuric
            acid — add two parts (by weight) concentrated  sulfuric acid
            to three  parts (by weight)  silica gel (extracted  and activated),
            mix with  a glass rod until  free of lumps, and store  in  a
            screw-capped glass bottle.

     7.1.4  Sulfuric  acid, concentrated  — ACS grade, specific gravity 1.84.

     7.1.5  Graphitized carbon black  (Carbopack C or equivalent), surface
            of approximately 12 m^/g, 80/100  mesh *-- mix thoroughly 3.6
            grams Carbopak C and 16.4 grams Celite 545®  in  a  40-mL  vial.
            Activate  at 130°C for six hours.   Store in a desiccator.

     7.1.6  Celite 545®, reagent grade,  or equivalent.

7.2  Membrane filters or filter paper with pore size of  ^25 urn;  rinse  with
     hexane before use.

7.3  Glass wool, silanized — extract with methylene chloride and hexane
     and air-dry before use.

7.4  Desiccating Agents

     7.4.1  Sodium sulfate — granular,  anhydrous; before use, extract it
            with methylene chloride for  6 hours (minimum of 3 cycles  per
            hour) and dry it for >4 hours in  a shallow tray placed  in an
            oven at 120°C.  Let it cool  in a  desiccator.

     7.4.2  Potassium carbonate—anhydrous, granular; use as  such.

7.5  Solvents — high purity, distilled  in glass:  methylene  chloride,
     toluene, benzene, cyclohexane, methanol, acetone, hexane; reagent
     grade:  tridecane.

7.6  Concentration calibration solutions (Table 1) — four  tridecane
     solutions containing   Cj2~l»2,3,4-TCDD  (recovery standard) and
     unlabeled 2,3,7,8-TCDD at varying  concentrations, and  ^Cj2-2,3,7 ,8-
     TCDD (internal standard, CAS RN  80494-19-5) at a constant concentration
     must be used to  calibrate the instrument.  These concentration calibra-
     tion solutions must be obtained  from the Quality Assurance  Division,
     US EPA, Environmental Monitoring Systems Laboratory (EMSL-LV),  Las
     Vegas, Nevada.  However, additional secondary standards  may be obtained

                                 D-7

-------
     from commercial sources, and solutions may be prepared in the con-
     tractor laboratory.  Traceability of standards must  be verified
     against EPAsupplied standard solutions.  Such procedures will be
     documented by laboratory SOPs as required in IFB Pre-award Bid Con-
     firmations, part 2.f.(4).  It is the responsibility  of the laboratory
     to ascertain that the calibration solutions received are indeed  at the
     appropriate concentrations before they are injected  into the instrument,

NOTE:  Serious overloading of the instrument may occur if the concentration
       calibration solutions intended for a low-resolution MS are injected
       into the high-resolution MS.

     7.6.1  The four concentration calibration solutions  contain unlabeled
            2,3,7,8-TCDD and labeled * C12~l»2,3,4-TCDD at nominal concen-
            trations of 2.0, 10.0, 50.0,  and 100 pg/uL, respectively, and
            labeled  3Cj2~2»3,7,8-TCDD at a constant nominal concentration
            of 10.0 pg/uL.

     7.6.2  Store the concentration calibration solutions in 1-mL mini-
            vials at 4°C.

7.7  Column performance check mixture — this solvent less mixture must be
     obtained from the Quality Assurance  Division, Environmental Monitoring
     Systems Laboratory, Las Vegas, Nevada, and dissolved by the Contractor
     in 1 mL tridecane.  This solution will then contain  the following
     components [including TCDDs (A) eluting closely to 2,3,7,8-TCDD, and
     the first- (F) and last-eluting (L)  TCDDs when using the columns
     recommended in Section 6.2] at a concentration of 10 pg/uL of each of
     these isomers:
          Analyte
     Unlabeled 2,3,7,8-TCDD

     13C12-2,3,7,8-TCDD

     1,2,3,4-TCDD (A)

     1,4,7,8-TCDD (A)

     1,2,3,7-TCDD (A)

     1,2,3,8-TCDD (A)

     1,3,6,8-TCDD (F)

     1,2,8,9-TCDD (L)
Approximate Amount Per Ampule

          10 ng

          10 ng

          10 ng

          10 ng

          10 ng

          10 ng

          10 ng

          10 ng
7.8  Sample fortification solution — an isooctane solution containing
     the internal standard at a nominal concentration of 10 pg/uL.

                                 D-8
                                             G5G

-------
     7.9  Recovery standard spiking solution — a  tridecane  solution contain-
          ing the recovery standard at  a nominal concentration of  10 pg/uL.
          Ten uL of this solution will  be spiked into each sample  extract
          (except for the fortified field blank A) before the final concentration
          step and HRGC/HRMS analysis.   It is also used for  the dilution of  the
          extracts from samples with high TCDD levels (Section 13.3, Exhibit D).
     7.10 Internal standard spiking solution — a tridecane  solution containing
          the internal standard (   C^2~2,3,7,8-TCDD)  at  a nominal  concentration  of
          10 pg/uL.  Ten uL of this solution will be  added to a fortified field
          blank extract (Section 4.2.1.1,  Exhibit E).   This  is the only case
          where   Cj22,3,7,8-TCDD is used  for recovery purposes.

     7.11 Field blank fortification solutions — isooctane solutions containing
          the following TCDD isomers:
          Solution A:
          Solution B:
10.0 pg/uL of unlabeled 2,3,7,8-TCDD
10.0 pg/uL of unlabeled 1,2,3,4-TCDD.
8.   SYSTEM PERFORMANCE CRITERIA

     System performance criteria are presented below.   The laboratory may use
     any of the recommended columns described in Section 6.2.   It  must be
     documented that all applicable system performance criteria specified in
     Sections 8.1, 8.2, 8.3 and 8.5 have been met before analysis  of any sample
     is performed.  Table 2 provides recommended conditions that can be used to
     satisfy the required criteria.  Table 3 provides  a typical 12-hour analysis
     sequence.  The GC column performance and mass resolution  checks must be
     performed at the beginning and end of each 12-hour period of  operation.

     8.1  GC Column Performance

          8.1.1  Inject 2 uL (Section 6.1.1) of the column performance check
                 solution (Section 7.7) and acquire selected ion monitoring
                 (SIM) data for m/z 258.930, 319.897,  321.894, 331.937 and
                 333.934 within a total cycle time of  <1 second (Section
                 8.3.4.1).

          8.1.2  The chromatographic peak separation between 2,3,7,8-TCDD and
                 the peaks representing any other TCDD isomers must  be resolved
                 with a valley of <25 percent, where
                                Valley Percent
                            (x/y)(100)
                 x  =  measured  as  in Figure 1

                 y  =  the peak  height of 2,3,7,8-TCDD.

                 It  is  the responsibility of the  laboratory  to verify the  con-
                 ditions  suitable  for the appropriate  resolution of 2,3,7,8-TCDD
                                      D-9
                                             657

-------
            from all other TCDD isomers.  The column performance check
            solution also contains the TCDD isomers eluting first and last
            under the analytical conditions specified in this protocol
            thus defining the retention time window for total TCDD determi-
            nation.  The peaks representing 2,3,7,8-TCDD and the first and
            the last eluting TCDD isomer must be labeled and identified as
            such on the chromatograms (F and L, resp.).  Any individual
            selected ion current profile or the reconstructed total ion
            current (m/z 259 + m/z 320 + m/z 322) constitutes an acceptable
            form of data presentation.

8.2  Mass Spectrometer Performance

     8.2.1  The mass spectrometer must be operated in the electron (impact)
            ionization mode.  Static resolving power of at least 10,000
            (10 percent valley) must be demonstrated before any analysis
            of a set of samples is performed (Section 8.2.2).  Static
            resolution checks must be performed at the beginning and at
            the end of each 12-hour period of operation.  However, it is
            recommended that a visual check (i.e<>, not documented) of the
            static resolution be made using the peak matching unit before
            and after each analysis.

     8.2.2  Chromatography time for TCDD may exceed the long-term mass
            stability of the mass spectrometer and thus mass drift correc-
            tion is mandatory.  A reference compound [high-boiling
            perfluorokerosene (PFK) is recommended] is introduced into the
            mass spectrometer.  An acceptable lock mass ion at any mass
            between m/z 250 and m/z 334 (m/z 318.979 from PFK is recommended)
            must be used to monitor and correct mass drifts.

     NOTE:  Excessive PFK may cause background noise problems and contami-
            nation of the source resulting in an increase in "downtime"
            for source cleaning.

            Using a PFK molecular leak, tune the instrument to meet the
            minimum required resolving power of 10,000 (10% valley) at
            m/z 254.986 (or any other mass reasonably close to m/z 259).
            Calibrate the voltage sweep at least across the mass range m/z
            259 to m/z 334 and verify that m/z 330.979 from PFK (or any
            other mass close to m/z 334) is measured within +5 ppn (i.e.,
            1.7 mmu, if m/z 331 is chosen) using m/z 254.986 as a reference.
            Documentation of the mass resolution must then be accomplished
            by recording the peak profile of the PFK reference peak m/z
            318.979 (or any other reference peak at a mass close to m/z
            320/322).  The format of the peak profile representation must
            allow manual determination of the resolution, i.e., the hori-
            zontal axis must be a calibrated mass scale (amu or ppm per
            division).  The result of the peak width measurement (performed
            at 5 percent of the maximum which corresponds to the 10%
            valley definition) must appear on the hard copy and cannot
            exceed 100 ppn (or 31.9 mmu if m/z 319 is the chosen reference
            ion).
                                 D-10


                                    p c: o
                                    b u O

-------
8.3  Initial Calibration

     Initial calibration is required  before  any  samples  are  analyzed  for
     2 ,3 ,7,8-TCDD.   Initial calibration  is also  required if  any routine
     calibration does not meet the required  criteria  listed  in Section 8.6.

     8.3.1   All concentration calibration  solutions listed in Table 1 must
            be utilized for the initial  calibration.

     8.3.2   Tune the instrument with  PFK as  described  in Section  8.2.2.

     8.3.3   Inject  2 uL of the column performance check  solution  (Section
            7.7) and acquire SIM mass spectral data  for  m/z  258.930,
            319.897, 321.894, 331.937 and  333.934 using  a total cycle time
            of ^ 1  second (Section 8.3.4.1).  The laboratory must not
            perform any further analysis until it has  been demonstrated
            and documented that the criterion listed  in  Section 8.1.2 has
            been met.

     8.3.4   Using the same GC (Section 8.1)  and  MS (Section  8.2)  conditions
            that produced acceptable  results with the  column performance
            check solution, analyze a 2-uL aliquot of  each of  the 4 concen-
            tration calibration solutions  in triplicate  with the  following
            MS operating parameters.

            8.3.4.1  Total cycle time for  data acquisition must be <_  1
                     second.  Total cycle  time includes  the  sum of all the
                     dwell times and  voltage reset times.

            8.3.4.2  Acquire SIM data for  the following  selected
                     characteristic ions:
m/z
258.930
319.897
321.894
331.937
333.934
Compound
TCDD - COC1
Unlabeled TCDD
Unlabeled TCDD
13C12-2,3,7,8-TCDD,
13C,,-2,3,7,8-TCDD,




13C12-1,2,3,4-TCDD
13C,,-1,2,3,4-TCDD
            8.3.4.3  The ratio of integrated ion current for m/z 319.897  to
                     m/z 321.894 for 2,3,7,8-TCDD must be between 0.67 and
                     0.90.

            8.3.4.4  The ratio of integrated ion current for m/z 331.937  to
                     m/z 333.934 for 13C12-2,3 , 7 ,8-TCDD and  3CJ2-1 ,2 , 3 ,4-
                     TCDD must be between 0.67  and 0.90.
                                 D-ll
                                     P CT r\
                                     bu.J

-------
                 8.3.A.5   Calculate  the relative response factors for unlabeled
                           2,3,7,8-TCDD  (RRF(I)] relative to 13C12~2,3,7,8-TCDD
                           and  for  labeled  13Cl2-2,3,7,8-TCDD  [RRF(II)]  relative
                              13C12-1,2,3,4-TCDD as follows:
              to
                    if,  - - -

                                      Ax * <*IS
                                       RRF(I)
                                                      * AIS
                                                  AIS  " QRS
                                      RRF(II)  =  	
                                                  QIS  * ARS

where

     Ax  »  sun of the integrated ion abundances of m/z 319.897 and m/z 321.894
            for unlabeled  2,3,7,8-TCDD.

     AJS  *  sutn J?f tne integrated  ion abundances of m/z 331.937 and m/z 333.934
            for 13C17-2,3,7,8-TCDD.
    ARS
sum of the integrated ion abundances for m/z 331.937 and n/z
333.934 for   C12-l,2,3,4-TCDD.
    QIS  -  quantity of  13C12-2,3,7,8-TCDD injected (pg).

    QRS  -  quantity of   Cj2-l,2,3,4-TCDD Injected (pg).

     Qx  -  quantity of  unlabeled 2,3,7,8-TCDD injected (pg).

    RRF is a dimensionless quantity;  the units used to express Qj$, QRS
must be the same.

                 8.3.4.6 Calculate  the four means (RRFs) and their respective
                          relative standard deviations (%RSD) for the response
                          factors from each of the triplicate analyses for both
                          unlabeled  and 13C12-2,3,7,8-TCDD (Form H-2).

                 8.3.4.7 Calculate  the grand means RRF(I) and RRF(II) and their
                          respective relative standard deviations (%RSD) using
                          the  four mean RRFs (Section 8.3.4.6) (Form H-2).

                 8.3.4.8 Calculate  the routine calibration permissible range
                          for  RRF(I) and RRF(II) using a ^20% window from the
                          grand  means RRF(I) and RRF(II) (Section 8.3.4.7)
                          (Form  H-2).

     8.4  Criteria for Acceptable Calibration

          The criteria listed  below  for acceptable calibration must be met
          before analysis of any sample is performed.

                                      D-12


                                           GGO

-------
     8.4.1  The percent relative  standard  deviation  (RSD)  for  the  response
            factors from each  of  the  triplicate  analyses  for both  unlabeled
            and   Ci2-2,3,7,8-TCDD must  be less  than 20 percent.

     8.4.2  The variation of the  4 mean  RRFs  for unlabeled and  Ci?"
            2,3,7,8-TCDD obtained from the triplicate analyses must be
            less than 20 percent  RSD.

     8.4.3  SIM traces for 2,3,7,8-TCDD  must  present a signal-to-noise
            ratio of ^2.5 for  m/z 258.930, m/z 319.897 and, m/z 321.894.

     8.4.4  SIM traces for *C12~2,3,7,8-TCDD must present a signal-to-
            noise ratio 2.2-5 for  m/z  331.937  and m/z 333.934.

     8.4.5  Isotopic ratios (Sections 8.3.4.3 and 8.3.4.4) must be within
            the allowed range.

     NOTE:  If the criteria for acceptable calibration listed  in Sections
            8.4.1 and 8.4.2 have  been met, the RRF can be  considered inde-
            pendent of the analyte quantity for  the  calibration concentra-
            tion range.  The mean RRF from 4  triplicate determinations for
            unlabeled 2,3,7,8-TCDD and  for   C12-2,3,7,8-TCDD  will be used
            for all calculations  until routine calibration criteria (Section
            8.6) are no longer met.   At  such  time, new mean RRFs will be
            calculated from a  new set of four triplicate  determinations.

8.5  Routine Calibrations

     Routine calibrations must be performed at  the beginning of a  12-hour
     period after successful mass resolution  and GC  column performance
     check runs.

     8.5.1  Inject 2 uL of the concentration  calibration  solution  which
            contains 10 pg/uL  of  unlabeled 2,3,7.8-TCDD,  10.0  pg/uL
            of 13C12-2,3,7,8-TCDD and 10 pg/uL    C12~l,2,3,4-TCDD.
            Using the same GC/MS/DS conditions  as used in Sections 8.1,
            8.2 and 8.3, determine and document  acceptable calibration as
            provided in Section 8.6.

8.6  Criteria for Acceptable Routine  Calibration

     The following criteria must  be met  before  further analysis  is per-
     formed.  If these criteria are  not  met,  corrective action must be
     taken and the instrument  must be recalibrated.

     8.6.1  The measured RRF for  unlabeled 2,3,7,8-TCDD must be within 20
            percent of the mean values  established (Section 8.3.4.8) by
            triplicate analyses of concentration calibration solutions.

     8.6.2  The measured RRF for  13Ci2-2,3,7,8-TCDD  must  be within 20 per-
            cent of the mean value established by triplicate analysis
            of the concentration  calibration  solutions (Section 8.3.4.8).

                                 D-13

                                   661

-------
          8.6.3  Isotopic ratios (Sections 8.3.4.3 and 8.3.4.4)  must be within
                 the allowed range.

          8.6.4  If one of the above criteria is not satisfied,  a second attempt
                 can be made before repeating the entire initialization process
                 (Section 8.3).

          NOTE:  An initial calibration must be carried out whenever the HRCC 2
                 solution is replaced by a new one from a different lot.

9.   QUALITY CONTROL PROCEDURES

     See Exhibit E for QA/QC requirements.

10.  SAMPLE PRESERVATION AND HANDLING

     10.1 Chain-of-custody procedures — see Exhibit G.

     10.2 Sample Preservation

          10.2.1 When received, each soil or sediment sample will be contained
                 in a 1-pint glass jar surrounded by vermiculite in a sealed
                 metal paint can.  Until a portion is to be removed for analysis,
                 store the sealed paint cans in a locked limited-access area
                 where the temperature is maintained between 25° and 35°C.
                 After a portion of a sample has been removed for analysis,
                 return the remainder of the sample to its original container
                 and store as stated above.

          10.2.2 Each aqueous  sample will be contained in a 1-liter glass
                 bottle.  The bottles with the samples are stored at 4°C in  a *
                 refrigerator  located in a locked limited-access area.

          10.2.3 To avoid photodecomposition, protect samples from light.

     10.3 Sample Handling

     CAUTION:  Finely divided  soils and sediments contaminated with 2,3,7,8-TCDD
               are hazardous, because of the potential for inhalation or ingestion
               of particles containing 2,3,7,8-TCDD.  Such samples should be
               handled in a confined environment (i.e., a closed hood or a
               glove box).

          10.3.1 Pre-extraction sample treatment

                 10.3.1*1 Homogenization — Although sampling personnel will
                          attempt to collect homogeneous samples, the contrac-
                          tor shall examine each sample and judge if it needs
                          further mixing.

                 NOTE:  Contractor personnel have the responsibility to take a
                        representative sample portion; this responsibility

                                      D-14


                                           Gb'2

-------
                        entails  efforts  to make  the sample  as  homogeneous  as
                        possible.   Stirring is recommended  when  possible.

                 10.3.1.2  Cent rifugat ion — When a soil  or  sediment  sample
                          contains  an  obvious liquid  phase,  it must  be
                          centrifuged  to separate the liquid from the solid
                          phase.  Place  the entire sample in a suitable  centri-
                          fuge  bottle  and centrifuge  for 10 minutes  at 2000 rpn.
                          Remove  the  bottle from the  centrifuge.   With a dis-
                          posable pipet, remove  the liquid  phase and discard
                          it.   Mix  the solid phase with  a stainless  steel
                          spatula and  remove a portion to be weighed and analyzed.
                          Return  the  remaining solid  portion to  the  original
                          sample  bottle  (which must be empty)  or to  a clean,
                          empty  sample bottle which is properly  labeled, and
                          store  it  as  described  in 10.2.1.

                 CAUTION:   The  removed liquid may contain TCDD and should  be
                           disposed as a liquid  waste.

                 10.3.1.3  Weigh  between  9.5 and  10.5  g of the  soil or sediment
                          sample  (+0.5 g) to 3 significant  figures.  Dry it to
                          constant  weight at 100°C.   Allow  the sample to cool
                          in a  desiccator.   Weigh the dried soil to  3 signifi-
                          cant  figures.   Calculate and report  percent moisture
                          on Form H-9.
11.  SAMPLE EXTRACTION

     11.1  Soil/Sediment  Extraction

          11.1.1  Immediately before  use,  the  Soxhlet  apparatus  is  charged
                 with 200 to 250 mL  benzene which  is  then  refluxed for  2 hours.
                 The apparatus  is allowed to  cool,  disassembled and the benzene
                 removed and retained  as  a blank for  later analysis if  required.

          11.1.2  Accurately weigh to 3 significant  figures a 10-g  (9.50 g  to
                 10.50 g) portion of the  wet  soil  or  sediment sample.   Mix 100
                 uL of the sample f ortif ication solution  (Section  7.8)  with
                 1.5 mL  acetone (1000  pg  of    C12~2 .3,7 ,8-TCDD) and deposit the
                 entire  mixture in small  portions  on  several sites on the
                 surface of the soil or sediment.

          11.1.3  Add 10  g anhydrous  sodium sulfate  and mix thoroughly using a
                 stainless steel spoon spatula.

          11.1.4  After breaking up any lumps,  place the soil-sodium sulfate
                 mixture in the Soxhlet apparatus  using a  glass wool plug  (the
                 use of  an extraction  thimble  is optional).   Add 200 to 250 mL
                 benzene to the Soxhlet apparatus  and reflux for 24 hours.  The
                 solvent must cycle  completely through the system  at least 3
                 times per hour.

                                      D-15

-------
     11.1.5 Transfer Che extract to a Kuderna-Danish apparatus and
            concentrate to 2 to 3 mL.  Rinse the column and flask with 5 mL
            benzene and collect the rinsate in the concentrator tube.
            Reduce the volume in the concentrator tube to 2 to 3 mL.
            Repeat this rinsing and concentrating operation twice more.
            Remove the concentrator tube from the K-D apparatus and care-
            fully reduce the extract volume to approximately 1 mL with a
            stream of nitrogen using a flow rate and distance such that
            gentle solution surface rippling is observed.

      NOTE:  Glassware used for more than one sample must be carefully
             cleaned between uses to prevent cross-contamination (Note on
             page D-6) .

11.2  Extraction of Aqueous Samples

      11.2.1  Nark the water meniscus on the side of the 1-L sample bottle
              for later determination of the exact sample volume.  Pour
              the entire sample (approximately 1 L) into a 2-L separatory
              f unne 1 .

      11.2.2  Mix 100 uL of the sample fortification solution with 1.5 mL
              acetone (1000 pg of   C12-2,3 ,7,8-TCDD) and add the mixture
              to the sample in the separatory funnel.

      NOTE:  A continuous liquid-liquid extractor may be .used in place of
             a separatory funnel.

      11.2.3  Add 60 mL methylene chloride to the sample bottle, seal and
              shake 30 seconds to rinse the inner surface.  Transfer the
              solvent to the separatory funnel and extract the sample by
              shaking the funnel for 2 minutes with periodic venting.
              Allow the organic layer to separate from the water phase for
              a minimum of 10 minutes.  If an emulsion interface between
              layers exists, the analyst must employ mechanical techniques
              (to be described in the final report) to complete the phase
              separation.  Collect the methylene chloride (3 x 60 mL)
              directly into a 500-mL Kuderna-Danish concentrator (mounted
              with a 10 -mL concentrator tube) by passing the sample extracts
              through a filter funnel packed with a glass wool plug and 5
              g of anhydrous sodium sulfate.  After the  third extraction,
              rinse the sodium sulfate with an additional 30 mL of methylene
              chloride to ensure quantitative transfer.

     11.2.4   Attach a Snyder column and concentrate the extract until  •
              the apparent volume of the liquid reaches  1 mL.  Remove  the
              K-D apparatus and allow it to drain and cool for at least
              10 minutes.  Remove the Snyder column, add 50 mL benzene,
              reattach the Snyder column and concentrate to approximately
              1 mL.  Rinse the flask and the lower joint with 1 to 2 mL
              benzene.  Concentrate the extract to 1.0 mL under a gentle
              stream of nitrogen.
                                 D-16

                                           '"*
                                            "

-------
     11.2.5   Determine the original  sample  volume  by  refilling  the  sample
              bottle to the mark  and  transferring the  liquid to  a  1000-mL
              graduated cylinder.  Record  the  sample  volume  to  the nearest
              5 mL.

11.3 Cleanup Procedures

     11.3.1 Prepare  an acidic silica  column  as follows:   Pack a  1  cm x  10
            cm chromatographic column with a glass  wool  plug, a  layer (1
            cm) of Na2804/K2003(1:1), 1.0  g  silica  gel (Section  7.1.2)  and
            4.0 g of 40-percent w/w sulfuric acid-impregnated silica gel
            (Section 7.1.3).  Pack a  second  chromatographic  column (1  cm x
            30 cm) with a glass wool  plug, 6.0 g  acidic  alumina  (Section
            7.1.1) and top with a 1-cm layer of sodium sulfate  (Section
            7.4.1).   Add hexane to the columns until  they are free of
            channels and air bubbles.

     11.3.2 Quantitatively transfer the benzene extract  (1 mL)  from  the
            concentrator tube to  the  top of  the silica gel column.  Rinse
            the concentrator tube with two 0.5-mL portions of hexane.
            Transfer the rinses to the top of  the silica gel column.

     11.3.3 Elute the extract from the silica  gel column with 90 mL  hexane
            directly into a Kuderna-Danish concentrator.  Concentrate  the
            eluate to 0.5 mL, using nitrogen blow-down as necessary.

     11.3.4 Transfer the concentrate  (0.5  mL)  to  the  top of  the  alumina
            column.   Rinse the K-D assembly  with  two  0.5-mL  portions of
            hexane and transfer the  rinses to  the top  of the alumina
            column.   Elute the alumina column  with  18 mL hexane  until the
            hexane level is just  below the top of the  sodium sulfate.
            Discard  the eluate.  Columns must  not be  allowed to  reach
            dryness  (i.e., a solvent  "head"  must be maintained.)

     11.3.5 Place 30 mL of 20-percent (v/v)  methylene chloride  in  hexane
            on top of the alumina and elute  the TCDDs  from the  column.
            Collect  this fraction in a 50-mL Erlenmeyer  flask.

     11.3.6 Prepare  an 18-percent Carbopak C/Celite 545® mixture by  thoroughly
            mixing 3.6 grams Carbopak C (80/100 mesh)  and 16.4  grams Celite
            545® in  a 40-mL vial.  Activate  at 130°C  for 6 hours.  Store
            in a desiccator.  Cut off a clean  5~mL  disposable glass  pipet
            (6 to 7mm ID) at the  4-mL mark.  Insert a plug of glass  wool
            (Section 7.3) and push to the  2-mL mark.   Add 340 to 600 mg of
            the activated Carbopak/Celite  mixture (see NOTE) followed by
            another  glass wool plug.   Using  two glass  rods,  push both
            glass wool plugs simultaneously  towards the  Carbopak/Celite
            mixture  and gently compress the  Carbopak/Celite  plug to  a
            length of 2 to 2.5 cm.  Preelute the  column  with 2 mL  toluene
            followed by 1 mL of 75:20:5 methylene chloride/methanol/benzene,
            1  mL of  1:1 cyclohexane  in methylene chloride, and  2 mL  hexane.
            The flow rate should  be  less than  0-5 mL/min. While the

                                 D-17
                                          r> i» r."
                                          "OOtJ

-------
                 column is still wet with hexane, add the entire eluate (30 mL)
                 from the alumina column (Section 11.3.5) to the top of the
                 column.  Rinse the Erlenmeyer flask which contained the extract
                 twice with 1 mL hexane and add the rinsates to the top of the
                 column.  Elute the column sequentially with two 1-mL aliquots
                 hexane, 1 mL of 1:1 cyclohexane in methylene chloride, and 1
                 mL of 75:20:5 raethylene chloride/ methanol/benzene.  Turn the
                 column upside down and elute the TCDD fraction with 6 mL tolu-
                 ene into a concentrator tube.  Warm the tube to approximately
                 60°C and reduce the toluene volume to approximately 1 mL using
                 a stream of nitrogen.  Carefully transfer the concentrate into
                 a 1-mL mini-vial and, again at elevated temperature, reduce the
                 volume to about 100 uL using a stream of nitrogen.  Rinse the
                 concentrator tube with 3 washings using 200 uL of 1% toluene
                 in CH2Cl2«  Add 10 uL of the tridecane solution containing the
                 recovery standard and store the sample in a refrigerator until
                 HRGC/HRMS analysis is performed.

                 NOTE:  The amount of activate Carbopak/Celite mixture required
                        to form a 2-to 2.5-cm plug in the column depends on the
                        density of the Celite being used.

12.  ANALYTICAL PROCEDURES

     12.1 Remove the sample extract or blank from storage and allow it to warm
          to ambient laboratory temperature.  With a stream of dry, purified
          nitrogen, reduce the extract/blank volume to 10 uL.

     12.2 Inject a 2-uL aliquot of the extract into the GC, operated under the
          conditions previously used (Section 8.1) to produce acceptable results
          with the performance check solution.

     12.3 Acquire SIM data according to 12.3.1.  Use the same acquisition and
          MS operating conditions previously used (Section 8.3.4) to determine
          the relative response factors.

          12.3.1 Acquire SIM data for the following selected characteristic ions:

                   ci/z                        Compound

                 258.930                      TCDD - COC1

                 319.897                      Unlabeled TCDD
                                                    •
                 321.894                      Unlabeled TCDD

                 331.937                      13C12-2,3,7,8-TCDD, 13C,2-1,2,3,4-
                                              TCDD

                 333.934                      13C12-2,3,7,8-TCDD, 13C12-1,2,3,4-
                                              TCDD
                                      D-18
                                      P ' * C'
                                      toh

-------
          NOTE:  The acquisition period must at least encompass  the TCDD  reten-
                 tion time window previously determined  (.Section 8.1.2, Exhibit
                 D).

     12.4 Identification Criteria

          12.4.1 The retention time (RT) (at maximum peak height)  of the  sample
                 component m/z 319.897 must be within -1 to +3 seconds  of the
                 retention time of the peak for the isotopically labeled  internal
                 standard at m/z 331.937 to attain a positive identification of
                 2,3,7,8-TCDD.  Retention times of other tentatively identified
                 TCDDs must fall within the RT window established by analyzing
                 the column performance check solution (Section  8.1).   Retention
                 times are required for all chromatograms.

       '   12.4.2 The ion current responses for m/z 258.930, 319.897 and 321.894
                 must reach maximum simultaneously (_+ 1  sec), and all  ion
                 current intensities must be >^ 2.5 times noise level for
                 positive identification of a TCDD or group of coeluting  TCDD
                 isomers.

          12.4.3 The integrated ion current at m/z 319.897 must  be between 67
                 and 90 percent of the ion current response at m/z 321.894.

          12.4.4 The integrated ion current at m/z 331.937 must  be between 67
                 and 90 percent of the ion current response at m/z 333.934.

          12.4.5 The integrated ion currents for m/z 331.937 and 333.934  must
                 reach their maxima within +_ 1 sec.

          12.4.6 The recovery of the internal standard 13C12~2,3,7,8-TCDD must
                 be between 40 and 120 percent.

13.   CALCULATIONS

     13.1 Calculate the concentration of 2,3,7,8-TCDD (or any other TCDD  isomer
          or group of coeluting TCDD isomers) using the  formula:

                                           AX ' QlS
                                       AIS *  W •  RRF(I)

where:

     GX  =  unlabeled 2,3,7,8-TCDD (or any other unlabeled TCDD isomer or group of
            coeluting TCDD isomers) concentration in pg/g.

     Ax  =  sum of the integrated ion abundances determined for m/z 319.897
            and 321.894.

         =  sum of the integrated ion abundances determined for m/z 331.937
            and 333.934 of   C12~2,3 , 7,8-TCDD (IS = internal standard).

                                      D-19


                                          G6V

-------
     QIS -  quantity  (in picogranis) of  3C12-2,3,7 ,8-TCDD added to the
            sample before  extraction (Qjg = 1000 pg).

      W  «  weight (in grams)  of dry soil or sediment sample or volume of
            aqueous sample converted to grams.

 RRF(I)  =  calculated
           mean relative response factor for unlabeled 2,3,7,8-TCDD
              Cj2~2,3,7 ,8-TCDD.  This represents the grand  mean  of
the RRF(I)'s obtained in Section 8.3.4.5.
            relative to     j2
     13.2 Calculate the recovery of the internal standard   C12~2 ,3,7,8-TCDD
          measured in the sample extract, using the formula:
          Internal standard
          percent recovery    -  Y        — -  • 100
                                   ARg  * RRF(II)
Where:
          =  sun of the integrated ion abundances determined for m/z 331.937
             and 333.934 of 1JC12-2,3,7,8-TCDD (IS  =  internal standard).

     ARS  =  sum °^ tlie integrated ion abundances determined for m/z 331.937
             and 333.934 of   C12-l ,2,3,4-TCDD (RS  =  recovery standard).

       Y  =  0.1 for the "10-yL extract" injection (to be reported on Forms H-l,
             H-5 and H-9).

   and Y  "  1.2 for the "24-uL extract" injection (Section 13.3) (to be reported
             on Form H-9 used for reporting the diluted extract analysis).

RRF(II)  »  calculated mean relative response factor for labeled   C|2-2,3,7,8-
            TCDD relative to   C12~l ,2,3,4-TCDD.  This represents the grand
            mean of the RRF(II)'s calculated in Section 8.3.4.5.

     13.3  If the concentration of the most abundant TCDD isomer (or group of
           coeluting TCDD isomers) exceeds 100 pg/uL in the 10 uL final extract,
           the linear range of response vs. concentration may have been exceeded,
           and a diluted aliquot of the original sample extract must be analyzed.
           Accurately dilute 2 uL of the remaining original extract with 22 uL
           of the tridecane solution containing 10 pg/uL of the recovery standard
           (Section 7.9, Exhibit D).

     13.4  Total TCDD concentration — all positively identified isomers of TCDD
           must be within the RT window and meet all identification criteria
           listed in Sections 12.4.2 and 12.4.3.  Use the expression in Section
           13.1 to calculate the concentrations of the other TCDD isomers, with
           Cx becoming the concentration of any unlabeled TCDD isomer or group
           of coeluting TCDD isomers.

c Total TCDD  -  Sum of the concentrations of the individual TCDDs including
                 2,3,7,8-TCDD.
                                      D-20

-------
     13.5  Estimated Detection Limit  — For samples  in  which  no  unlabeled
           2,3,7,8-TCDD was detected, calculate  the  estimated minimum  detectable
           concentration.   The background area is  determined  by  integrating the
           ion abundances  for n/z 319.897 and  321.894  in the  appropriate region
           of the selected ion current profiles, multiplying  that  area by  2.5,
           and relating the product area to an estimated concentration that
           would produce that product area.

           Use the formula:

                                          (2.5)  '  (Ax)  ' (QIS)
                                  CE
                                         (AIS)  '  (RRF(I))  '  (W)

where

     CE  =  estimated concentration of unlabeled  2,3,7,8-TCDD required  to
            produce Ax.

     Ax  =  sum of integrated ion abundances  for  m/z  319.897 and 321.894 in  the
            same group of >5 scans used to measure Ajg.

     AJS =  sum °^ integrated ion abundances  for  the  appropriate ion  character-
            istic of the internal standard, m/z 331.937  and  m/z  333.934.

QiS» RRF(I), and W retain the definitions previously  stated  in Section- 13.1.
Alternatively, if peak height measurements are  used for  quantification, measure
the estimated detection limit by the peak height  of the  noise in the  2,3,7,8-
TCDD RT window.

     13.6  The relative percent  difference (RPD)  is calculated as follows:

                        I  S! - S2 |           |  S1 - S2  |
             RPD  =  	  = 	' x  100
                     Mean Concentration      (Sj  + S2)/2

          Sj and $2 represent sample and duplicate sample  results.

References

1.  "Carcinogens - Working with  Carcinogens", Department  of  Health, Education
     and Welfare, Public Health  Service, Center for Disease  Control,  National
     Institute for Occupational  Safety and Health, Publication No.  77-206, Aug.
     1977.

2.   "OSHA Safety and Health Standards, General Industry"  (29 CFR1910),
     Occupational Safety and Health Administration, OSHA 2206 (Revised  January
     1976).

3.   "Safety in Academic Chemistry Laboratories", American Chemical Society
     Publication, Committee on Chemical Safety, 3rd Edition  1979.


                                      D-21


                                      660

-------
TABLE 1 .  COMPOSITION OF CONCENTRATION CALIBRATION SOLUTIONS

HRCC1
HRCC2
HRCC3
HRCC4

Recovery Standard
13C12-1,2,3,4-TCDD
2.0 pg/uL
10.0 pg/uL
50.0 pg/uL
100.0 pg/uL
Sample
Analyte
2,3,7,8-TCDD
2.0 pg/uL
10.0 pg/uL
50.0 pg/uL
100.0 pg/uL
Fortification Solution
Internal Standard
13C12-2,3,7,8-TCDD
10.0 pg/uL
10.0 pg/uL
10.0 pg/uL
10.0 pg/uL

             10.0 pg/uL of 13C12-2,3,7,8-TCDD



             Recovery Standard Spiking Solution

               10.0 pg/uL 13C12-1,2,3,4-TCDD
            Field Blank Fortification Solutions

          A)  10.0 pg/uL of unlabeled 2,3,7,8-TCDD

          B)  10.0 pg/uL of unlabeled 1,2,3,4-TCDD



             Internal Standard Spiking Solution

              10 pg/uL of 13C12-2,3,7,8-TCDD
             (Used only in Section 4.2.1.1,  Exhibit E)
                             D-22

-------
                 TABLE 2.   RECOMMENDED GC  OPERATING  CONDITIONS
Column coating

Film thickness

Column dimensions

Helium linear velocity


Initial temperature

Initial time

Temperature program
Approximate 2,3,7,8-TCDD
retention time
SP-2330 (SP-2331)

0.2 urn

60 m x 0.24 mm

28-29 cm/sec
at 240°C

150°C

4 min

Rapid increase to .200°C
(15°C/min)
200°C to 2508C
at 4°C/min

27 min
CP-SIL 88

0.22 urn

50 m x 0.22 mm

28-29 cm/sec
at 240°C

2008C

1 min

Program from 200°C
to 240°C
at 4°C/min


22 min
         TABLE 3.  TYPICAL 12-HOUR SEQUENCE FOR 2,3,7,8-TCDD ANALYSIS
1. Static mass resolution check and mass
measurement error determination 10/20/84
2 . Co lumn
3 . HRCC2
4. Sample
5 . Co lumn
6. Static
performance check

1 through Sample "N"
performance check
mass resolution check
10/20/84
10/20/84
10/20/84
10/20/84
10/20/84
0700h
0730h
0800h
0830h
1800h
1830h
                                      D-23
                                       f! 7 I
                                       w I JL-

-------
     100i
o

K>
*>
       80-
 M

 § 60
 «•*
 c,

 9


 <0 40
"3
DC



   20-
                00
                
-------
    EXHIBIT E
QA/QC Requirements
           B73

-------
SUMMARY OF QC ANALYSES

o  Initial and periodic calibration and instrument performance checks.

o  Field blank analyses (Section A.I); a minimum of one fortified field blank
   pair shall be analyzed with each sample batch; an additional fortified field
   blank pair must be analyzed when a new lot of absorbent and/or solvent is used.

o  Analysis of a batch of samples with accompanying QC analyses:

        Sample Batch —_<24 samples, including field blank and rinsate sample(s).

        Additional QC analyses per batch:


        Fortified field blanks              2

        Method blank                       (1*)

        Duplicate sample                    1

                              TOTAL         3(4)


        * A method blank is required whenever a fortified field blank shows a
          positive response as defined in Section 3.11, Exhibit D.

o  "Blind" QC samples may be submitted to the contractor as ordinary soil,
   sediment or water samples included among the batch of samples.  Blind samples
   include:

        Uncontaminated soil, sediment and water,

        Split samples,

        Unidentified duplicates, and

        Performance evaluation samples.

QUALITY CONTROL

1.   Performance Evaluation Samples — Included among the samples in all batches
     will be samples containing known amounts of unlabeled 2,3,7,8-TCDD and/or
     other TCDDs that may or may not be marked as other-than-ordinary samples.

2.   Performance Check Solutions

     2.1  At the beginning of each 12-hour period during which samples are to
          be analyzed, an aliquot each of the 1) GC column performance check
          solution and 2) high-resolution concentration calibration solution
                                      E-l
                                          r* >"i
                                          6 <

-------
     No. 2 (HRCC2) shall  be analyzed Co demonstrate adequate GC resolution
     and sensitivity,  response factor reproducibility, and mass range
     calibration.  A mass resolution check shall also be performed to
     demonstrate adequate mass resolution using an appropriate
     reference compound  (PFK is recommended).

     These procedures  are described in Section 8 of Exhibit D.  If the
     required criteria are not met, remedial action must be taken before
     any samples are analyzed.

2.2  To validate positive sample data, the GC column performance check
     and the mass resolution check must be performed also at the end of
     each 12-hour period  during which samples are analyzed.

     2.2.1  If the contractor laboratory operates only during one period
            (shift) each  day of 12 hours or less, the GC performance check
            solution must be analyzed twice (at the beginning and end of
            the period) to validate data acquired during the interim
            period.  This applies also to the mass resolution check.

     2.2.2  If the contractor laboratory operates during consecutive
            12-hour periods (shifts), analysis of the GC performance check
            solution at the beginning of each 12-hour period and at the
            end of the final 12-hour period is sufficient.  This applies
            also to the mass resolution check.

2.3  Results of at least  two analyses of the GC column performance check
     solution and the  mass resolution check must be reported with the
     sample data collected during a 12-hour period.

2.4  Deviations from criteria specified for the GC performance check or
     for the mass resolution check (Section 8, Exhibit D) invalidate all
     positive sanple data collected between analyses of the performance
     check solution, and  the extract from those positive samples shall be
     reanalyzed Exhibit C).

The GC column performance check mixture, concentration calibration solu-
tions, and the sample  fortification solutions are to be obtained from the
EMSL-LV.  However, if  not available from the EMSL-LV, standards can be
obtained from other sources, and solutions can be prepared in the contractor
laboratory.  Concentrations of all solutions containing unlabeled 2,3,7,8-
TCDD which are not obtained from the EMSL-LV must be verified by comparison
with the unlabeled 2,3,7,8-TCDD standard solution (concentration of 7.87
ug/raL) that is available  from the EMSL-LV.  When a lower-concentration
standard solution becomes available from the EMSL-LV, it will be substituted
for the 7.87 ug/tnL standard.
                                 E-2
                                      7
                                      f

-------
4.   Blanks
     4.1  A method blank is required whenever a positive  response (Section  3.11,
          Exhibit D) is  obtained for a fortified field  blank.   To that  effect,
          perform all steps detailed in the analytical  procedure (Section  11,
          Exhibit D) using all reagents,  standards,  equipment,  apparatus,
          glassware, and solvents that would be used for  a sample analysis,  but
          omit addition  of the soil, sediment or aqueous  sample portion.

                                                                  13
          4.1.1  The method blank must contain the same amount  of   Cj2~2,3,7,8-
                 TCDD that is  added to samples before extraction.

          4.1.2  An acceptable method blank exhibits no positive response  (Section
                 3.11, Exhibit D) for any of the characteristic ions monitored.
                 If the  method blank which was extracted  along  with  a batch of
                 samples is contaminated, all positive  samples  must  be  rerun
                 (Exhibit C).

                 4.1.2.1  If the above criterion is  not met,  check solvents,
                          reagents, fortification solutions,  apparatus, and
                          glassware to locate and eliminate the source  of
                          contamination before any samples are  extracted and
                          analyzed.

                 4.1.2.2  If new batches  of reagents or solvents contain
                          interfering contaminants,  purify or discard them.

     4.2  Field blanks — Each batch of samples contains  a field blank  sample
          of uncontaminated soil/sediment or water that is to be fortified
          before analysis according to Section 4.2.1, Exhibit E.  In addition
          to this field  blank, a batch of samples may include a rinsate, that
          is a portion of solvent (usually trichloroethylene) that was  used to
          rinse sampling equipment.  The  rinsate is  analyzed  to assure  that the
          samples have not been contaminated by the  sampling  equipment.

          4.2.1  Fortified field blank pair

                 4.2.1.1  Fortified field blank A:  2,3,7,8-TCDD

                 4.2.1.1.1  Weigh a 10-g  portion or  use 1 liter (for aqueous
                            samples) of the specified field blank sample and
                            add 100 uL of the solution  containing 10.0  pg/uL of
                            2,3,7,8-TCDD  (Table 1, Exhibit D) diluted in 1.5 mL
                            of acetone (Section 11.1.2, Exhibit D).

                 4.2.1.1.2  Extract using the procedures  beginning in Sections
                            11.1 or 11.2  of Exhibit  D,  as applicable, add  10 uL
                            of the internal standard solution (Section  7.10,
                            Exhibit D) and analyze a 2-uL aliquot of the con-
                            centrated extract.
                                      E-3
                                             f

-------
       NOTE:  This is the only case where the recovery standard is
              used for other than recovery purposes.

       4.2.1.1.3  Calculate the concentration (Section 13.1, Exhibit
                  D) of 2,3,7,8-TCDD and the percent recovery of
                  unlabeled 2,3,7,8-TCDD.  If the percent recovery at
                  the measured concentration of 2,3,7,8-TCDD is <40
                  percent or >120 percent, report the results and
                  repeat the fortified field blank extraction and
                  analysis with a second aliquot of the specified
                  field blank sample (Exhibit C).

       4.2.1.1.4  Extract and analyze a new fortified simulated field
                  blank whenever new lots of solvents or reagents are
                  used for sample extraction or for column chromato-
                  graphic procedures.  When a fortified simulated
                  field blank produces a positive response (Section
                  3.11, Exhibit D) for any m/z being monitored at the
                  retention time of 1,2,3,4-TCDD, a method blank
                  (Section 4.1, Exhibit E) is required.

       NOTE:  For this purpose only, the Contractor will simulate
              field blanks by using clean sand or distilled water.

       4.2.1.2  Fortified field blank B:  1,2,3,4-TCDD

       4.2.1.2.1  Repeat steps 4.2.1.1.1 to 4.2.1.1.3 using unlabeled
                  1,2,3,4TCDD (instead of 2,3,7.8-TCDD) and 13C
                  12-1,2,3,4-TCDD (instead of 13C12-2,3,7,8-TCDD)
                  recovery standard.
as
       4.2.1.2.2  Extract and analyze a new fortified simulated field
                  blank whenever new lots of solvents or reagents are
                  used for sample extraction or for column chromato-
                  graphic procedures.  When a fortified simulated
                  field blank produces a positive response (Section
                  3.11, Exhibit D) for any m/z being monitored at the
                  retention time of 2,3,7,8-TCDD, a method blank
                  (Section 4.1, Exhibit E) is required.

4.2.2  Rinsate sample

       4.2.2.1  The rinsate sample must be fortified as a regular
                sample.

       4.2.2.2  Take a 100-mL aliquot of sampling equipment rinse
                solvent (rinsate sample), filter, if necessary, and
                add 100 uL of the solution containing 10.0 pg/uL of
                13C12-2,3,7,8-TCDD (Table 1, Exhibit D).
                            E-4
                              r. 7 V
                              o J  '

-------
                 4.2.2.3  Using a Kuderna-Danish apparatus,  concentrate  to
                          approximately 5 mL.

                 4.2.2.4  Transfer the 5-mL concentrate in 1-mL portions  to a  1-
                          mL mini-vial, reducing the  volume  as  necessary  with  a
                          gentle stream of dry nitrogen;  see Exhibit  D,
                          Section 11.1.5 for volume reduction procedures.

                 4.2.2.5  Rinse the container with two 0.5-mL portions  of hexane
                          and transfer the rinses to  the 1-mL mini-vial.

                 4.2.2.6  Just before analysis,  add 10 uL tridecane recovery
                          standard spiking solution (Table 1, Exhibit D), and
                          reduce the volume to a final volune of 10 uL  (no
                          column chromatography is required).

                 4.2.2.7  Analyze an aliquot following the same procedures used
                          to analyze samples (Section 12, Exhibit D).

                 4.2.2.8  Report percent recovery of  the internal standard and
                          the level of contamination  by any  TCDD isomer  (or
                          group of coeluting TCDD isomers) on Form H-5  in pg/mL
                          of rinsate solvent.
5.   Duplicate Analyses

     5.1  Laboratory duplicates — in each batch of samples, locate the  sample
          specified for duplicate analysis and analyze a second 10-g soil or
          sediment sample portion or 1-L water sample.

          5.1.1  The results of laboratory duplicates (percent  recovery  and
                 concentrations of 2,3,7,8-TCDD and total TCDD) must  agree
                 within 50 percent relative difference (difference expressed as
                 percentage of the mean).  If the relative difference is >50
                 percent, the Contractor shall immediately contact the  Sample
                 Management Office for resolution of  the problem.  Report all
                 results.

          5.1.2  Recommended actions to help locate problems:

                 5.1.2.1  Verify satisfactory instrument performance
                          (Section 8, Exhibit  D).

                 5.1.2.2  If possible, verify that no error  was made while
                          weighing sample portions.

                 5.1.2.3  Review the analytical  procedures with the performing
                          laboratory personnel.

6.   Percent Recovery of the Internal Standard ^C,--2,3,7,8-TCDD — For each
     sample, method blank and rinsate, calculate the  percent recovery (Section
     13.2,  Exhibit D) of the measured concentration of 13C12-2,3,7,8-TCDD.   If


                                      E-5


                                        G78

-------
     the percent recovery  is  <40 percent or >120 percent for a sample, analyze
     a second portion of that  sample and report both results (Exhibit C).

NOTE:  A low or high percent  recovery for a blank does not require discarding
       analytical data but  it  may indicate a potential problem with future
       analytical data.

7.   Identification Criteria

     7.1  If either of the  two identification criteria (Sections 12.4.1 and
          12. A. 2, Exhibit  D)  is not met, it is reported that the sample does
          not contain unlabeled 2,3,7,8-TCDD at the calculated detection limit
          (Section 13.5, Exhibit D).

     7.2  If the first two  initial identification criteria are met, but the
          third, fourth, fifth or sixth criterion (Sections 12.4.3 through
          12.4.6, Exhibit D)  is not met, that sample is presumed to contain
          interfering contaminants.  This must be noted on the analytical
          report form and  the  sample must be rerun or the extract reanalyzed.
          Detailed sample  rerun and extract reanalysis requirements are
          presented in Exhibit C.

8.   Blind QC Samples — Included among soil, sediment and aqueous samples may
     be QC samples that are not specified as such to the performing laboratory.
     Types that may be included are:

          8.1    Uncontaminated soil, sediment or water.

                 8.1.1  If  a  false positive is reported for such a sample,
                        the Contractor shall be required to rerun the entire
                        associated batch of samples (Section 2.3.3, Exhibit C).

          8.2    Split samples — composited sample portions sent to more than
                 one laboratory.

          8.3    Unlabeled  field duplicates — two portions of a composited
                 sample.

          8.4    Performance  evaluation sample — soil/sediment or water sample
                 containing a  known amount of unlabeled 2,3,7,8-TCDD and/or
                 other TCDDs.

                 8.4.1  If  the performance evaluation sample result falls
                        outside the acceptance windows established by EPA, the
                        Contractor shall be required to rerun the entire associ-
                        ated  batch of samples (Exhibit C).

                 NOTE: EPA  acceptance windows are based on previously generated
                       data.

9.   Records - At each contractor laboratory, records must be maintained on


                                      E-6
                                         r: 7
                                         *_/ (

-------
     site for six months after contract  completion  to document  the quality of
     all data generated during the contract  performance.   Before  any records are
     disposed, written concurrence from  the  Contracting Officer must be obtained.

10.  Unused portions of samples and sample extracts must  be preserved for
     six months after sample receipt;  appropriate samples may be  selected
     by EPA personnel for further analyses.

11.  Reuse of glassware is to be minimized to avoid the risk of contamination.

LABORATORY EVALUATION PROCEDURES

1.   On a quarterly basis, the EPA Project Officer  and/or designated
     representatives may conduct an evaluation of the laboratory  to ascertain
     that the laboratory is meeting contract requirements.  This  section outlines
     the procedures which may be used  by the Project Officer or his authorized
     representative in order to conduct  a successful evaluation of laboratories
     conducting dioxin analyses according to this protocol.  The  evaluation
     process consists of the following steps:  1) analysis of a performance
     evaluation (PE) sample, and 2) on-site  evaluation of the laboratory to
     verify continuity of personnel, instrumentation, and quality assurance/
     quality control functions.  The following is a description of these
     two steps.

2.   Performance Evaluation Sample Analysis

     2.1  The PE sample set will be sent to  a participating laboratory to
          verify the laboratory's continuing ability to produce acceptable
          analytical results.  The PE  sample will be representative of the
          types of samples that will be  subject to analysis under this contract.

     2.2  When the PE sample results are received,  they are scored using the
          PE Sample Score Sheet shown  in Figure 1.   If a  false  positive
          (e.g., a PE sample not containing  2,3,7,8-TCDD  and/or other TCDDs
          but reported by the laboratory to  contain it and/or them) is reported,
          the laboratory has failed the  PE analysis requirement.   The Project
          Officer will notify the laboratory immediately  if such  an event
          occurs.

     2.3  As a general rule, a laboratory should achieve  75 percent or more of
          the total possible points for  all  three categories, and 75 percent  or
          more of the maximum possible points in each category  to be considered
          acceptable for this program.  However, the Government reserves the
          right to accept scores of less than 75 percent.

     2.4  If unanticipated difficulties  with the PE samples are encountered,
          the total points may be adjusted by the Government evaluator in an
          impartial and equitable manner for all participating  laboratories.
                                      E-7
                                         G80

-------
     Number of          Maximum Possible        Recommended  Passing
     PE Samples             Score	          Score  (75Z)	
        1                    290                       218

        2                    475                       356

        3                    660                       495

        4                    845                       634

        5                   1030                       773


On-Site Laboratory Evaluation

3.1  An on-site laboratory evaluation is performed •to verify that  (1)  the
     laboratory is maintaining the necessary minimum level in instrumen-
     tation and levels of experience in personnel committed to the con-
     tract and (2) that the necessary quality control/quality assurance
     activities are being carried out.  It also serves as  a mechanism  for
     discussing laboratory weaknesses identified through routine data
     audits, PE sample analyses results, and prior on-site evaluation.
     Photographs may be taken during the on-site laboratory evaluation
     tour.

3*2  The sequence of events for the on-site evaluations is shown in
     Figure 2.  The Site Evaluation Sheet (SES) (Figure 3) is used to
     document the results of the evaluation.
                                    E-8

                                      GB1

-------
Laboratory
                     PERFORMANCE EVALUATION SAMPLE SCORE  SHEET

                                                         Date
                                                               False Positive
I.   False Positive - If a laboratory reports a false
     positive on any PE sample, the laboratory may be
     disqualified, i.e., rendered ineligible for
     contract award based on the failure to pass the
     PE sample analysis requirement.

     2,3,7,8-TCDD

     Other TCDD(s)
                                                               (  ) Yes  ( ) No

                                                               (  ) Yes  ( ) No
                                                            Possible
                                                              Score
II.   Calibration Data

     1.   Method Blank:

         a.   Results properly recorded on Forms H-l,  H-5 and    5
             H-9.

         b.   No native  TCDD isomers  at/or above method
             quantitative limit.                                 5

         c.   Results documented by selected ion
             monitoring (SIM) traces for m/z being
             monitored  to detect  TCDDs.                         5

         d.   Percent recovery of  13C,2-2,3,7,8-TCDD
             240 and £120%.                                     5

     2.   Initial Concentration Calibration:

         a.   Results properly recorded on Forms H-2             5
             and H-8.

         b.   The percent  relative standard deviation
             (RSD)  for  the response  factors for each
             of the triplicate analyses for both unlabeled
             and 13C12-2,3,7,8-TCDD  less than 20%.              5

         c.   The variation of the 4  mean RRFs for both
             unlabeled  and labeled 2,3,7,8-TCDD obtained
             from the triplicate  analyses less than 20% RSD.    5

         d.   For unlabeled 2,3,7,8-TCDD the abundance ratio
             must be 20-67 and £0.90 for m/z 319.897  to
             321.894.                                           5

             Figure 1.  Performance  evaluation sample score sheet.

                                      E-9^
                                        O 
-------
                                                   Possible     Score
                                                     Score    Achieved
e.  The abundance ratios must be X).67 and <0.90
    for 331.937 to 333.934 for 13C12-2,3, 7,8-TCDD
    and 13C12-1,2,3,4-TCDD.

f.  Results must be documented with appropriate
    SIM traces, labeled with the corresponding EPA
    sample numbers, and calculations.

 Performance Checks:

a.  GC resolution and MS resolution checks performed
    at the beginning and end of each 12-hour period.

b.  Results of performance checks properly recorded
    on Form H-4.

c.  MS Resolution:  PFK (or alternate) tune shows
    appropriate mass resolution (Section 8.2,
    Exhibit D) with mass assignment accuracy
    within +5 ppm.

d.  GC Resolution:  chromatograms meet the criteria
    specified in Section 8.1, Exhibit D.

Routine Calibration:

a.  Performed each 12 hours, after MS and  GC
    resolution checks, using HRCC2.

b*  Results of routine calibrations properly
    reported on Forms H-3 and H-8.

c.  For un labeled 2,3 , 7,8-TCDD:  abundance
    ratio must be X).67 and £0.90 for m/z
    319.897 to 321.894.
Abundance ratio correct for isotopically
labeled standards (e.g., 331.937/333.934
     be >0.67 and £0
     3C-1,2,3,4-TC
    must be >0.67 and £0.90 for   C-2 ,3 , 7,8-TCDD
    and 13C12-1,2,3,4-TCDD).
    Response factors [RRF(I) and RRF(II)] are
    within j+20Z of the mean of the respective
    initial calibration response factors.              5

    Signal-to-Noise (S/N) .Ratio:  SIM traces
    for 2, 3, 7,8-TCDD demonstrate S/N of ^2.5.          5

    Results documented with appropriate SIM
    traces and calculations.                           5 .
                                        Subtotal II   105

                   Figure 1.  (Continued).

                             E-10
                                G83

                                                           Possible    Score
                                                             Score    Achieved
III.  Performance Evaluation (PE)  Sample  Data
     (Scores to be determined  for each  sample
      in the PE set)

     1.   Forms H-l and H-9 properly  filled  out  for  sample.       5

     2.   Measured concentration  of unlabeled
         2,3,7,8-TCDD within acceptance  window
         established  by EPA.                                  40

     3.   Estimated concentration of  total TCDDs
         within acceptance window established by
         EPA.                                                 20

     4.   Identification Criteria for 2,3,7,8-TCDD:

         a.   Retention time (RT)  (at maximum peak
             height)  of the sample component tn/z
             319.897  is within -1 to +3  seconds
             of the m/z 331.937  13C122,3,7,8-TCDD
             internal standard peak.                          10

         b.   The ion  current responses  for  m/z
             258.930, 319.897  and 321.894 must reach
             a maximum simultaneously (_+! second)
             and must be J>2*5  times  noise level.               10

         c.   The m/z  319.897/321.894 ratio  is X).67
             and £0.90.                                       10

         d.   The m/z  331.937/333.934 ratio  is X).67
             and £0.90.                                         5

         e.   The S./N  ratio for m/z 331.937  and
             333.934  is ^2.5.                                    5

     5.   Identification Criteria for other  TCDDs:

         a.   Retention time must  fall into  window
             established by GC performance  check.                5

         b.   The ion  current responses  for  m/z
             258.930, 319.897, and 321.894  reach
             a maximum simultaneously (_+l second)
             and are  >2.5  times  noise level.                   10

         c.   The ra/z  319.897/321.894 ratio  is
             X).67 and £0.90.                                    5

                            Figure 1.   (Continued).
                                     E-ll
                                        684

-------
                                                        Possible    Score
                                                          Score    Achieved
 6.  Concentrations of unlabeled TCDDs
     are calculated according to D-13.1.                   10

 7.  Duplicate analysis values agree within
     _*502.                                                 10

 8.  Estimated detection limits calculated
     according to D-13.5.                                  10

 9.  Percent recovery of 13C,,-2,3,7,8-TCDD
     >40 and O20%.                                        10

10.  Results documented with appropriate
     SIM traces and calculations.                          20
                                           Subtotal III   185

                                                  Total   290

                        Figure 1.  (Continued).
                                  E-12
                                     f~" f*. v •
                                     b o O

-------
                EVENT SEQUENCE FOR ON-SITE LABORATORY EVALUATION
I.    Meeting with Laboratory Manager and  Project Manager

      Introduction; discuss purpose of visit;  discuss problems  with data
      submitted by the laboratory.

II.   Verification of Personnel

      Review qualification of contractor personnel in place and committed  to
      project (Section I,  SES).

III.  Verification of Instrumentation

      Review equipment in place and committeed to project  (Section II,  SES).
      The Contractor must  demonstrate adequate equipment  redundancy,  as defined
      in SES, Section II.D.,  to ensure his capability to  perform the required
      analyses in the required time.

IV.   Quality Control Procedures

      Walk through the laboratory to  review:

       1.  Sample receiving and logging procedures,
       2.  Sample and extract storage area,
       3.  Procedures to prevent sample contamination,
       4.  Security procedures for  laboratory  and samples,
       5.  Safety procedures,
       6.  Conformance to written SOPs,
       7.  Instrument records and logbooks,
       8.  Sample and data control  systems,
       9.  Procedures for handling  and disposing of hazardous materials,
      10.  Glassware cleaning procedures,
      11.  Status of equipment and  its availability,
      12.  Technical and managerial review of  laboratory  operations and
             data package preparations,
      13.  Procedures for  data handling, analysis, reporting and case
             file preparation, and
      14.  Chain-of-custody procedures.

V.    Review of Standard Operating  Procedures  (SOPs)

      Review SOPs with the Project  Manager to  assure that  the laboratory  under-
      stands the dimensions and requirements  of the program.

VI.   Identification of Needed Corrective  Actions

      Discuss with the Project Manager the actions needed  to correct weaknesses
      identified during the site inspection,  PE sample  analysis or production  of

          Figure 2.  Event Sequence for On-Site Laboratory Evaluation.

                                      E-13

                                         686

-------
       reports  (hard  copies  and,  if  appropriate, manual calculations) and documen-
       tation.  Determine how and when corrective actions will be documented,
       how and  when  improvements  will be demonstrated, and identify the contractor
       employee responsible  for corrective actions.

 VII.  Previously  Identified Problems

       Check the most recent SES  to  verify that all previously identified
       problems have  been corrected.

VIII.  Identification of New Problems

       a.  Discuss any weaknesses identified in the performance evaluation
           sample analyses and reports.

       b.  Discuss any weaknessess identified in this site inspection.
                                Figure 2.  (Continued)


                                       E-14

                                              Gr; •> ;
                                              Cj .(

-------
                             SITE EVALUATION  SHEET

Laboratory:  	  Date;


Location:
                                   EVALUATORS
              Name                                        Organization

1.
2.
3.
4.
5.
6.
7.

I.    Laboratory Personnel Committed to Project:

      A.  Project Manager (responsible for  overall  technical  effort)

          Name:	
          Title:
      B.  GC/MS Operator:
          Experience:*
                          (one year minumum)
      C.  GC/MS Data Interpreter:
          Experience:*  	
                          (two year minimum)

      D.  Person responsible for sample exraction,  column chromatography
            and extract concentration:  	
          Experience:*  	
                          (one year minumum)

      E.  Person(s) responsible for calculations  and  report  preparation:
          Hardcopy Reports:   	
      F.  Person responsible for handling,  storage  and  (if appropriate)
            preparation of solutions of standard compounds:
*Experience is deemed to mean "more than 50 percent of the person's productive
 work time."

                       Figure 3.   Site Evaluation Sheet.

                                     E-15

                                        688

-------
      G.  Person responsible for standards preparation/storage:


      H.  Person responsible for record keeping:


      I.  Quality Assurance Officer: 	_____

      J.  Personnel checklist
                                                                (  ) Yes   (  ) No
          1.  Do personnel assigned to this project  have
              the appropriate level and type of experience
              to successfully accomplish the objectives of
              this program?

          2.  Is the organization adequately staffed to          (  ) Yes   (  ) No
              meet project requirements in a tmely
              manner?

          3.  Does the Laboratory Quality Assurance  officer     (  ) Yes   (  ) No
              report to senior management levels?

          4.  Was the Quality Assurance officer available       (  ) Yes   (  ) No
              during the evaluation?

II.   Laboratory Equipment

      A.  Gas chromatograph(s)*

          Manufacturer and Model:  	
          Installation Date:
          Type of Capillary Column Injection System:  	___
          Capillary Column to be used (length,  ID,  coating,  etc.):
          Necessary Ancillary Equipment (gases, syringes,  etc.):
      B.  High Resolution Mass Spectrometer^ s)*

          Static Resolution Capability (10,000  min.):
          Peak matching system:  	
          Manufacturer and Model:  	
          Installation Date:
          Pertinent Modifications:  _____________
          Peak Matching System/Accuracy (Mfg. spec.):

      C.  Data System(s)*
          Manufacturer and Model:
  If more than one GC/MS/DC, indicate system 1,2,3,  etc., by numbering
  components with 1,2,3, etc.

                            Figure 3.  (Continued).

                                      E-16

-------
          Installation Date:
          Software  Version  Identifier:
          Appropriate selected  ion monitoring  software/hardware (  ) Yes  ( ) No
          Capability  to  produce hard copies of computer-
            generated information                               (  ) Yes  ( ) No

      D.   Evidence that  at  least one GC/MS/DS  system  can be  reasonably
          expected to be operating acceptably  at any  given time:

          (  )   More than one adequate  GC/MS/DS system is available in-house,
               (i.e.,meeting requirements  specified in SOW Section 6.1,
               Exhibit D).

          (  )   Appropriate  in-house  replacement  parts and  trained  service
               personnel are available.

          (  )   A service contract  is  in  place  with guaranteed  response  time
               (specify  type of contract and  limitations). 	
          (  )  Voltage control devices  are  used  on major  instruments;  isolated
               circuits  are  used.

          (  )  Other (specify) 	

III.   Facilities Checklist

      A.  Does the laboratory appear to have adequate            (  )  Yes   (  )  No
          workspace (120 sq. feet,  6 linear feet of
          unencumbered bench space  per  analyst)?

      B.  Does the laboratory have  a source of distilled/       (  )  Yes   (  )  No
          demineralized  water?

      C.  Is the analytical  balance located away from            (  )  Yes   (  )  No
          draft and areas subject to rapid  temperature
          changes or vibration?

      D.  Has the balance been calibrated within one year       (  )  Yes   (  )  No
          by a certified technician?

      E.  Is the balance routinely  checked  with  class  S         (  )  Yes   (  )  No
          weights before each use and the results recorded
          in a logbook?

      F.  Is the laboratory  maintained  in a clean and            (  )  Yes   (  )  No
          organized manner?
                            Figure 3.   (Continued)

                                      E-17

                                     690

-------
      G.  Is the  facility designed for hazardous organic        (  )  Yes  (  )  No
          chemical analysis?

          1.  Is  ventilation provided in the sample             (  )  Yes  (  )  No
              preparation areas?

          2.  Are vented hoods available and adequately         (  )  Yes  (  )  No
              vented in the sample preparation areas?

          3.  Are the hoods equipped with charcoal              (  )  Yes  (  )  No
              and HEPA filters?

          4.  Are instruments, including GC/MS pumps,           (  )  Yes  (  )  No
              vented into hoods or control devices such
              as  charcoal traps?

      H.  Are adequate secured facilities provided for          (  )  Yes  (  )  No
          storage of samples, extracts, and calibration
          standards, including cold storage?

      I.  Are the temperatures of the cold storage units        (  )  Yes  (  )  No
          recorded daily in logbooks?

      J.  Are chemical waste disposal policies/procedures       (  )  Yes  (  )  No
          in place?

      K.  Is the  laboratory secure?                             (  )  Yes  (  )  No

IV.   Analysis Control Checklist

      A.  Do the  project personnel have SOPs for the required   (  )  Yes  (  )  No
          activities?

      B.  Is a logbook maintained for each instrument and       (  )  Yes  (  )  No
          is information such as calibration data and
          instrument maintenance continually recorded?

      C.  Do the analysts record bench data in a neat           (  )  Yes  (  )  No
          and accurate manner?

      D.  Standards

          1.  Are fresh analytical standards prepared           (  )  Yes  (  )  No
              at a frequency consistent with good QC?

          2.  Are reference materials properly labeled with     (  )  Yes  (  )  No
              concentrations, date of preparation, and the
              identity of the person preparing the sample?

          3.  Is a standards preparation and tracking           (  )  Yes  (  )  No
              logbook maintained?

                             Figure 3. (Continued).

                                      E-18

                                           G31

-------
          4.  Are working standards traceable to EPA            (  )  Yes  (  )  No
              standards or validated against EPA
              standards?

V.    Documentation/Tracking Checklist

      A.  Is a sample custodian designated?  If yes,            (  )  Yes  (  )  No
          name of sample custodian.
          Name:  	

      B.  Are the sample custodian's procedures and             (  )  Yes  (  )  No
          responsibilities documented?  If yes, where
          are these documented?
          Are the chain-of-custody procedures documented?       ( ) Yes  (  ) No

      C.  Are written Standard Operating Procedures (SOPs)       ( ) Yes  (  ) No
          developed for receipt of samples?  If yes, where
          are the SOPs documented (laboratory manual,
          written instructions, etc.)?

      D.  Are quality assurance procedures documented           ( ) Yes  (  ) No
          and available to the analysts?  If yes,  where
          are these documented?

      E.  Are written Standard Operating Procedures (SOPs)       ( ) Yes  (  ) No
          developed for compiling and maintaining  sample
          document files?  If yes, where are the SOPs
          documented (laboratory manual, written
          instructions, etc.)?

      F.  Are the magnetic tapes stored in a secure area?       ( ) Yes  (  ) No

      G.  Are samples that require preservation stored          ( ) Yes  (  ) No
          in such a way as to maintain their integrity?
          If yes, how are the samples stored?

      Documentation/Notebooks Checklist

      A.  Is a permanently bound notebook with preprinted,       ( ) Yes  (  ) No
          consecutively numbered pages being used?

      B.  Is the type of work clearly displayed on the          ( ) Yes  (  ) No
          notebook?

      C.  Is the notebook maintained in a legible  manner?       ( ) Yes  (  ) No

      D.  Are entries noting anomalies routinely recorded?       ( ) Yes  (  ) No

                            Figure 3.  (Continued).

                                      E-19

                                             ^ '•} 2
                                             \_7 «J C~*

-------
      E.  Has  Che analyst  avoided obliterating entries or the   ( ) Yes  (  )  No
          use  of a pencil?

      F.  Are  inserts (i.e. chromatograros, computer print'      ( ) Yes  (  )  No
          outs, etc.) permanently affixed to the notebook
          and  signed across insert edge and page?

      G.  Has  the supervisor of the individual maintaining the  ( ) Yes  (  )  No
          notebook personally examined and reviewed the notebook
          periodically, and signed his/her name therein, together
          with the date and appropriate comments as to whether or
          not  the notebook is being maintained in an appropriate
          manner?

      H.  Where applicable, is the notebook holder              ( ) Yes  (  )  No
          referencing reports or memoranda pertinent to
          the  contents of  an entry?

VI.   Quality  Control Manual Checklist

      Does the laboratory maintain a Quality Assurance/         ( ) Yes  (  )  No
      Quality  Control (QA/QC) Manual?

      Does the manual address the important elements            ( ) Yes  (  )  No
      of a QA/QC program,  including the following:

      A.  Personnel                                             ( ) Yes  (  ')  No

      B.  Facilities and equipment                              ( ) Yes  (  )  No

      C.  Operation of instruments                              ( ) Yes  (  )  No

      D.  Documentation of Procedures                           ( ) Yes  (  )  No

      E.  Procurement and inventory practices                   ( ) Yes  (  )  No

      F.  Preventive maintenance                                ( ) Yes  (  )  No

      G.  Reliability of data                                   ( ) Yes  (  )  No

      H.  Data validation                                       ( ) Yes  (  )  No

      I.  Feedback and corrective action                        ( ) Yes  (  )  No

      J.  Instrument calibration                                ( ) Yes  (  )  No

      K.  Recordkeeping                                         ( ) Yes  (  )  No

      L.  Internal audits                                       ( ) Yes  (  )  No

                            Figure 3.  (Continued).


                                      E-20

                                           C o o
                                           I/ J < >

-------
       Are QA/QC responsibilities and reporting relationships     (  )  Yes   (  )  No
       clearing defined?

       Have standard curves been adequately documented?           (  )  Yes   (  )  No

       Are laboratory standards traceable?                        (  )  Yes   (  )  No

       Are quality control charts maintained for each            (  )  Yes   (  )  No
       routine analysis?

       Do QC records show corrective action when                 (  )  Yes   (  )  No
       analytical results fail to meet QC criteria?

       Do supervisory personnel review the data and  QC results?   (  )  Yes   (  )  No

 VII.  Data Handling Checklist

       Are data calculations checked by a second person?        ( ) Yes  ( )  No

       Are data calculations documented?                        ( ) Yes  ( )  No

       Do records indicate corrective action that has           ( ) Yes  ( )  No
       been taken on projected data?

       Are limits of detection determined and reported          ( ) Yes  ( )  No
       properly?

       Are all data and records retained for the                ( ) Yes  ( )  No
       required amount of time?

       Are quality control data (e.g., standard curve           ( ) Yes  ( )  No
       duplicates) accessible for all analytical
       results?

VIII.  Summary

       Do responses to the evaluation indicate that              ( ) Yes  ( )  No
       project and supervisory personnel are aware
       of QA/QC and its application to the project?

       Do project and supervisory personnel place               ( ) Yes  ( )  No
       positive emphasis on QA/QC?

       Have responses with respect to QA/QC aspects  of          ( ) Yes  ( )  No
       the project been open and direct?

       Has a cooperative attitude been displayed by  all         ( ) Yes  ( )  No
       project and supervisory personnel?
                             Figure 3.   (Continued)
                                       E-21

                                            694

-------
Does the organization place the proper emphasis           (  ) Yes  ( ) No
on quality assurance?

Have any QA/QC deficiencies been discussed  before         (  ) Yes  ( ) No
leaving?

Is the overall quality assurance adequate to              (  ) Yes  ( ) No
accomplish the objectives of the project?

Have corrective actions recommended during                (  ) Yes  ( ) No
previous evaluations been implemented?

Are any corrective actions required?  If  so,              (  ) Yes  ( ) No
list the necessary actions below.
                                E-22
                                   631)

-------
                                   TECHNICAL REPORT DATA
                            (flettt read Instructions on lite reverse before completing)
1. REPORT NO.
                                                           3. RECIPIENT S ACCESSION NO.
4. TITLE AND SUBTITLE                                           8. REPORT DATE
  PROTOCOL FOR  THE ANALYSIS OF 2,3,7,8-TETRACHLORODIBENZOt
  p-DIOXIN BY HIGH-RESOLUTION GAS CHROMATOGRAPHY/HIGH-
  RESOLUTION MASS  SPECTROMETRY
              i. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
  J. S. Stanley and  T.  M.  Sack
                                                            . PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Midwest Research  Institute
  425 Volker Boulevard
  Kansas City, Missouri  64110
                                                            10. PROGRAM ELEMENT NO.
             11. CONTRACT/GRANT NO.

                Contract Number SAS  157 6X
12. SPONSORING AGENCY NAME AND ADDRESS
  Environmental Monitoring  Systems Laboratory  - LV,  NV
  Office of Research  and Development
  U.S. Environmental  Protection Agency
  Las Vegas, NV 89114
                                                            13. TYPE OF REPORT AND PERIOD COVERED
             14. SPONSORING AGENCY CODE

              EPA/600/07
16. SUPPLEMENTARY NOTES
  Project Officer - Werner F. Beckert, Environmental Monitoring Systems  Laboratory
                                                     Las Vegas, NV 89114
16. ABSTRACT
       An analytical  protocol for the determination of 2,3,7,8-tetrachlorodibenzo-p-
  dioxin (TCDD) and total TCDDs in soil, sediment  and aqueous samples  using high-
  resolution gas chromatography/high-resolution  mass spectrometry  (HRGC/HRMS) was
  developed using the best features of several candidate methods and input from experts
  in the field.  Preliminary tests led to refinements of the chromatographic cleanup
  procedures and corresponding changes in the protocol.  A final single-laboratory
  evaluation of the refined protocol, consisting of triplicate analyses  of five solid
  and five aqueous samples showed that the method  is useful for the determination of
  2,3,7,8-TCDD and total  TCDDs at concentrations from 10 to 200 pg/g (ppt) in soils and
  100 to 2,000 pg/L (ppq)  in aqueous samples.  Based on the data generated and on the
  evaluation of several options,  parts of the protocol were modified at  the EMSL-LV to
  lower the quantitation  limit for TCDD to 2 ppt in soil/sediments and to  20 ppq in
  aqueous samples.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS  C.  COSATI Field/Group

18. DISTRIBUTION STATEMENT

 RELEASE  TO PUBLIC
18. SECURITY CLASS (This Report)
    UNCLASSIFIED
                           21. NO. Of PAGES
20. SECURITY CLASS (Thispagel
    UNCLASSIFIED
                           22. PRICE
EPA f*rm 2220-1 (••». 4-77)   PMKVIOUS COITION i» OBSOLETE

-------
 136
Federal  Register / Vol. 49, No. 209 /  Friday.  October 26,  1984  / Rules  and Regulations
 Method 613—2,3,7,8-Tetrachlorodibenzo-p-
 Dioxin
 1. Scope and Application
   l.l  This method covers the determination
 of 2,3,7.8-tetrachlorodibenzo-p-dioxin (2.3,7,8-
 TCDD). The following parameter may be
 determined by this method:
Parameter
237 8-TCOD

STORET
No.
34675

GAS No.
1 746-01 -6

   1.2  This is a gas chromatographic/mass
 spectrometer (GC/MS) method applicable to
 the determination of 2,3,7.8-TCDD in
 municipal and industrial discharges as
 provided under 40 CFR 136.1. Method 625
 may be used to screen samples for 2.3,7,8-
 TCDD. When the screening test is positive.
 the final qualitative confirmation and
 quantification mlust be made using Method
 613.
   1.3  The method detection limit (MDL,
 defined in Section 14.1)' for 2,3.7.8-TCDD is
 listed in Table 1. The MDL for a specific
 wastewater may be different from that listed.
 depending upon the nature of interferences in
 the sample matrix.
   1.4  Because of the extreme toxicity of this
 compound, the analyst must prevent
 exposure to himself, of to others, by materials
 knows or believed to contain 2.3,7,8-TCDD.
 Section 4 of this method contains guidelines
 and protocols that serve as  minimum safe-
 handling standards  in a limited-access
 laboratory.
   1.5  Any modification of this method.
 beyond those expressly permitted, shall be
 considered as a major modification subject to
 application and approval of alternate test
 procedures under 40 CFR 136.4 and 136.5.
   1.6  This method is restricted to use by or
 under the supervision of analysts
 experienced in the use of a gas
 chromatograph/mass spectrometer and in the
 interpretation of mass spectra. Each analyst
 must demonstrate the ability to generate
 acceptable results with this method using the
 procedure described in Section 8.2.

 2. Summary of Method
   2.1   A measured volume of sample,
 approximately 1-L, is spiked with an internal
 standard of labeled  2.3,7.8-TCDD and
 extracted with methylene chloride using a
 separatory funnel. The methylene chloride
 extract is exchanged to hexane during
 concentration to a volume of 1.0 mL or less.
 The extract is then analyzed by capillary
 column CC/MS to separate  and measure
 2.3,7,8-TCDD."
   2.2   The method provides selected column
 chromatographic cleanup proceudres to aid in
 the elimination of interferences that may be
 encountered.

3. Interferences
  3.1   Method interferences may be caused
 by contaminants in solvents, reagents.
glassware, and other sample processing
hardware that lead to discrete artifacts and/
or elevated backgrounds at the masses (m/z)
monitored. All of these materials must be
routinely demonstrated to be free from
                               interferences under the conditions of the
                               analysis by running laboratory reagent
                               blanks as described in Section 8.1.3.
                                 3.1.1  Glassware must be scrupulously
                               cleaned.4 Clean all glassware as soon as
                               possible after use by rinsing with the last
                               solvent used in it. Solvent rinsing should be
                               followed by detergent washing with hot
                               water, and rinses with tap water and distilled
                               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 the treatment. Solvent rinses
                               with acetone and pesticide quality hexane
                               may be substituted for the muffle furnace
                               heating. Thorough rinsing with such solvents
                               usually eliminates PCB interference.
                               Volumetric  ware should not be heated in a
                               muffle furnace. After drying and cooling,
                               glassware should be sealed and stored in a
                               clean environment to prevent any
                               accumulation of dust or other contaminants.
                               Store inverted or capped with aluminum foil.
                                 3.1.2  The use of high purity reagents and
                               solvents helps to mininmize interference
                               problems. Purification of solvents by
                               distillation in all-glass systems may be
                               required.
                                 3.2 Matrix interferences may be caused
                               by contaminants that are coextracted from
                               the sample. The extent of matrix
                               interferences will vary considerably from
                               source to source, depending upon  the nature
                               and diversity of the industrial complex or
                               municipality being sampled. 2.3,7,8-TCDD is
                               often associated with other interfering
                               chlorinated compounds which are at
                               concentrations several magnitudes higher
                               than that of 2.3,7,8-TCDD. The cleanup
                               producers in Section 11 can be used to
                               overcome many of these interferences, but
                               unique samples may require additional
                               cleanup approaches '•>1to eliminate false
                               positives and achieve the MDL listed in Table
                               1.
                                 3.3  The primary column, SP-2330 or
                               equivalent,  resolves 2.3,7,8-TCDD  from the
                               other 21 TCDD insomers. Positive  results
                               using any other gas chromatographic column
                               must be confirmed using the primary column.

                               4. 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 laboratory is responsible for
                               maintaining a current awareness file of
                               OSHA regulations regarding the safe
                               handling of  the chemicals specified in this
                               method. A reference file of material data
                               handling sheets should also be made
                               available to all personnel involved in the
                               chemical analysis. Additional references to
                               laboratory safety are available and have
                               been identified •'•• for the information of the
                               analyst. Benzene and 2,3.7.8-TCDD have been
                               identified as suspected human or mammalian
                               carcinogens.
                                4.2  Each laboratory must develop a strict
                               safety program for handling 2,3,7,8-TCDD.
                               The following laboratory practices are
                               recommended:
  4.2.1  Contamination of the laboratory will
 be minimized by conducting all
 manipulations in a hood.
  4.2.2  The effluents of sample splitters for
 the gas chromatograph and roughing pumps
 on the GC/MS should pass through either a
 column of activated charcoal or be bubbled
 through a trap containing oil or high-boiling
 alcohols.
  4.2.3  Liquid waste should be dissolved in
 methanol or ethanol and irradiated with
 ultraviolet light with a wavelength greater
 than 290 nm for several days. (Use F 40 BL
 lamps or equivalent). Analyze  liquid wastes
 and dispose of the solutions when 2.3,7.8-
 TCDD can no longer be detected.
  4.3  Dow Chemical U.S.A. has issued the
 following precautions (revised November
 1978) for safe handling of 2,3,7,8-TCDD in the
 laboratory:
  4.3.1  The following statements on safe
 handling are as complete as possible on the
 basis of available toxicological information.
 The precautions for safe handling and use an:
 necessarily  general in nature since detailed.
 specific recommendations can be made only
 for the particular exposure and circumstances
 of each individual use. Inquiries about
 specific operations or uses may be addressed
 to the Dow  Chemical Company. Assistance in
 evaluating the health hazards of particular
 plant conditions may be obtained from
 certain consulting laboratories and from
 State Departments.of Health or of Labor.
 many of which have an industrial health
 service. 2.3,7,8-TCDD  is extremely toxic to
 laboratory animals. However,  it has been
 handled for years without injury in analytical
 and biological laboratories. Techniques used
 in handling  radioactive and infectious
 materials are applicable to 2,3,7,8,-TCDD.
  4.3.1.1  Protective equipment—Throw-
 away plastic gloves, apron or lab coat, safety
 glasses, and a lab hood adequate for
 radioactive  work.
  4.3.1.2  Training—Workers must be
 trained in the proper method of removing
 contaminated gloves and clothing without
 contacting the exterior surfaces.
  4.3.1.3  Personal hygiene—Thorough
 washing of hands and forearms after each
 manipulation and before breaks (coffee.
 lunch, and shift).
  4.3.1.4  Confinement—Isolated work area.
 posted with signs, segregated glassware and
 tools, plastic-backed absorbent paper on
 benchtops.
  4.3.1.5  Waste—Good technique includes
 minimizing contaminated waste. Plastic bag
 liners should be used in waste  cans. Janitors
 must be trained in the safe handling of waste.
  4.3.1.6  Disposal of  wastes—2,3.7,8-TCDD
 decomposes above 800 "C. Low-level waste
 such as absorbent paper, tissues, animal
 remains, and plastic gloves may be burned in
 a good incinerator. Gross quantities
 (milligrams) should be packaged securely and
 disposed through commercial or
 governmental channels which are capable of
 handling high-level radioactive wastes or
 extremely toxic wastes. Liquids should be
allowed to evaporate in a good hood and in a
disposable container. Residues may then be
handled as above.
                                                                   G 9 7

-------
               Federal  Register /  Vol. 49.  No.  209  / Friday, October  26, 1984 /  Rules and  Regulations         137
              ^^^M^^B^H^HM^BIHH^B^MBB^H^^BHMiMVM^MB^HMMMM^^^^H^^^^HM^^MMi^^B^^Mi^^^^i^MHlHi^Mai^^^>M^Bi^^aHM^H^HHMH'**M^M|HM^H
V.  4.3.1.7  Decontamination—For personal
  decontamination, use any mild soap with
  plenty of scrubbing action. For
  decontamination of glassware, tools, and
  surfaces. Chlorothene NU Solvent
  (Trademark of the Dow Chemical Company)
  is the least toxic solvent shown to be
  effective. Satisfactory cleaning may be
  accomplished by rinsing with Chlorothene.
  then washing with any detergent and water.
  Dishwater may be disposed to the sewer. It is
  prudent to minimize solvent wastes because
  they may require special disposal through
  commercial sources  which are expensive.
    4.3.1.8  Laundry—Clothing known to be
  contaminated should be disposed with the
  precautions described under Section 4.3.1.6.
  Lab coats or other clothing worn in 2.3.7.8-
  TCDD work areas may be laundered.
    Clothing should be collected in plastic
  bags. Persons who convey the bags and
  launder the clothing should be advised of the
  hazard and trained in proper handling. The
  clothing may be put  into a washer without
  contact if the launderer knows the problem.
  The washer should be run through a cycle •
  before being used again for other clothing.
    4.3.1.9  Wipe tests—A useful method of
  determining cleanliness of work surfaces and
  tools is to wipe the surface with a piece of
  filter paper. Extraction and analysis by gas
  chromatography can achieve a limit of
  sensitivity of 0.1 fig per wipe. Less than 1 jig
  of 2.3,7,8-TCDD per sample indicates
  acceptable cleanliness: anything higher
  warrants further cleaning. More than 10 jig
  on a wipe sample constitutes an acute hazard
  and requires prompt cleaning before further
  use of the equipment or work space. A high
  (>10 jig) 2,3,7.8-TCDD level indicates  that
  unacceptable work practices have been
  employed in the past.
   4.3.1.10  Inhalation—Any procedure that
  may produce airborne contamination must be
  done with good ventilation. Gross losses to a
  ventilation system must not be allowed.
  Handling of the dilute solutions normally
  used in analytical and animal work presents
  no inhalation hazards except in the case of
  an accident.
   4.3.1.11  Accidents—Remove
  contaminated clothing immediately, taking
  precautions not to contaminate skin or other
  articles. Wash exposed skin vigorously and
  repeatedly until medical attention is
  obtained.

  5. Apparatus and Materials
   5.1  Sampling equipment, for discrete or
  composite sampling.
   5.1.1  Grab sample bottle—1-L or 1-qt,
  amber glass, fitted with a screw cap lined
  with Teflon. Foil may be substituted for
  Teflon if the sample is not corrosive. If amber
  bottles are not available, protect samples
  from light. The bottle and cap liner must be
  washed, rinsed with acetone or methylene
  chloride, and dried before use to minimize
 contamination.
   5.1.2  Automatic sampler (optional)—The
 sampler must incorporate glass sample
 containers for the collection of a minimum of
 250 mL of sample. Sample containers must be
 kept refrigerated at 4 °C and protected  from
 light during compositing. If the sampler uses a
 peristaltic pump, a minimum length of
 compressible silicone rubber tubing may be
 used. Before use, however, the compressible
 tubing should be thoroughly rinsed with
 methanol, followed by repeated rinsings with
 distilled water to minimize the potential for
 contamination of the sample. An integrating
 flow meter is required to collect flow
 proportional composites.
   5.1.3  Clearly label all samples as
 "POISON" and ship according to U.S.
 Department of Transportation regulations.
   5.2  Glassware (All specifications are
 suggested. Catalog numbers are included for
 illustration only.):
   5.2.1  Separatory funnels—2-L and 125-mL,
 with Teflon stopcock.
   5.2.2  Concentrator tube, Kuderna-
 Danish—lOmL graduated (Kontes K-570050-
 1025 or equivalent). Calibration  must be
 checked at the volumes employed in the test.
 Ground glass stopper is used to  prevent
 evaporation of extracts.
   5.2.3  Evaporative flask. Kuderna-
 Danish—500-mL (Kontes K-570001-0500 or
 equivalent). Attach to concentrator tube with
 springs.
   5.2.4  Snyder column. Kuderna-Danish—
 Three-ball macro (Kontes K-503000-0121 or
 equivalent).
   5.2,5  Snyder column. Kuderna-Danish—
 Two-ball micro (Kontes K-569001-0219 or
 equivalent).
   5.2.6  Vials—10 to 15-mL, amber glass.
 with Teflon-lined screw cap.
   5.2.7  Chromatographic column—300 mm
 long X 10 mm ID, with Teflon stopcock and
 coarse frit filter disc at bottom.
   5.2.8  Chromatographic column—400 mm
 long X 11 mm ID, with Teflon stopcock and
 coarse frit filter disc at bottom.
   5.3  Boiling chips—Approximately 10/40
 mesh. Heat to 400 °C for 30 min or Soxhlet
 extract with methylene chloride.
   5.4  Water bath—Heated, with concentric
 ring cover, capable of temperature control
 (±2 *C). The bath should be used in a hood.
   5.5  GC/MS system:
   5.5.1  Gas chromatograph—An analytical
 system complete with a temperature
 programmable gas chromatograph and all
 required accessories including syringes,
 analytical columns, and gases. The injection
 port must be designed for capillary columns.
 Either split, splitless. or on-column injection
 techniques may be employed, as long as the
 requirements of Section 7.1.1 are achieved.
   5.5.2  Column—60 m long X 0.25 mm ID
 glass or fused silica, coated with SP-2330 (or
 equivalent) with a Him thickness of 0.2 u.m.
 Any equivalent column must resolve 2, 3, 7.
 8-TCDD from the other 21 TCDD isomers.16
   5.5.3  Mass spectrometer—Either a low
 resolution mass spectrometer (LRMS) or a
 high resolution mass spectrometer (HRMS)
 may be used. The mass spectrometer must be
 equipped with a 70 V (nominal) ion source
 and be capable  of aquiring m/z abundance
 data in real time selected ion monitoring
 (SIM) for groups of four or more masses.
  5.5.4  GC/MS interface—Any GC to MS
 interface can be used that achieves the
 requirements of Section 7.1.1. GC to  MS
 interfaces constructed of all glass or glass-
lined materials are recommended. Glass
surfaces can be  deactivated by silanizing
with dichlorodimethylsilane. To achieve
 maximum sensitivity, the exit end of the
 capillary column should be placed in the ion
 source. A short piece of fused silica capillary
 can be used as the interface to overcome
 problems associated with straightening the
 exit end of glass capillary columns.
  5.5.5  The SIM data acquired during the
 Chromatographic program is defined as the
 Selected Ion Current Profile (SICP). The SICP
 can be acquired under computer control or as
 a real time analog output. If computer control
 is used, there must be software available to
 plot the SICP and report peak height or area
 data for any m/z in the SICP between
 specified time or scan number limits.
  5.6  Balance—Analytical, capable of
 accurately weighing 0.0001 g.

 6. Reagents
  6.1   Reagent water—Reagent water is
 defined as a water in which an interferent is
 not observed at the MDL of 2. 3. 7. 8-TCDD.
  6.2  Sodium hydroxide solution (10 N)—
 Dissolve 40 g of NaOH  (ACS) in reagent
 water and dilute to 100 mL. Wash the
 solution with methylene chloride and hexane
 before use.
  6.3  Sodium thiosulfate—(ACS) Granular.
  6.4  Sulfuric acid—Concentrated (ACS. sp.
 gr.  1.84).
  6.5  Acetone, methylene chloride, hexane,
 benzene, ortho-xylene,  tetradecane—
 Pesticide quality or equivalent.
  6.6  Sodium sulfate—(ACS) Granular,
 anhydrous. Purify by heating at 400 °C for 4 h
 in a shallow tray.
  6.7  Alumina—Neutral. 80/200 mesh
 (Fisher Scientific Co.. No. A-540 or
 equivalent). Before use, activate for 24 h at
 130 'C in a foil-covered glass container.
  • 6.8   Silica gel—High purity grade, 100/120
 mesh (Fisher Scientific  Co., No. S-679 or
 equivalent).
  6.9   Stock standard solutions (1.00 jig/
 fiL)—Stock standard solutions can be
 prepared from pure standard materials or
 purchased as certified solutions. Acetone
 should be used as the solvent for spiking
 solutions: ortho-xylene  is recommended for
 calibration standards for split injectors: and
 tetradecane  is recommended for splitless or
 on-colum injectors. Analyze stock internal
 standards to verify the absence of native
 2,3,7.8-TCDD.
  6.9.1  Prepare stock standard solutions of
 2.3,7,8-TCDD (mol wt 320) and either 37Cl4
 2.3,7,8-TCDD (mol wt 328) or 13Cl,j 2.3.7.8-
 TCDD (mol wt 332) in an isolated area by
 accurately weighing about 0.0100 g of pure
 material. Dissolve the material in pesticide
 quality solvent and dilute to volume in a 10-
 mL volumetric flask. When compound purity
 is assayed to be 96% or  greater, the weight
 can be used  without correction to calculate
 the concentration of the stock standard.
 Commercially prepared stock standards can
 be used at any concentration if they are
certified by the manufacturer or by an
 independent source.
  6.9.2  Transfer the stock standard
solutions into Teflon-sealed screw-cap
bottles. Store in an isolated refrigerator
protected from light. Stock standard solutions
should be checked frequently for signs of
degradation or evaporation, especially just
                                                                     698

-------
 138         Federal Register  /  Vol.  49. No. 209 /  Friday.  October 26.  1984  / Rules  and Regulations
 prior to preparing calibration standards or
 spiking solutions from them.
   6.9.3  Stock standard solutions must be
 replaced after six months, or sooner if
 comparison with check standards indicates a
 problem.
   6.10   Internal standard spiking solution (25
 ng/mL)—Using stock standard solution,
 prepare a spiking solution in acetone of
 either'3Cl,2 or "Cl4 2,3.7.8-TCDD at a
 concentration of 25 ng/mL. (See Section 10.2)
   6.11   Quality control check sample
 concentrate—See Section 8.2.1.

 7.   Calibration
   7.1   Establish gas chromatograhic
 operating conditions equivalent to those
 given in Table 1 and SIM  conditions for the
 mass spectrometer as described in Section
 12.2 The GC/MS system must be calibrated
 using the internal standard technique.
   7.1.1   Using stock standards,  prepare
 calibration standards that will allow
 measurement of relative response factors of
 at least three concentration ratios of 2,3,7,8-
 TCOO  to internal  standard. Each calibration
 standard must be prepared to contain the
 internal standard  at a concentration of 25 ng/
 mL. If any interferences are contributed by
 the internal standard at m/z 320 and 322, its
 concentration may be reduced in the
 calibration standards and in the internal
 standard spiking solution (Section 6.10). One
 of the calibration  standards should contain
 2,3,7,8-TCDD  at a  concentration near,  but
 above,  the MDL and the other 2.3,7,8-TCDD
 concentrations should correspond to the
 expected range of concentrations found in
 real samples or should define the working
 range of the GC/MS system.
   7.1.2   Using injections of 2 to 5 fiL, analyze
 each calibration standard according to
 Section 12 and tabulate peak height or area
 response against the concentration of 2,3,7,8-
 TCDD and internal standard. Calculate
 response factors (RF) for 2,3,7,8-TCDD using
 Equation 1.
   Equation 1.
             RF="
(A.) (Cu)

(A(s) (C.)
 where:
   A,=SIM response for 2,3,7,8-TCDD m/z
     320.
   Ai5 = SIM response for the internal
     standard, m/z 332 for 13C,2 2,3.7.8-TCDD
     m/z 328 for "C14 2,3,7,8-TCDD.
   Cu= Concentration of the internal standard
  C, = Concentration of 2,3,7,8-TCDD (fig/L).
If the RF value over the working range is a
constant (< 10% relative standard deviation,
RSD). the RF can be assumed to be invariant
and the average RF can be used for
calculations. Alternatively, the results can be
used to plot a calibration curve of response
ratios, A./AU, vs. RF.
  7.1.3  The working calibration curve or RF
must be verified on each working day by the
measurement of one or more 2,3.7,8-TCDD
calibration standards. If the response for
2,3,7,8-TCDD varies from the predicted
response by more than ±15%, the test must
 be repeated using a fresh calibration
 standard. Alternatively, a new calibration
 curve must be prepared.
   7.2  Before using any cleanup procedure.
 the analyst must process a series of
 calibration standards through the procedure
 to validate elution patterns and the absence
 of interferences from the reagents.

 8. Quality Control
   8.1  Each laboratory that uses this method
 is required to operate a formal quality control
 program. The minimum requirements of this
 program consist of an initial demonstration of
 laboratory capability and an ongoing
 analysis of spiked samples to evaluate and
 document  data quality. The laboratory must
 maintain records to document the quality of
 data that is generated. Ongoing data quality
 checks are compared with established
 performance criteria to determine if the
 results of analyses meet the performance
 characteristics of the method. When results
 of sample  spikes indicate atypical method
 performance, a quality control check   .
 standard must be analyzed to confirm that
 the measurements were performed in an in-
 control mode of operation.
   8.1.1   The analyst must make an initial,
 one-time, demonstration of the ability to
 generate acceptable accuracy and precision
 with this method. This ability is  established
 as described in Section 8.2.
   8.1.2   In recognition of advances that are
 occurring in chromatography, the analyst is
 permitted certain options  (detailed in
 Sections 10.5,11.1, and 12.1) to improve the
 separations or lower the cost of
 measurements. Each time such a modification
 is made to the method, the analyst is required
 to repeat the procedure in Section 8.2
   8.1.3   Before processing any samples, the
 analyst must analyze a reagent water  blank
 to demonstrate that interferences from the
 analytical  system and glassware are under
 control. Each time a set of samples is
 extracted or reagents are changed, a reagent
 water blank must be processed as a
 safeguard  against laboratory contamination.
   8.1.4   The laboratory must, on an ongoing
 basis, spike and analyze a minimum of 10% of
 all samples with native 2.3,7,8-TCDD to
 monitor and evaluate laboratory data  quality.
 This procedure is described in Section 8.3.
   8.1.5  The  laboratory must, on an ongoing
 basis, demonstrate through the analyses of
 quality control check standards that the
 operation of the measurement system is in
 control. This procedure is  described in
 Section 8.4. The frequency of the check
 standard analyses is equivalent to 10% of all
 samples analyzed but may be reduced if
 spike recoveries from samples (Section 8.3)
 meet all specified quality control criteria.
  8.1.6  The laboratory must maintain
 performance records to document the quality
 of data that is generated. This procedure is
 described in Section 8.5.
  8.2 To establish the ability to generate
 acceptable accuracy and precision, the
 analyst must perform the following
 operations.
  8.2.1  A  quality control  (QC) check sample
concentrate is required containing 2,3,7,8-
TCDD at a  concentration of 0.100 fig/mL in
acetone. The QC check sample concentrate
must be obtained from the U.S.
Environmental Protection Agency,
Environmental Monitoring and Support
Laboratory in Cincinnati, Ohio, if available. If
not available from that source, the QC check
sample concentrate must be obtained from
another external source. If not available from
either source above, the  QC check sample
concentrate must be prepared by the
laboratory using stock standards prepared
independently from those used for
calibration.
  8.2.2  Using a pipet, prepare QC check
samples at a concentration of 0.100 fig/L (100
ng/L] by adding 1.00 mL of QC check sample
concentrate to each of four 1-L aliquots of
reagent water.
  8.2.3  Analyze the well-mixed QC check
samples according to the method beginning in
Section 10.
  8.2.4  Calculate the average recovery (X)
in ng/L. and the standard deviation of the
recovery (s) in u£/L, for  2,3.7,8-TCDD using
the four results.
  8.2.5  Compare s and  (X) with the
corresponding acceptance criteria for
precision and accuracy,  respectively, found in
Table 2. If s and X meet  the acceptance
criteria, the system performance is
acceptable and analysis of actual samples
can begin. If s exceeds the precision limit or
X falls outside the range for accuracy, the
system performance is unacceptable for
2,3,7,8-TCDD. Locate and correct the source
of the problem and repeat the test beginning
with Section 8.2.2.
  8.3   The laboratory must, on an ongoing
basis, spike at least 10% of the samples from
each sample site being monitored to assess
accuracy. For laboratories analyzing one to
ten samples per month, at least one spiked
sample per month is required.
  8.3.1  The concentration of the spike in the
sample should be determined as follows:
  8.3.1.1   If, as in compliance monitoring, the
concentration of 2,3,7,8-TCDD in the sample
is being checked against a regulatory
concentration limit, the spike should be at
that limit or 1 to 5 times higher than the
background concentration determined in
Section 8.3.2. whichever  concentration would
be larger.
  8.3.1.2   If the concentration of 2,3,7,8-
TCDD in the sample is not being checked
against a limit specific to that parameter, the
spike should be at 0.100 fig/L or  1 to 5 times
higher than the background concentration
determined in Section 8.3.2. whichever
concentration would be larger.
  8.3.1.3   If it is impractical to determine
background levels before spiking (e.g.,
maximum holding times will be exceeded),
the spike concentration should be (1) the
regulatory concentration limit, if any: or, if
none (2) the larger of either 5 times higher
than the expected background concentration
or 0.100 ng/L.
  8.3.2  Analyze one sample aliquot to
determine the background concentration (B)
of 2,3.7,8-TCDD. If necessary, prepare a new
QC check sample concentrate (Section 8.2.1)
appropriate for the background concentration
in the sample. Spike a second sample aliquot
with 1.0 mL of the QC check sample
concentrate and analyze it to determine the
                                                                  699

-------
              Federal Register  /  Vol.  49, No. 209  / Friday. October 26. 1984 /  Rules and Regulations         139
 concentration after spiking (A) of 2,3.7.8-
 TCDD. Calculate percent recovery (P) as
 100(A-B)%T, where T is the known true value
 of the spike.
   8.3.3  Compare the percent recovery (P) for
 2.3.7,8-TCDD with the corresponding QC
 acceptance criteria found in Table 2. These
 acceptance criteria were calculated to
 include an allowance for error in
 measurement of both the background and
 spike concentrations, assuming a spike to
 background ratio of 5:1. This error will be
 accounted for to the extent that the analyst's
 spike to background ratio approaches 5:1."If
 spiking was performed at a concentration
 lower than 0.100 ftg/L, the analyst must use
 either the QC acceptance criteria in Table 2.
 or optional QC acceptance criteria calculated
 for the specific spike concentration. To
 calculate optional acceptance criteria for the
 recovery of 2.3,7,8-TCDD: (1) calculate
 accuracy (X') using the equation in Table 3.
 substituting the spike concentration (T) for C:
 (2) calculate overall precision (S') using the
 equation in Table 3. substituting X' for X; (3)
 calculate the range for recovery at the spike
 concentration as (100 X'/T)±2.44(100 S'f
 T)%. "
   8.3.4  If the recovery of 2,3,7,8-TCDD falls
 outside the designated range for recovery, a
 check standard must be analyzed as
 described in Section 8.4.
   8.4  If the recovery of 2.3,7,8-TCDD fails
 the acceptance criteria for recovery in
 Section 8.3, a QC check standard must be
 prepared and analyzed.
   Note.—The frequency for  the required
 analysis of a QC check standard will depend
 upon the complexity of the sample matrix
 and the performance of the laboratory.
   8.4.1   Prepare the QC check standard by
 adding 1.0 mL of QC check sample
 concentrate (Section 8.2.1 or 8.3.2) to 1 L of
 reagent water.
   8.4.2  Analyze the QC check standard to
 determine the concentration measured (A) of
 2,3,7,8-TCDD. Calculate the percent recovery
 (P,) as 100 (A/T)%, where T is  the true value
 of the standard concentration.
   8.4.3  Compare the percent recovery (P.)
 with the corresponding QC acceptance
 criteria found in Table 2. If the recovery of
 2.3.7,8-TCDD falls outside the designated
 range, the laboratory performance is judged
 to be out of control, and the problem must be
 immediately identified and corrected. The
 analytical result for 2,3,7,8-TCDD in the
 unspiked sample is suspect and may not be
 reported for regulatory compliance purposes.
  8.5  As part of the QC program for the
 laboratory, method accuracy for wastewater
 samples must be assessed and records must
 be maintained. After the analysis of five
 spiked wastewater samples as in Section 8.3,
 calculate the average percent recovery (P)
 and the spandard deviation of the percent
 recovery (sp). Express the accuracy
 assessment as a percent recovery interval
 from P-2sp to P + 2sD. If P=90% and sp=10%,
 for example, the accuracy interval is
 expressed as 70-110%. Update the accuracy
assessment on a regular  basis (e.g. after each
five to ten new accuracy measurements).
  8.6  It is recommended that the
laborataory adopt additional quality
assurance practices for use with this method.
 The specific practices that are most
 productive depend upon the needs of the
 laboratory and the nature of the samples.
 Field duplicates may be analyzed to assess
 the precision of the environmental
 measurements. Whenever possible, the
 laboratory should analyze standard reference
 materials and participate in relevant
 performance evaluation studies.

 9. Sample Collection. Preservation, and
 Handling
   9.1  Grab samples must be collected in
 glass containers. Conventional sampling
 practices " should be followed, except that
 the bottle must not be prerinsed with sample
 before collection. Composite samples should
 be collected in refrigerated glass containers
 in accordance with the requirements of the
 program. Automatic sampling equipment
 must be as free as possible of Tygon tubing
 and other potential sources of contamination.
   9.2  All samples must be iced or
 refrigerated at 4 "C and protected from light
 from the time of collection until extraction.
 Fill the sample bottles and. if residual
 chlorine is present, add 80 mg of sodium
 thiosulfate per liter of sample and mix well.
 EPA Methods 330.4 and 330.5 may be used for
 measurement of residual chlorine.13 Field test
 kits are available for this purpose.
   9.3  Label all samples and containers
 "POISON" and ship according to applicable
 U.S. Department of Transportation
 regulations.
   9.4  All samples must be extracted within
 7 days of collection and completely analyzed
 within 40 days of extraction.2

 10. Sample Extraction
   Caution: When using this method to
 analyze  for 2,3,7,8-TCDD, all of the following
 operations must be performed in a limited-
 access laboratory with the analyst wearing
 full protective covering for all exposed skin
 surfaces. See Section 4.2.
   10.1  Mark the water meniscus on the side
 of the sample bottle for later determination of
 sample volume. Pour the entire sample into a
 2-L separatory funnel.
   10.2  Add 1.00 mL of internal standard
 spiking solution to the sample in the
 separatory funnel. If the final extract will be
 concentrated to a fixed volume below 1.00
 mL (Section 12.3), only that volume of spiking
 solution  should be added to the sample so
 that the final extract will contain 25 ng/mL of
 internal standard at the time of analysis.
  10.3  Add 60 mL of methylene chloride to
 the sample bottle, seal, and shake 30 s to
 rinse the inner surface. Transfer the solvent
 to the separatory funnel and extract the
 sample by shaking the funnel for 2 min with
 periodic  venting to release excess pressure.
 Allow the organic layer to separate from the
 water phase for a minimum of 10 min. If the
 emulsion interface between layers is more
 than one-third the volume of the solvent
 layer, the analyst must employ mechanical
 techniques to complete the phase separation.
 The optimum technique depends upon  the
 sample, but may include stirring, filtration of
 the emulsion through glass wool,
centrifugation, or other physical methods.
Collect the methylene chloride extract  in a
250-mL Erlenmeyer flask.
  10.4  Add a second 60-mL volume of
 methylene chloride to the sample bottle and
 repeat the extraction procedure a second
 time, combining the extracts in the
 Erlenmeyer flask. Perform a third extraction
 in the same manner.
  10.5  Assemble a Kudema-Danish (K-D)
 concentrator by attaching a 10-mL
 concentrator tube to a 500-mL evaporative
 flask. Other concentration devices or
 techniques may be used in place of the K-D
 concentrator if the requirements of Section
 8.2 are met.
  10.6  Pour the combined extract into the
 K-D concentrator. Rinse the Erlenmeyer flask
 with 20 to 30 mL of methylene chloride to
 complete the quantitative transfer.
  10.7  Add one or two clean boiling chips to
 the evaporative flask and attach a three-ball
 Snyder column. Prewet the Snyder column by
 adding about 1 mL of methylene chloride to
 the top. Place the K-D apparatus on a hot
 water bath (60 to 65 *C) so  that the
 concentrator tube is partially immersed in the
 hot water, and the entire lower rounded
 surface of the flask is bathed with hot vapor.
 Adjust the vertical position of the apparatus
 and the water temperature as required to
 complete the concentration in 15 to 20 min. At
 the proper rate of distillation the balls of the
 column will actively chatter but the chambers
 will not flood with condensed solvent. When
 the apparent volume of liquid reaches 1 mL,
 remove the K-D apparatus and allow it to
 drain and cool for at least 10 min.
  10.8  Momentarily remove the Snyder
 column, add 50 mL of hexane and a new
 boiling chip, and reattach the Snyder column.
 Raise the temperature of the water bath to 85
 to 90'C. Concentrate the extract as in Section
 10.7, except use hexane to prewet the column.
 Remove the Snyder column and rinse the
 flask and its lower joint into the concentrator
 tube with 1 to 2 mL of hexane. A 5-mL syringe
 is recommended for this operation. Set aside
 the K-D glassware for reuse in Section 10.14.
  10.9  Pour the hexane extract from the
 concentrator tube into a 125-mL separatory
 funnel. Rinse the concentrator tube four times
 with 10-mL aliquots of hexane. Combine all
 rinses in the 125-mL separatory funnel.
  10.10   Add 50 mL of sodium hydroxide
 solution to the funnel and shake for 30 to 60 s.
 Discard the aqueous phase.
  10.11   Perform a second  wash of the
 organic layer with 50 mL of reagent water.
 Discard the aqueous phase.
  10.12   Wash the hexane  layer with a least
 two 50-mL aliquots of concentrated sulfuric
 acid. Continue washing the hexane layer with
 50-mL aliquots of concentrated sulfuric acid
 until the acid layer remains colorless. Discard
 all acid  fractions.
  10.13   Wash the hexane layer with two 50-
 mL aliquots of reagent water. Discard the
 aqueous phases.
  10.14   Transfer the hexane extract into  a
125-mL Erlenmeyer flask containing 1 to 2  g
of anhydrous sodium sulfate. Swirl the flask
for 30 s and decant the hexane extract  into
the  reassembled K-D apparatus. Complete
the  quantitative transfer with two 10-mL
hexane rinses of the Erlenmeyer flask.
                                                                   700

-------
 140
Federal  Register / Vol. 49. No. 209 /  Friday,  October 26,  1984  / Rules and  Regulations
   10.15  Replace the one or two clean boiling
 chips and concentrate the extract to 6 to 10
 mL as in Section 10.8.
   10.16  Add a clean boiling chip to the
 concentrator tube and attach a two-ball
 micro-Snyder column. Prewet the column by
 adding about  1 mL of hexane to the top. Place
 the micro-K-D apparatus on the water bath
 so that the concentrator tube is partially
 immersed in the hot water. Adjust the
 vertical position of the apparatus and the
 water temperature as required to complete
 the concentration in 5 to 10 min. At the
 proper rate of distillation the balls of the
 column will actively chatter but the chambers
 will not flood. When the apparent volume of
 liquid reaches about 0.5 mL. remove the K-D
 apparatus and allow it to drain and cool for
 at least 10 min. Remove the micro-Snyder
 column and rinse its lower joint into the
 concentrator tube with 0.2 mL of hexane.
   Adjust the extract volume to 1.0 mL with
 hexane. Stopper the concentrator tube and
 store refrigerated and protected from light if
 further processing will not be performed
 immediately. If the extract will be stored
 longer than two days, it should be transferred
 to a Teflon-sealed screw-cap vial. If the
 sample extract requires no further cleanup,
 proceed with GC/MS analysis  (Section 12). If
 the sample requires further cleanup, proceed
 to Section 11.
   10.17  Determine the original sample
 volume by refilling  the sample bottle to the
 mark and transferring the liquid to a 1000-mL
 graduated cylinder. Record the sample
 volume to the nearest 5 mL.

 11. Cleanup and Separation
   11.1  Cleanup procedures may not be
 necessary for  a relatively clean sample
 matrix. If particular circumstances demand
 the use of a cleanup procedure, the analyst
 may use either procedure below or any other
 appropriate procedure.us'7However, the
 analyst first must demonstrate  that the
 requirements of Section 8.2 can be met using
 the method as revised to incorporate the
 cleanup procedure.  Two cleanup column
 options are offered to th° analyst in this
 section. The alumina column should be  used
 first to overcome interferences. If background
 problems are still encountered, the silica gel
 column may be helpful.
   11.2  Alumina column cleanup for 2,3,7,8-
 TCDD:
   11.2.1  Fill a 300 mm long x 10 mm ID
 chromatographic column with activated
 alumina to the 150 mm level. Tap the column
 gently to settle the alumina and add 10 mm of
 anhydrous sodium sulfate to the top.
   11.2.2  Preelute the column with 50 mL of
 hexane. Adjust the elution rate to 1 mL/min.
 Discard the eluate and just prior to exposure
 of the sodium sulfate layer to the air,
 quantitatively  transfer the 1.0-mL sample
 extract onto the column using two 2-mL
 portions of hexane to complete the transfer.
  11.2.3  Just prior to exposure of the sodium
 sulfate layer to the air, add 50 mL of 3%
 methylene chloride/97% hexane (V/V) and
 continue the elution of the column. Discard
 the eluate.
  11.2.4  Next, elute the column with 50 mL
of 20% methylene chloride/80% hexane (V/V)
into a 500-mL K-D flask equipped with a 10-
                               mL concentrator tube. Concentrate the
                               collected fraction to 1.0 mL as in Section
                               10.16 and analyze by GC/MS (Section 12).
                                 11.3  Silica gel column cleanup for 2.3.7,8-
                               TCDD:
                                 11.3.1 Fill a 400  mm long x 11 mm ID
                               chromatographic column with silica gel to the
                               300 mm level. Tap the column gently to settle
                               the silica gel and add 10 mm of anhydrous
                               sodium sulfate to the top.
                                 11.3.2 Preelute the column with 50 mL of
                               20% benzene/80% hexane (V/V). Adjust the
                               elution rate to 1 mL/min. Discard the eluate
                               and just prior to exposure of the sodium
                               sulfate layer to the  air, quantitatively transfer
                               the 1.0-mL sample extract onto the column
                               using two 2-mL portions of 20% benzene/80%
                               hexane to complete the transfer.
                                 11.3.3 Just prior to exposure of the sodium
                               sulfate layer to the  air, add 40 mL of 20%
                               benzene/80% hexane to the column. Collect
                               the eluate in a clean 500-mL K-D flask
                               equipped with a 10-mL concentrator tube.
                               Concentrate the collected fraction to 1.0 mL
                               as in Section 10.16 and analyze by GG/MS.

                               12. GC/MS Analysis
                                 12.1. Table 1 summarizes the
                               recommended operating conditions for the
                               gas chromatograph. Included in this table are
                               retention times and MDL that can be
                               achieved under these conditions. Other
                               capillary columns or chromatographic
                               conditions may be used if the requirements of
                               Sections 5.5.2 and 8.2 are met.
                                 12.2  Analyze standards and samples with
                               the mass spectrometer operating in the
                               selected ion monitoring (SIM) mode using a
                               dwell time to give at least seven points per
                               peak. For LRMS, use masses at m/z 320, 322,
                               and 257 for 2,3,7,8-TCDD and either m/z 328
                               for "CU 2,3,7,8-TCDD or m/z 332 for I3C,2
                               2,3,7,8-TCDD. For HRMS, use masses at m/z
                               319.8965 and 321.8936 for 2,3,7,8-TCDD and
                               either m/z 327.8847 for "CU 2,3,7,8-TCDD or
                               m/z 331.9367 for 13Cu 2,3,7,8-TCDD.
                                 12.3  If lower detection limits are required,
                               the extract may be carefully evaporated to
                               dryness under a gentle stream of nitrogen
                               with the concentrator tube in a water bath at
                               about 40 'C. Conduct this operation
                               immediately before GC/MS analysis.
                               Redissolve the extract in the desired final
                               volume of ortho-xylene or tetradecane.
                                 12.4  Calibrate the system daily as
                               described in Section 7.
                                 12.5  Inject 2 to 5 /xL of the sample extract
                               into the gas chromatograph. The volume of
                               calibration standard injected must  be
                               measured, or be the same as all sample
                               injection volumes.
                                 12.6  The presence of 2,3,7,8-TCDD is
                               qualitatively confirmed if all of the following
                               criteria are achieved:
                                 12.6.1  The gas chromatographic column
                               must resolve 2,3,7,8-TCDD from the other 21
                               TCDD isomers.
                                 12.6.2  The masses for native 2,3.7,8-TCDD
                               (LRMS-m/z 320, 322. and 257 and HRMS-m/z
                               320 and 322) and labeled 2,3,7,8-TCDD (m/z
                               328 or 332) must exhibit a simultaneous
                               maximum at a retention time that matches
                               that of native 2,3,7,8-TCDD in the calibration
                               standa-d, with the performance specifications
                               of the analytical system.
  12.6.3  The chlorine isotope ratio at m/z
320 and m/z 322 must agree to within±10% of
that in the calibration standard.
  12.6.4  The signal of all peaks must  be
greater than 2.5 times the noise level.
  12.7  For quantitation, measure the
response of the m/z 320 peak for 2.3,7.8-
TCDD and the m/z 332 peak for 13C,2 2.3.7.8-
TCDD or the m/z 328 peak for "CU 2.3.7,8-
TCDD.
  12.8  Co-eluting impurities are suspected if
all criteria are achieved except those in
Section 12.6.3. In this case, another SIM
analysis using masses at m/z 257, 259, 320
and either m/a 328 or m/z 322 can be
performed. The masses at m/z 257 and m/z
259 are indicative of the loss of one chlorine
and one carbonyl group from 2,3,7,8-TCDD. If
masses m/z 257 and m/z 259 give a chlorine
isotope ratio that agrees  to within ±10% of
the same cluster in the calibration standards.
then the presence of TCDD can be confirmed.
Co-eluting ODD, DDE, and PCB residues can
be confirmed, but will require another
injection using the appropriate SIM masses or
full repetitive mass scans. If the response for
"CU 2.3,7,8-TCDD at m/z 328 is too large,
PCB contamination is suspected  and can be
confirmed by examining the response  at both
m/z 326 and m/z 328. The "CU 2,3.7,8-TCDD
internal standard gives negligible response at
m/z 326. These pesticide residues can  be
removed using the alumina column cleanup
procedure.
  12.9  If broad background interference
restricts the sensitivity of the GC/MS
analysis, the analyst should employ
additional cleanup procedures and reanalyze
by GC/MS.
  12.10  In those circumstances  where these
procedures do not yield a definitive
conclusion, the use of high resolution mass
spectrometry is suggested.5

13. Calculations
  13.1  Calculate the concentration of 2,3,7,8-
TCDD in the sample using the response factor
(RF) determined in Section 7.1.2 and Equation
2.
  Equation 2:
    Concentration (/ig/L) =
                            (AS)(IS)
                         (Als)(RF)(V0)


where:
  A, = SIM response for 2,3.7.8-TCDD at m/z
    320.
  Ai»=SIM response for the internal
    standard at m/z 328 or 332.
  I, = Amount of internal standard added to
    each extract (u.g).
  V0=Volume of water extracted (L).
  13.2  For each sample, calculate the
percent recovery of the internal standard by
comparing the area of the m/z peak
measured in the sample to the area of the
same peak in the calibration standard. If the
recovery  is below 50%. the analyst should
review all aspects of his analytical technique.
  13.3  Report results in jig/L without
correction for recovery data. All QC data
                                                             701

-------
              Federal  Register  / Vol. 49.  No.  209 /  Friday. October  26, 1984  / Rules  and Regulations         141
 obtained should be reported with the sample
 results.

 14. Method Performance
   14.1  The method detection limit (MDL) is
 defined as the minimum concentration of a
 substance that can be measured and reported
 with 99% confidence that the value is above
 zero.' The MDL concentration listed in Table
 1 was obtained using reagent water."The
 MDL actually achieved in a given analysis
 will vary depending on instrument sensitivity
 and matrix effects.
   14.2  This method was tested by 11
 laboratories using reagent water, drinking
 water, surface water, and three industrial
 wastewaters spiked at six concentrations
 over the range 0.02 to 0.20 ng/Lls Single
 operator precision, overall precision, and
 method accuracy were found to be directly
 related to the concentration of the parameter
 and essentially independent of the sample
 matrix. Linear equations to describe these
 relationships are presented in Table 3.

 References
   1. 40 CFR Part 136, Appendix B.
   2. "Determination of 2.3.7.8-TCDD in
 Industrial and Municipal Wastewaters,"
 EPA-600/4-S2-028. U.S.  Environmental
 Protection Agency. Environmental Monitoring
 and Support  Laboratory. Cincinnati, Ohio
 45268. June 1982.
  3. Buser, H.R., and Rappe, C. "High
 Resolution Gas Chromatography of the 22
Tetrachlorodibenzo-p-dioxin Isomers."
Analytical Chemistry, 52. 2257 (1980).
  4. ASTM Annual Book of Standards, Part
31. D3694-78. "Standard  Practices for
 Preparation of Sample Containers and for
Preservation of Organic Constituents,"
American Society for Testing and Materials.
Philadelphia.
  5. Harless, R. L.. Oswald. E. O., and
Wilkinson, M. K. "Sample Preparation and
Gas Chromatography/Mass Spectrometry
Determination of 2.3.7,8-Tetrachlorodibenzo-
p-dioxin." Analytical Chemistry. 52. 1239
(1980).
  6. Lamparski, L. L.. and Nestrick. T. J.
"Determination of Tetra-, Hepta-, and
Octachlorodibenzo-p-dioxin Isomers in
Particulate Samples at Parts per Trillion
Levels." Analytical Chemistry. 52. 2045
(1980).
  7. Longhorst, M. L. and Shadoff, L. A.
"Determination of Parts-per-Trillion
Concentrations of Tetra-. Hexa-. and
Octachlorodibenzo-p-dioxins in Human
Milk," Analytical Chemistry. 52. 2037 (1980).
  8. "Carcinogens—Working with
Carcinogens." Department of Health,
Education, and Welfare. Public Health
Service, Center for Disease Control, National
Institute for Occupational Safety and Health,
Publication No. 77-206, August 1977.
  9. "OSHA Safety and Health Standards.
General Industry." (29 CFR 1910),
Occuptional Safety and Health
Administration, OSHA 2206 (Revised.
January 1976).
  10. "Safety in Academic Chemistry
Laboratories," American Chemical Society
Publication, Committee on Chemical Safety.
3rd Edition, 1979.
  11. Provost, L. P.,  and Elder, R. S.,
"Interpretation of Percent Recovery Data."
American Laboratory. 15, 58-63 (1983). (The
value 2.44 used in the equation in Section
8.3.3 is two times  the value 1.22 derived in
this report.)
  12. ASTM Annual Book of Standards. Part
31. D3370-76, "Standard Practices for
Sampling Water." American Society for
Testing and Materials, Philadelphia.
  13. "Methods, 330.4 (Titrimetric, DPD-FAS)
and 330.5 (Spectrophotometric DPD) for
Chlorine, Total Residual." Methods for
Chemical Analysis of Water and Wastes.
EPA-600/4-79-020, U.S. Environmental
Protection Agency, Environmental Monitoring
and Support Laboratory. Cincinnati. Ohio
45268. March 1979.
  14. Wong. A.S. et al. 'The Determination of
2.3.7.8-TCDD in Industrial and Municipal
Wastewaters. Method 613, Part 1—
Development and Detection Limits." G.
Choudhay. L. Keith, and C. Ruppe. ed..
Butterworth Inc., (1983).
  15. "EPA Method Validation Study 23.
Method 613 (2.3,7.8-Tetrachlorodibenzo-p-
dioxin)," Report for EPA Contract 68-03-2863
(In  preparation).

  TABLE 1.—Chromatographic Conditions and
           Method Detection Limit
Parameter
2.3.7.8.-TC
CO 	 _

Retention
time
(min)
13.1
Method
detection
limit (tig/
U
0.002
  Column conditions: SP-2330 coated on a 60 m long x
0.25 mm  ID glass cotumn with hydrogen carrier gas at 40
cm/sec linear velocity, splitless injection using tetradecane.
Column temperature held isothermal at 200*C tor 1 min, then
programmed at 8'C/min to 250 "C and held. Use of helium
carrier gas will approximately double  the retention  time.
 TABLE 2.—QC Acceptance Criteria—Method
                    613
Parameter
2,3.7.8-TCOD 	
Test
cone.
°3'
o.too
Limit
tors
"L?'
0.0276
Range for X
ig/L (Section 8.2.4).
  X= Average recovery for four recovery measurements, in
ng/L (Section 8.2.4).
  P, P1=Percent recovery measured (Section 8.3.2. Section
8.4.2).
  Note.—These criteria are based directly upon the method
performance data in Table 3. Where necessary, trie limits for
recovery have been broadened to assure applicability of the
limits to concentrations below those used to develop Table
                       TABLE. 3.—METHOD ACCURACY AND PRECISION AS FUNCTIONS OF CONCENTRATION—METHOD 613
Parameter
237 8-TCOD 	

Accuracy, as
recovery. X '
Oig/U
0 86C+0 00145

Single analyst,
precision, s,'
(M/L)
0 13)? -f-000129

Overall
precision. S '
(ji/g/U


   X' = Expected recovery for one or more measurements, of a sample containing a concentration of C. in ug/L
   V = Expected single analyst standard deviation of measurements at an average concentration found of X, in jig/L
   S = Expected interlaboratory standard deviation of measurements at an average concentration found of X. in pg/L
   C = True value for the concentration, in ug/L,
   X = Average recovery found for measurements of samples containing a concentration of C. in pg/L.
                                                             702

-------
                                                      By Lloyd P Provost and Robert S. Elder
 fiterpretacion   of   oercent
 Bff*          J                                     t
                                   L  ...
     'COvery  ciata
  fit SPIKED  (FORTIFIED) sample studies, known
  famounts of a compound or compounds of inter-
  fest are added to aliquots of a sample, and the
     ntage of analyte recovered by a test method is
     to evaluate the performance of that method.
    Environmental Protection Agency (EPA), for
    aple, uses spiking studies in method develop-
     (e.g., Ref. 1) and has proposed the use of
   ked samples  in  quality control programs under
   itional Pollutant Discharge Elimination System
   t>DES) permits.2 Thus, the proper conduct and
    pretation of spiking programs are critical to the
Ifevelopment and implementation of the analytical
.methods upon which important environmental pro-
-ams are based.
> Spiking is particularly useful in wastewater anal-
 yses because the variety of sample matrices and the
 number of analytes of interest in each sample make
 realistic standard  reference materials difficult to
 produce. Spiking permits flexibility in the choice of
 sample matrix and  in the combinations and levels of
.analytes  that  can be evaluated. The  usefulness of
 spiked-sample analyses is not limited to wastewater
.or environmental samples, however, and  proper in-
 terpretation of data from such analyses (percent re-
 covery data) is important whatever the application.
  This paper  will  describe statistical  properties of
 percent recovery data when analytical bias and pre-
cision are proportional  to sample concentration.
The impact of the presence of the analyte of interest
in the unspiked sample (i.e., nonzero background
concentration) will be examined, and some  of the
potential pitfalls in the interpretation of percent re-
covery data in  method development and quality
control applications will be discussed.


Assumptions

  In investigating  the statistical  properties of per-
cent recovery data, we assume that the expected val-
Mr. Provost is Quality Assurance Director. Raitiun Corporation.
Mr. E/rfer is Senior Statistician. JRB Associates.
ue of a  concentration  measurement (X) for a
sample with true concentration B is
                 E (X) = pB
                               (1)
where 100 p is the  mean percent recovery of the
method. If p =  1, the method is unbiased; other-
wise, its absolute bias is proportional to true con-
centration. We also assume that the variance of a
sample with concentration B is
              Var(X) =
                               (2)
where 100 C is the coefficient of variation of the
method. That is, analytical precision is proportion-
al to concentration; the smaller C, the more precise
the method (on an absolute basis).
  It is important to keep in mind that p and C, the
parameters that characterize the bias and precision
of the analytical method, are assumed constant with
respect to concentration. This is a realistic assump-
tion for many methods within their ranges of ap-
plicability. However, the values of p and C some-
times depend on the sample matrix  involved, and
the value of C sometimes increases at low concen-
trations. These possible departures from the simple
properties assumed above often are investigated in
method development studies through  statistical
analyses of percent recovery data (Ref. 1,  for exam-
ple).
  To estimate the mean and variance of percent re-
covery  for a test method at a particular concentra-
tion, one typically analyzes n aliquots of a sample
spiked at that level and computes
and
5- = (n - !)-'!(/>, -
                                         (3)
                                         (4)
(e.g., Ref. 2), where P, denotes the observed percent
recovery for the /th aliquot. The statistical proper-
                                                          AMERICAN LABORATORY   :  57

-------
 ties of P" and s* are described separately below for
 the cases of zero and nonzero background concen-
 tration.


 Zero background

   'if the sample background concentration is known
 to be zero (B = 0), the percent recovery is defined
 as
                  P = 100 Y/T
                            (5)
 where T is the spike level and X is the measured con-
 centration for a spiked aliquot. The most common
 case in which background concentration is known
 to be zero is when  spikes are added  to distilled
 water. It can be shown using the assumed properties
 of the analytical method that in this case the mean
 and variance of 7*" [defined in Eq. (3)] are
                                   percent recovery for the method.
                                   Nonzero background

                                     If the sample background concentration is non-
                                   zero (B>0), percent recovery may be defined as
                                                 P = 100 (Y -
where T is the spike level, Y is the measured concen-
tration of a spiked aliquot, and ~X is an estimate of
the background concentration based on the mean of
measurements  on m unspiked aliquots [e.g., Eq
(3)].  If n aliquots of the same sample are spiked at
level 7" and analyzed, ~F and 5: can be computed as
described  in Eqs. (3) and (4) and used to estimate
the mean  and variance of percent recovery for the
analytical method. It can be shown in this case that
the mean and variance of 7r are                $
 and
                        -  lOOp
              VarC?5") = (lOOpC)2/n
                            (6)
                            (7)
                                   and
 Thus the sample average percent recovery (7s") is an
 unbiased estimator of the mean percent recovery of
 the method. It also can be shown that the  mean of s2
 [defined in Eq. (4)] is

                E(s>) = (lOOpCY            (8)

 that is, s' is an unbiased estimator of the variance of
      QOOpCy-
          n
                £(7r) = \00p
                   Var(F) =

                                                                         \
                                   where k = T/B (k may be termed the spike/bad^
                                   ground ratio). These results show that 7s" also is a*
                                   unbiased estimator of mean percent recovery in
                                   case, but  that the  variance of 7s"  is  greater in
                                   zero-background case  by  a  factor  that depend**
                                                                            •"'$
                                                                             *i?j
                                  ,	  ''"••'*
 Table 1
Impact of spike-to-background rates on variability of percent recoveries
                                          ;'.*;

                                          M
:> Spike-to-background
ratio (k)
Zero background
100
50
10
•>" ' ., 5
'•'"" - ' "1
!-!fT-'- ' 0.5
"•'" "->• -• ->. -.0.1

-------
 RECOVERY DATA continued
 the numbers of spiked and unspiked aliquots ana-
 lyzed (n and m) and the ratio of spike and back-
 ground  concentrations (k).  As the spike/back-
 ground ratio decreases, it can be seen from Eq. (11)
 that Var(7r) increases, and T5" becomes a poorer es-
 timator of analytical percent recovery.
   The consequence of this result is easily illustrated
 by some examples. Table 1 shows the impact of k
 on Var(75") and the expected range in recoveries for
 three cases with only one spiked and one unspiked
 aliquot analyzed (m = n =  1). The expected range
 in recovery is based on a 95% tolerance interval for
 a normal distribution:
             [100 p ± 1.96VVar(FT]

As can be seen from Table 1, when k = 1,
is five times the zero-background value; when k  =
0.1, Va^T5") is about 221 times the zero-background
value.

   It can also be shown that the mean of s2 in the
nonzero-background case is
           E(s~) = (100 pC)1 (1 +  \/kY
(12)
 This is greater than the result for B  = 0 by a factor
 that depends once again on k. For example, when
 spike and background levels are equal (k = l),E(s2)
 in Eq. (12) is four times the zero-background value;
 when k  =  1/5,  £ (s1) is 36  times the zero-back-
 ground value. Thus, s2 is a biased estimator of the
 variance of percent recovery, and s: overestimates
 that variance to a greater extent the smaller the
 spike/background ratio.
Alternate definitions for percent recovery

   The definitions above are not the only ones used
for  percent  recovery.  One  alternative  definition
(based on expressing recovery as a percentage of the
total spiked sample concentration) is

                  />=  100X
                       T+X

Another alternative (applicable when the spike level
is a  multiple, h, of the estimated background con-
centration) is

               p =   IOOQ-- T)


Regardless of hov\  percent  recovery is  defined, it


60    :   DECEMBER 1983
can be shown that percent recovery data tend to be
unreliable  when  the spike/background ratio is
small.
                               .'* •- '.. *
                                   '*;
Interpreting percent recovery data

  Two issuei:were investigated in the method eval-
uation studies for all of EPA's 600 series methods:
  1.  Does  met hod performance depend on  the
sample matrix involved; e.g., do p and ~C values dif-
fer for distilled, natural water, and waisitewater sam-
ples?
  2.  Does performance depend on the "sample con-
centration; e.g., is C larger at lower concentrations?
  These  questions were investigated'by analyzing
spiked aliquots of both distilled and natural water
samples and by spiking aliquots of given samples at
different levels (e.g., Ref. 1).
  We have shown that  for samples with nonzero
background, such as wastewater samples, VarfP'jis
large when k is small. Thus the estimate of p fora
wastewater sample with fl>0 and k small may differ
greatly from the true mean percent recovery of the
method and, therefore, may appear to differ from;
the corresponding  estimate for a  distilled water.
sample. That is,  the different  statistical properties]
of percent recovery data in the zero- and nonzero-"
background cases may mislead one to conclude that}
matrix differences affect  mean  analytical percent
recovery.                       '            ..•*'
  We also  have shown that for samples with noil-
zero background, s: tends to overestimate the van-
ance of  percent  recovery to a greater  extent  the
smaller the spike/background ratio.  Thus, if s' vat.
ues from different spike levels are compared whes
fi>0, it is likely to appear that relative precision ».
poor at lower concentrations even when it is not.   -
  The  discussion  above shows two misconception*.
that can  arise in method development due to
terprctation of percent recovery data. The
cal properties of such data may also lead to ""
rection in analytical quality control  pr
example. EPA's Handbook for Analytical
Control  in  Water  and  U'astewater
suggests  that when analytical precision  varies
concentration, separate control charts
kept  for different concentration ranges.' I/
tionship  of precision to concentration  is
gated by estimating variances  of percent.
data at different spike levels, one may eirc
conclude (if 5>0) that separate charts are
for low concentrations. This would-increase   ^
of proems control activities unnecessarily
in ihe  QA appendix to the proposed^
                                           705

-------
M.., ''.}..' RECOVERY DATA continued
•ri.- "'•: . methods,2 EPA recommends spiking at levels equal
        to 1, 9, and 99 times'background, computing T5" and
        s2 at each level, and comparing these estimates for
        different levels. Results at the lowest spike level
        may well appear to be more variable than results at
        higher levels because of the dependence of E (s2) on
  .'•>.;•• .the spike/background ratio.
      •    An example of the importance of  knowing  the
   " "  statistical properties of percent recovery data is il-
 .   .    lustrated in a report by the Chemical Manufacturers
-I'-; .">" Association (CMA) on results of a joint CMA/EPA
        study .of the quality of wastewater from five organic
..>;-.'  .•;- chemicals plants." One objective of the study was to
•v.---.   ,v characterize the mean  and standard deviations of
"-''•''   1, percent  recovery for the analytical methods used to
     v  "measure organic priority pollutants in the industry's
    '    wastewater. The report  generally was critical of the
        capabilities of the analytical methods  used. How-
     p••; ever, in planning the study it  was decided  that the
 -  .  , spike level should  approximately equal the back-
        ground  concentration; thus, overestimation of per-
        cent recovery  variability was  built  into the study.
        Furthermore,  spike/background ratios  less than
        one sometimes were employed.  The consequences
        were reflected  most dramatically in results for acryl-
        onitrile: the 12 influent percent recovery results for
        this compound ranged from - 7000% to 400% and
        had mean and  standard deviation of -465 and
                                                            2060, respectively, due to the two extreme result
                                                            The estimated spike/background ratios for the ab-\
                                                            quots  with  -7000% and 400%  recovery  wotj-
                                                            10,000/890,000  =  0.011 and 10,000/210,000  "4i
                                                            0.048, respectively. By Eq. (11), it can be seen that'
                                                            the standard deviation of 7s" is about 14,200 pC fat,
                                                            k = 0.01 and m  = n = 1; thus, a result of - 7000%
                                                            is not surprising when k is this small. Proper consid*'
                                                            eration of the statistical properties of percent reco*^
                                                           :ery data would  have led to the choice of a higbff,
                                                            spike/background ratio  in planning the study
                                                            to the exclusion from summary statistics  of
                                                            values made meaningless by the use of too
                                                            spike/background ratios.
                                                              Another example of potential misinterpretati*
                                                            of data from a spiking study can be found in
                                                            As described, spiking studies were used to
                                                            performance of laboratories. The authors
                                                            eluded that overall performance by the five 1
                                                            tories  in the study was  poor.  In one test,
                                                            known freshwater  sample was analyzed
                                                            without spikes of various minerals. The
                                                            spike/background ratios for the six minerals^
                                                            as follows: 0.14, 1.1,0.21, 5.0, 0.71, and 5.0--
                                                            of the variability  in  recoveries attributed
                                                            laboratory performance may have been du.
                                                            statistical   properties  of  recoveries  WB»J
                                                            spike/background ratios.
       62
                 DECEMBER 1983

-------
V  It should be noted that these examples of prob-
 lems in the interpretation of percent recqy.ery data
 from the area of wastewater analysis were selected
 because  this is the application with which the au-
 thors are most familiar. There is no reason to doubt
 that similar examples could be found in other areas
 such as clinical and agricultural chemistry..
of a method near the detection limit), it may be difr
ficult to obtain low background levels in the .sample.
matrix  of interest.  If it then becomes necessary to
perform  spiking' studies. with  a  low spjkWback-
ground ratio, the statistical properties of the recov-
eries should be considered in interpreting the'results
and in  comparing^them to  results at other concen-
trations or in other matrices.         ,'.   '.'!','•
 Summary      .  • .                     '.'..;

   The statistical properties of percent recovery data
 are important to consider when interpreting results
 of analytical studies. When background quantities
 of the spiked artalyte are present, percent ^recoveries
 can be highly' variable a'hid estimates  of analytical
 precision can be b iased.
   In  designing method  evaluation studies, spike
 kvels should first be chosen to cover the  range of
. concentration of interest. Once these levels  are de-
 termined,  then  sample  matrices with  background
 kvels that are small compared to these spiking lev-
 els should be chosen for the study.  If this is done,
 ^en the statistical properties  of percent, recovery
 data will not affect the evaluation of method bias
 and precision.
   In some situations (e.g., studying the properties
References    •:!•:"•"'•-'   •  '.  'r, •..:'1>"

1. WINTER, J.P.. BRjTJON; P.. CLEMENTS, H.,.and KRONER.
   R., "EPA Method,Study 8, Total Mercury in Water," U.S.
   Environmental Protection Agency, Environmental Monitor-
   ing and Support Laboratory, Cincinnati (1977).  "
2. Environmental Protection Agency, "Guidelines'"iEstabJishing
   Test Procedures for.tJijiAnalysis of Po{\iirai\ts'i''j]federat Reg-
   ister 44 (223), 694^MJ9J75,(December 3, 1979).•\'l, " '
3. Environmental Monitoring and,Support.Labpratory,, Hand-
   book/or Analytical Quality Control in WaKr..qnd Waste-
   water Laboratories, EPA-600/4-79^9, 'Office of Research.
   and Developrnent;:u;S. -t'^A, Cincihnati', M"a'r'cn, J979.
4. Engineering-'SL-ience, Inc.,^MA/EPA Five-Planf Study, pre-
   pared for Chemical Manufacturers Association, 3-20-3-22
   VD-3. Austin, TX, April, 1982.        '    's'l;
5. EDWARDS, R.R., SCHILLING, D.L., and ROS'MILLER. T.L.,
   "A performance evaluation of certified water analysis labora-'
   tories,"y. Wat. Poll. Contr. Fed. 49, 1704(1977),   :
                                                                     AMERICAN LABORATORY  :  63
                                                      '0.

-------