jT/Jf
             £*t>
             Il10'%
(V
METHOD 513.  DETERMINATION OF 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN
             IN DRINKING WATER BY GAS CHROMATOGRAPHY  WITH HIGH
                  RESOLUTION MASS SPECTROMETRY
                                     July 1990
               This method Is taken from the SW-846 Methods Manual,
                    Method 8280,  and adapted to drinking water.
                                 A.  Alford-Stevens

                               James W.  Eichelberger
                    ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
                        OFFICE OF RESEARCH AND DEVELOPMENT
                       U.S. ENVIRONMENTAL PROTECTION AGENCY
                              CINCINNATI, OHIO 45268

                                HEADQUARTERS LIBRARY
                                ENVIRONMENTAL PROTECTION AGENCY
                                WASHINGTON, D.C. 20460
    o
    oo
                             33

                      HEADQu.., '   .
                      ENVIRONMLM*. ;
                      WASHINGTON, D.C. i
                                                 •LCTION AGENCY
                                                 +00

-------
                                  METHOD 513

DETERMINATION OF 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN  IN  DRINKING WATER BY
        GAS CHROMATOGRAPHY WITH HIGH-RESOLUTION MASS SPECTROMETRY
1.    SCOPE AND APPLICATION       §         1

      1.1   This method provides procedures for identification and measurement
            of 2,3,7,8-tetrachlorodibenzorp-dioxin  (TCDD,  CASRN 1746-01-6)  at
            concentrations of 20 pg/L to 2 ng/L in water sample extracts.  The
            minimum measurable concentration will  vary among samples, depending
            on the presence or absence of interfering compounds  in a particular
            sample.

      1.2   A  water  sample  may  contain  floating,  suspended,  and   settled
            particulate matter, which should not be  removed by filtering before
            extraction.  The  estimated solubility of 2,3,7,8-TCDD in water  is
            <50 ng/L  (1),  but larger measured concentrations can be caused  by
            TCDD associated with particulates.

      1.3   Because  2,3,7,8-TCDD  may  be  extremely toxic,  safety  procedures
            described  in  Section 5 should be  followed to prevent exposure  of
            laboratory personnel to materials  containing this compound.

2.    SUMMARY OF METHOD

      2.1
             ^ iJM I I WJH» %f\r VVIJip WMIIV* )  s?Vy  VLMV*   «*•«* fe y W ) / y V/  llsLJU  ^lllt«t^lll%Ai  ^OUltNjlUlt*}
             IS), are  added to the water.   The sample container is rinsed  with
             methylene chloride,  which is  then added to the water sample.   The
             water sample is extracted sequentially with three 60-mL portions of
             methylene chloride.   AN optional liquid-solid extraction  procedure
             using Empore disk technology is also  described in this method.  When
             using this  option,  all surrogate compounds and  internal  standards
             and other solutions are added just as in the liquid-liquid extraction
             procedure.The combined extract is subjected to column chromatographic
             procedures  to remove sample components  that  may interfere  with
             detection and measurement of  TCDD.   A 10-juL aliquot  of a solution
             containing 13Ci;z-l,2,3,4-TCDD, which is  used as  a recovery standard
             (RS),  is  addecf to the extract  before concentration  and  analysis.
             The  sample  extract is concentrated to  10  /iL,  and a  l-p,l or  2-/*L
             aliquot is  injected  into a gas chromatograph (GC) equipped  with a
             fused silica capillary column and interfaced with a high resolution
             mass spectrometer (MS).  Selected characteristic ions  are  monitored
             with high resolution MS  (10,000 resolving power).  Identification
             of  a  sample component r'as  TCDD  is  based  on detection, of  two
             characteristic;'lions'  (m/z 320 and 322) in the molecular ion cluster,
             measurement of acceptable relative abundance of those  two  ions, and
                                       34

-------
            relative to the IS,  "C^.S^.S-TCDD.  Because the IS is a labeled
            analog of the analyte,  the procedure presumes that IS losses during
            method procedures are  equal  to  unlabeled  TCDD  losses.   Therefore,
            each calculated sample TCDD concentration has been compensated for
            losses during sample preparation.

3.    DEFINITIONS

      3.1   Calibration limits  —  the minimum (20 pg/L) and  maximum (2 ng/L)
            concentration of 2,3,7,8-TCDD in solutions used to calibrate detector
            response.  In  some samples, <20 pg/L of 2,3,7,8-TCDD may be detected
            but measured concentrations will only be estimated concentrations.
            In  other samples,  interferences may  prevent  identification  and
            measurement of 20 pg/L.

      3.2   Concentration calibration  solution —  a  solution  containing known
            amounts  of the  analyte  (unlabeled  2,3,7,8-TCDD),  the  IS  (13C  "
            2,3,7,8-TCDD), the SC (37Cl4-2,3,7,8-TCDD), and the  RS (13C12-l,2,3,
-------
      for the same Ion.  The same number of scans must be integrated for
      both areas.   (The ratio of  peak heights may be used instead of peak
      areas.)

3.10  Surrogate compound (SC) — a compound  (37Cl4-2,3,7,8-TCDD)  that is
      present  in   each calibration  solution  and  is  added  at  a  low
      concentration (20 pg/L) to  each sample and blank before extraction.
      Successful  detection  and  measurement  of  the SC  in each  sample
      provides  some  assurance   that  unlabeled  2,3,7,8-TCDD  would  be
      detectable if present in  the sample at >20 pg/L.

INTERFERENCES

4.1   An organic  compound  that  has approximately the  same GC retention
      time 2,3,7,8-TCDD (within  a  few scans  of the  IS) and produces the
      ions  that are  monitored   to  detect 2,3,7,8-TCDD  is a  potential
      interference.  Most frequently encountered interferences are other
      sample components that are  extracted along with TCDD;  some potential
      interferences are listed in Table 1.  To minimize  interference, high
      purity reagents and solvents must  be used and  all equipment must be
      meticulously cleaned.  Laboratory  reagent blanks must be analyzed
      to  demonstrate lack  of  contamination that  would  interfere  with
      2,3,7,8-TCDD measurement.   Column  chromatographic  procedures are
      used  to  remove some  sample  components;  these procedures  must be
      performed carefully to minimize loss of 2,3,7,8-TCDD during attempts
      to enrich its concentration relative to other sample components..

4.2   False positive identifications are produced only when  an interfering
      compound  elutes from the  GC column within  3  sec  of the  IS and
      produces ions with exact masses and relative abundances very similar
      to  those for  2,3,7,8-TCDD.    The  specified  GC  column  separates
      2,3,7,8-TCDD from all 21 other TCDD isomers.

SAFETY

5.1   Because 2,3,7,8-TCDD has been identified as an  animal  carcinogen and
      a  possible  human carcinogen,  exposure  to  this compound  and its
      isotopically  labeled  analogs  must  be  minimized   (2,3).    The
      laboratory  is  responsible  for maintaining a  file  of current OSHA
      regulations  regarding the  safe  handling  of chemicals specified in
      this  method.   A  reference file  of material  data  handling sheets
      should also be made available to all personnel  involved in analyses.

5.2   Each  laboratory  must  develop a  strict  safety  program for handling
      2,3,7,8-TCDD.  The following laboratory practices are recommended:

      5.2.1   Minimize   laboratory  contamination  by  conducting   all
              manipulations in a hood.

      5.2.2   Effluents  of GC  sample  splitters and GC/MS vacuum pumps
              should  pass through  either  a column of activated carbon or
                                36

-------
              be bubbled  through  a trap containing oil  or high-boiling
              alcohols.

      5.2.3   Liquid waste should  be dissolved in methanol or ethanol and
              irradiated with ultraviolet light at a wavelength  <290 nm
              for several  days.  Analyze the liquid wastes and dispose of
              the solutions when 2,3,7,8-TCDD can  no  longer be detected.

5.3   The following precautions for safe handling  of  2,3,7,8-TCDD in the
      laboratory are presented  as  guidelines only.  These precautions are
      necessarily   general   in   nature,   because   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 conditions may be obtained from certain
      consulting laboratories or from  state Departments  of Health  or of
      Labor, many of which  have an industrial health service.   Although
      2,3,7,8-TCDD  is  extremely  toxic to certain  kinds of  laboratory
      animals, it has been handled  for  years without injury in analytical
      and biological laboratories.  Techniques used to handle radioactive
      and infectious materials  are applicable to 2,3,7,8-TCDD.

      5.3.1   Protective  equipment:     Laboratory  hood  adequate  for
              radioactive work, safety glasses, and disposable  plastic
              gloves, apron or  lab coat.

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

      5.3.3   Person hygiene:   Thorough washing  of  hands  and  forearms
              after each manipulation  and before  breaks  (coffee,  lunch,
              and shift).

      5.3.4   Confinement:   Isolated  work  area,  posted  with  signs,
              segregated glassware and tools,   plastic-backed  absorbent
              paper on benchtops.

      5.3.5   Waste:   Good  technique  includes  minimizing  contaminated
              waste.  Plastic liners  should  be used in  waste cans.

      5.3.6   Disposal  of Wastes:   Refer to the November 7,  1986,  issue
              of the  Federal Register on Land  Ban  Rulings for  details
              concerning handling  wastes  containing dioxin.

      5.3.7   Decontamination:   Personnel  -  any  mild  soap with  plenty of
              scrubbing action.  Glassware,  tools, and surfaces  -  rinse
              with 1,1,1-trichloroethane, then wash with any detergent and
              water.   Dish  water  may be disposed  to  the sewer  after
              percolation  through  a carbon filter.  Solvent wastes should
              be minimized,  because they require special disposal  through
              commercial  sources that  are  expensive.
                                37

-------
5.3.8   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.

5.3.9   Wipe tests:   A useful method to  determine cleanliness  of
        work  surfaces and  tools  is  to  wipe a  surface  area  of
        2 in. X 1  ft.  with  an acetone-saturated  laboratory wiper
        held in a  pair of clean  stainless steel  forceps.   Combine
        wipers  to  make one  composite sample  in  an extraction jar
        containing 200 mL acetone. Place  an equal  number of unused
        wipers in 200 mL acetone and use as a control.   Extract each
        jar with a wrist-action shaker for 20 min.   Transfer extract
        to   a   Kuderna-Danish  (K-D)  apparatus   fitted   with   a
        concentrator tube and a three-ball Snyder column.  Add two
        boiling chips  and  concentrate the extract to an apparent
        volume  of  1.0  mL with the same  techniques used for sample
        extracts.  Add 100  jitL of  the sample fortification solution
        that has  not been  diluted with  acetone  or 1.5 mL of the
        acetone-diluted solution  (Section 7.14),  and continue all
        extract   preparation   steps  and  analytical   procedures
        described  for  samples.    If  any  2,3,7,8-TCDD  is detected,
        report  the result as  a quantity (picograms) per wipe test.
        A lower limit  of  calibration of 20 pg/composite wipe test
        is expected.  A positive  response  for  the  control sample is
        8 pg/wipe test. When  the  sample contains  >25 pg, steps must
        be  taken  to correct the  contamination.   First vacuum the
        working places (hoods, benches, sink) using a vacuum cleaner
        equipped with a high-efficiency particulate absorbent filter
        and then wash with a detergent.  Analyze a new set of wipers
        before  personnel  are  allowed in  work in  the previously
        contaminated area.

5.3.10  Inhalation:    Any  procedure that may   produce   airborne
        contamination  must  be carried out with  good  ventilation.
        Gross  losses  to  a  ventilation  system  must not be  allowed.
        Handling  the  dilute solutions  normally used in analytical
        and  animal work presents  no  significant  inhalation hazards
        except  in  case of an  accident.

5.3.11  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.
                           38

-------
,6.    APPARATUS AND  EQUIPMENT

      6.1    Computerized  GC/MS System

             6.1.1    The  GC must be capable  of temperature programming and be
                     equipped with  all  required accessories, such as syringes,
                     gases,  and  capillary columns.   The GC  Injection port must
                     be designed  for capillary  columns.  Splltless or on-column
                     Injection technique Is recommended.   With  this method,  a
                     Z-pL  Injection is  used consistently.   A  1-jttL injection
                     volume  can   be  used, but  the  injection volume should be
                     constant throughout analyses of calibration solutions  and
                     related blanks,  sample   extracts,  and  quality  control
                     samples.

             6.1.2    GC/MS interface components should withstand  temperatures up
                     to  280°C.    The  interface  should  be  designed  so  that
                     separation  of  2,3,7,8-TCDD  from  all   other  TCDD isomers
                     achieved in the GC column is not appreciably degraded.  Cold
                     spots or active surfaces  (adsorption sites) in the interface
                     can cause peak tailing and  broadening.  The GC column should
                     be  inserted  directly into  the MS ion source without being
                     exposed to  the ionizing  electron beam.  Graphite  ferrules
                     should  be  avoided  in the  injection  port because they  may
                     adsorb  TCDD. Vespel or equivalent ferrules are recommended.

             6.1.3    The static resolving power of the MS must be maintained at
                     >10,000 (10% valley).  The  MS must be operated in a  selected
 ;                    ion monitoring (SIM) mode,  and data must be acquired for  the
                     ions  listed  in Table 2 during a  total cycle  time (including
                     instrument overhead time)  of <1  s.  Selection of the lock-
                     mass  ion is left to the performing laboratory.   Recommended
                     MS tuning conditions are provided in Section 9.1.  The  ADC
                     zero  setting must  allow peak-to-peak measurement of  baseline
                     .noise for every monitored channel and allow  good estimation
 ;                    of instrument  resolving  power.

             6.1.4    A dedicated  data  system  is used to control  rapid  SIM data
                     collection.  Quantitation data (peak areas or peak  heights)
                     must  be acquired continuously and stored.   The  data system
                     must  be capable of  producing selected  ion current  profiles
                     (SICPs, which are displays of ion intensities as a  function
                     of  time)  for each  monitored  ion,  including the lock-mass
                     ion.   Quantitation  may be  based  on computer-generated peak
                     areas or on measured peak heights.  The data system must be
                     capable of  acquiring data for > five  ions and generating
                     hard  copies  of  SICPs  for  selected   GC.  retention  time
                     intervals and  permit measurement of baseline noise.

      6.2    GC Column. ' Two narrow bore,  fused silica capillary columns coated
             with phenyl cyanppropyl si 11 cone are recommended; one is a  60-m  SP-
                                      39

-------
      2330 and the other is a 50-m CP-SIL 88.   Any capillary column that
      separates 2,3,7,8-T€DD from all other TCDO isomers may be used, but
      this separation must  be demonstrated.  At the beginning of each 12-h
      period during which analyses are to be performed, column operating
      conditions must be demonstrated to achieve 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 3.

6.3   Miscellaneous Equipment.

      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   Glove box.

      6.3.6   Drying oven.

      6.3.7   Minivials  —  1-mL amber borosilicate  glass  with
              conical-shaped  reservoir  and  screw  caps  lined with
              Teflon-faced silicone disks.

      6.3.8   Pipets, disposable, Pasteur, 150 mm X 5 mm i.d.

      6.3.9   Separatory funnels, 2 L with Teflon stopcock.

      6.3.10  Kuderna-Danish  concentrator,  500  ml,  fitted  with  10-mL
              concentrator tube and three-ball Snyder column.

      6.3.11  Teflon boiling chips washed with hexane before use.

      6.3.12  Chromatography column, glass,  300 m X 10.5 mm i.d., fitted
              with Teflon stopcock.

      6.3.13  Adapters for concentrator tubes.
                                                         ^
      6.3.14  Continuous liquid-liquid extractor  (optional).

      6.3.15  Glass funnels, appropriate size  to accommodate filter paper
              used to filter extract (volume of approximately 170 mL).

      6.3.16  Desiccator.

6.4   CAUTION:   To  avoid the risk of  using contaminated glassware, all
      glassware  that  is  reused  must be meticulously  cleaned  as  soon as

                                40

-------
7.
      •possible after use.  Rinse glassware with the last solvent used In
      It.  Wash with hot Water containing detergent.  Rinse with copious
      amounts of tap water and several portions of distilled.  Drain dry
      and  heat  in  a muffle furnace  at 400°C  for  15-30  win.   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.

6.5   TCDD concentrations of concern in  water are much lower than those
      of concern in many other sample types.  Extreme care must be taken
      to prevent cross-contamination between water and other samples.  The
      use of separate glassware for  water samples is recommended.

6.6   Empore extraction disks, C-18, 47mm.

6.7   Mi 111 pore Standard Filter Apparatus (or equivalent) to  hold disks,
      all glass

REAGENTS AND CONSUMABLE MATERIALS

7.1   Alumina, acidic  (BioRad Lab. #132-1240  or equivalent).  Extract in
,  .,   a Soxhlet apparatus with methylene chloride for 6 h (>  3 cycles/h)
      and  activate it  by heating  in a foil-covered glass  container for
      24 h at 190°C.

7.2   Carbon, (Amoco PX-21 or equivalent).

7.3   Glass wool.  Extract with methylene chloride and hexane  and air-dry
      before use.  Store  in a clean  glass jar.
1                  '''•
7.4   Potassium hydroxide, ACS grade.

7;5   Potassium silicate.  Slowly dissolve 56 g of reagent grade potassium
      hydroxide  in 300 mL of anhydrous  methanol  in a  1-L  round bottom
     : flask.  Add  slowly with swirling  100 g silica gel  (prewashed and
      activated).   With a  rotary  evaporation apparatus with no vacuum
      applied, rotate  the  flask  and heat to 55°C  for 90 min.  After the
      mixture cools to  room temperature,  pour  it into a large glass column
      containing a plug of glass wool at the  end.  Wash the mixture into
      the column with methanol, and then  add 200 mL of methanol.  When the
      methanol level reaches the bed of  sorbent, add 200 mL of methylene
      chloride  to  the   column. Push  the  methylene  chloride  through the
      column of sorbent by applying nitrogen pressure to dry or partially
      dry the sorbent,  which is then activated at 130°C overnight.

7.6   Silica gel,  high  purity grade, type 60, 70-230 mesh.  Extract in a
      .Soxhlet apparatus with methylene chloride for  6 h  (>3 cycles/h) and
                                      41

-------
      activate by heating in a foil-covered glass container  for 24 h at
      130°C.

7.7   Silica gel  impregnated with 40% (w/w) 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.

7.8   Silica gel/Carbon.   To a  20-g portion of  silica  gel  add  500 mg
      carbon, and blend until the mixture is a uniform color.

7.9   Sodium sulfate, granular,  anhydrous.

7.10  Solvents, high purity, distilled-in-glass,  or  highest  available
      purity: methylene chloride,  hexane,  benzene,  methanol, tridecane,
      isooctane,  toluene,  cyclohexane, and acetone.

7.11  Sulfuric acid, concentrated, ACS grade, specific gravity 1.84.

7.12  Concentration  Calibration  Solutions (Table 4)  —  Five  (or more)
      tridecane solutions  (CAL 1-5) containing unlabeled 2,3,7,8-TCDD and
      isotopically labeled TCDDs.  All  five  solutions  contain  unlabeled
      2,3,7,8-TCDD at  varying concentrations and the  IS (13C1?-2,3,7,8-
      TCDD, CASRN 80494-19-5)  and the RS  (13C12-1,2,3,4-TCDD)  each at a
      constant concentration.  Three of these solutions also contain the
      surrogate compound  (SC, 37Cl4-2,3,7,8-TCDD,  CASRN 85508-50-5)  at
      varying concentrations.  All standards required for preparing CALs
      are commercially available but must be verified by comparison with
      the National Bureau  of Standards certified  solution SRM-1614, which
      contains 67.8 ng/mL  of unlabeled 2,3,7,8-TCDD and 65.9 ng/mL of 13C-
      labeled  2,3,7,8-TCDD  at 23°C.   Note:   CALs can  be  prepared by
      diluting calibration solutions used in Contract Laboratory Program
      procedures for 2,3,7,8-TCDD determinations with low resolution MS;
      to obtain appropriate  IS concentrations for CAL 4, however, solvent
      containing the IS must be used  for dilution.  Calibration solutions
      used for USEPA Method.8290 can also be used  to  determine RFs for
      2,3,7,8-TCDD;   with  those   solutions   the   lower   calibration
      concentration may be higher (25 pg/L rather than  20 pg/L) or lower,
      depending on   injected  volume  of calibration solution.    Because
      Method 8290 solutions do not contain the SC, however, one or three
      additional solutions containing that compound will  be necessary to
      measure   its   RF   relative   to  the   IS.    Assuming   adequate
      reproducibility  of  RF  measurements, triplicate  analyses  of one
      solution  (recommended SC  concentration  of 1.2  pg/A*L)  or single
      analysis  of three  solutions  (0..6 to  1.8  pg/MU  Table  4)  are
      acceptable.

      7.12.1  Each  of CALs  1-5  contains  the  IS at a concentration of
              50 pg/ML; if  100%  of the IS is  extracted, 10 ML of this
              solution is equivalent  to a 10-jxL extract  of a 1-L sample
              to which 500 pg of IS was  added before extraction.   If 100%
                                42

-------
                    of  the  analyte is  extracted,  CALs 1-5  contain unlabeled
                    2,3,7,8-TCDD at concentrations  that are equivalent to 10-nl
                    extracts of 1-L samples containing 20 pg to 2 ng.

            7.12.2  CALs   1-3   contain   the   SC   (37C1-2,3,7,8-TCDD)   at  a
                    concentration  of 0.6  pg/^L,  1.2  pg/*iL,  and  1.8  pg//iL,
                    respectively;  10  jxL of those  solutions  are equivalent to
                    10 juL extracts containing 30%,  60%, and  90%, respectively,
                    of  the  amount of  SC  added  to  each  1-L  sample  before
                    extraction.

            7.12.3  Store CALS  in  1-mL  amber  minivials at room temperature in
                  , the dark.

      7.13  Column  Performance  Check Solution --  contains a  mixture of TCDDs
            including  the  IS,  the  SC,  unlabeled  2,3,7,8-TCDD,  1,2,3,4-TCDD
            (CASRN  30746-58-8),  1,4,7,8-TCDD(CASRN 40581-94-0), 1,2,3,7,-TCDD
            (CASRN  67028-18-6),  1,2,3,8-TCDD  (CASRN 53555-02-5), 1,2,7,8-TCDD
            (CASRN  34816-53-0),  and  1,2,6,7-TCDD  (CASRN 40581-90-6).   Other
            TCDDs can be present.  Except for  the IS and SC, solution component
            concentrations  are  not critical.   The IS  concentration should be
          •  10 ±  1  pg/ML and the  SC  concentration should be  0.6  ±0.1 pg/L,
            because  ions  produced  by these  compounds  will  be  used  to check
            signal-tornoise ratios.

      7.14  Sample  Fortification Solution.  A solution containing the IS at a
            concentration of 5 to 25 pg/^L and the SC at a concentration of 0.2
.  ,  '        to  1  pg/^L,  but with  the  ratio of IS to SC  always  25:1.   The
     ,     ,  solution solvent is not critical; mix  20 to 100 pi, as  appropriate
            to  produce  needed  IS  and SC concentrations  (50  pg/L  and 2 pg/L,
            respectively)  of this  solution with 1.5  mL  of  acetone.   Add the
            resulting solution to  each  sample and  blank before  extraction.

      7J5  Recovery Standard  Solution.  A tridecane  solution containing the
            recovery standard,  13C12-1,2,3,4-TCDD  at a concentration of 50 pg/ML.
            A lQ-fj.1 aliquot  of  this  solution  is  added  to  each  sample  and blank
            extract before  concentrating the extract  to  its final  volume for
            analysis (Section 11.4.1).

8.    SAMPLE COLLECTION. PRESERVATION. AND STORAGE

      8.1   Samples must be collected in  glass containers. The container should
            not be  rinsed with sample before  collection.

      8.2   Samples  may  be  stored  under   ambient   conditions  as  long  as
            temperature  extremes  (below  freezing  or  >90°F)  are  avoided.   To
            prevent photo-decomposition,  samples must  be protected from light
            from the time of collection  until extraction.

      8.3   All samples must be extracted within 90 days after collection and
            completely analyzed within  40 days after extraction.

                                      43

-------
9.    GC/MS SYSTEM CALIBRATION.

      9.1   MS Performance.
            9.1.1   The MS must be operated In the electron ionization mode, and
                    a static resolving  power  of >10,000 (10% valley definition)
                    at >m/z  334  must be  demonstrated  before any  analysis  is
                    performed.   The resolving  power  must  be  documented  by
                    recording the  mass profile of the  reference  peak.   The
                    format of the peak  profile representation must allow manual
                    determination of resolution (i.e., the horizontal axis must
                    be a calibrated mass scale (amu or ppm per division).  The
                    peak width at 5% peak height must appear on the hard copy
                    and cannot exceed 100 ppm.  Static resolving power must be
                    checked  at the  beginning and  end of each  12-h  period  of
                    operation.    A  visual   check  of  static  resolution  is
                    recommended before  and after each analysis.

            9.1.2   Chromatography time may exceed  the long-term mass stability
                    of the high resolution MS, and mass  drift of a few ppm can
                    affect the accuracy of measured masses.   Therefore, a mass
                    drift correction is required.   A lock  mass ion  from the
                    reference compound  (high  boiling perfluorokerosene, PFK, is
                    recommended)  is  used  to  calibrate the MS.   An acceptable
                    lock mass is  an  ion with  mass larger than the lightest mass
                    monitored but less  than  the  heaviest ion monitored.   The
                    amount  of. PFK   introduced  into  the ion  chamber  during
                    analysis  should  be  adjusted  so that the  amplitude of the
                    lock mass ion is <10% full scale.   Excessive PFK may cause
                    noise problems and  ion source contamination.

            9.1.3   Using a PFK molecular leak, tune the MS to obtain resolving
                    power  of >10,000  (10%  valley)  at  m/z  334.    Using   a
                    reference peak near m/z 320,  verify  that  the exact mass of
                    the reference peak  is within  5 ppm of the known mass.  The
                    low-  and high-mass  reference  ions  must be  selected  to
                    provide the voltage jump  required to detect ions from m/z
                    320  through  m/z  334.    (Note:    With  a  qualitative
                    confirmation option in Section 11.5.5, detected ion range
                    will be m/z 257 to  m/z 334.)

            9.1.4   MS resolving power must  be demonstrated  by recording the
                    mass peak profile of the high-mass reference signal obtained
                    using  the low-mass  ion  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   peak  profile
                    representation  must  allow  manual  determination  of  the.
                    resolution (i.e., the horizontal  axis must be a calibrated
                    mass  scale in amu  or  ppm per division.   The measured peak

                                      44

-------
              width at 5% of the peak maximum must appear on the hard copy
              and cannot exceed 100 ppm at the high mass.

9.2   Initial Calibration

      9.2.1   GC  column performance.   The  laboratory  must verify  GC
              conditions necessary for required separation of 2,3,7,8-TCDD
              from other TCDD  isomers.   Inject 2 ML  of the performance
              check solution and  acquire  SIM data for  the  five  ions  in
              Table 2 (nominal  m/z 320, 322,  328, 332, and 334)  within a
              total cycle time of <1 s.  Acquire at least five scans for
            ;  each  ion  across each  GC  peak  and use the  same  data
              acquisition  time  for  each  ion  monitored.     The  peak
              representing 2,3,7,8-TCDD and peaks representing any other
              TCDD isomers must be resolved with  a valley of <25% (Figure
              1), Valley %  - 100  x/y, where y  is peak height of 2,3,7,8-
              TCDD and x is measured as shown  in Figure 1 between 2,3,7,8-
              TCDD and  its  closest  eluting  isomer.  CAUTION:   The same
              data acquisition  parameters must  be  used to  analyze all
              calibration and performance check solutions.

      9.2.2   MS calibration and sensitivity check.  Ratio of integrated
              ion current for  m/z 320 to m/z  322  produced  by unlabeled
              2.3,7,8-TCDD  and  for  m/z 332  and  334 produced by  the  IS
              ("C-labeled 2,3,7,8-TCDD) must be >0.67  and <0.87.  The S/N
              ratio for m/z 328 produced  by the  SC  (13C-labeled  1,2,3,4-
              TCDD) must be >2.5  and  the  S/N  ratio for  m/z  332  produced
              by the IS must be >10.

.,,    9.2.3   Using the same GC and  MS conditions, analyze a 2-/iL aliquot
              of the medium concentration CAL (CAL 3).  Check ion ratios
              specified in Section  9.2.2.  If criteria  are  met,  analyze
              a 2-/iL,aliquot of each of the four (or more), remaining CALS.

      9.2.4   For each  CAL, ensure  that  ion ratios (Section  9.2.2) are
              acceptable.  For  CAL 1 (the  lowest concentration CAL) data,
              ensure that each  ion produces a signal-to-noise (S/N) ratio
              of >2.5.  Display a SICP  for a region of  the  chromatogram
              near the elution  time of 2,3,7,8-TCDD but  where no analyte
              or interference peak is present.  The preferred width of the
              display is about  10 X full  width at  half  height of the IS
              peak.  The "noise" is the height (measured from the lowest
              point in  the display window)  of  the largest  signal  not
              attributable to any eluting substance.

 •>.    9.2.5   RF  Measurements.   Using  data  acquired  for  each  CAL,
              calculate the RF  for unlabeled  2,3,7,8-TCDD,  the SC (37CL-
              2,3,7,8-TCDD), and  the  RS (13C12-2,3,7,8-TCDD)  relative  to
              the IS (13C12-2,3,7,8-TCDD) with the following  equation:


                        RF-  VQi./"AI.QX

                                45

-------
                  where  Ax=  the sum of integrated ion abundances of m/z
                             320 and 322 for unlabeled 2,3,7,8-TCDD, the
                             abundance  of m/z  328 for  the  SC, or the
                             abundances of m/z  332 and 334 for  the  RS.

                        A,  « the sum of integrated ion abundances of m/z
                             332 and 334 for the  IS,

                        Qis - injected quantity of IS,  and

                         Qx = injected quantity of  unlabeled 2,3,7,8-TCDD,
                             the SC, or the RS.

                  RF  is  a   unitless   number;   units  used  to  express
                  quantities  must  be equivalent.
      9.2.6   For each compound (unlabeled 2,3,7,8-TCDD, the SC, and the
              RS), calculate a mean RF and the relative standard deviation
              (RSD) of  the five measured RFs.    When  RSD  exceeds  20%,
              analyze additional aliquots of appropriate  CALs  to obtain
              an  acceptable RSD of  RFs  over  the entire  concentration
              range, or take action to improve  GC/MS performance.

9.3   Routine Calibration.   If a laboratory operates during only one <12-h
      period  (shift)  each  day, routine  calibration  procedures  must  be
      performed at the  beginning (after  mass calibration  and successful
      analysis  of the  performance check solution  to  ensure  adequate
      sensitivity  and acceptable  ion ratios)   of that  shift,   and  the
      performance check solution must  be analyzed again at the end of that
      shift to validate  data acquired during the  shift.   If the laboratory
      operates during consecutive shifts, routine calibration procedures
      must be performed at  the  beginning  of each  shift,  but analysis  of
      the performance check solution  at  the beginning of  each  shift and
      at the end of the final  12-h  period is sufficient.

      9.3.1   Inject a 2-/iL aliquot of CAL 3, and analyze with  the same
              conditions used during Initial  Calibration.

      9.3.2   Demonstrate  acceptable  performance  for  ions  abundance
              ratios, and demonstrate  that each measured RF for unlabeled
              2,3,7,8-TCDD, the SC,  and  the RS  is within  20% of  the
              appropriate  mean  RF  measured during  initial  calibration.
              If  one  or more  of these criteria  are  not met, up  to  two
              additional attempts can be  made before remedial  action  is
              necessary  and the entire  initial  calibration process  is
              repeated.   Corrective  action may include  increasing  the
              detector sensitivity,  baking the GC column, clipping a short
              length  (about 0.3-0.5 m)  of the  injector side of  the  GC
              column, washing or replacing the GC column, and  cleaning the
                                46

-------
                     ion  source.   If  degradation of the  standards in CALs  Is
          .-..          suspected,  a fresh set of CALs should be. used for repeating
          ••:          initial  calibration  procedures.

 10.   QUALITY CONTROL

      10.1   Laboratory  Reagent Blank.   Perform  all  steps  in the  analytical
             procedure  using  all   reagents,  standards,  equipment,  apparatus,
             glassware,  and solvents that would be used for a  sample  analysis,
             but  omit a water sample,  and substitute 1  L of reagent water.

             10.1.1   Analyze  two laboratory reagent blanks (LRBs)  before  sample
                     analyses begin  and when a new batch of solvents or reagents
                     is used  for sample extraction.  Do not add any IS, SC or RS
                     to  one blank;  this will  allow demonstration  that reagents
                     contain  no impurities  producing  any  ion current  above  the
                     level  of background  noise  for monitored  ions.

             10.1.2   Criteria for acceptable  LRBs.

                     10.1.2.1   When no IS, SC, or RS  is present,  no ion current
.:-...                           above  the level- of background S/N is detected for
!,               .                any  monitored ion within 20 s of  the  retention
                                times  previously measured for labeled 1,2,3,4-
                                TCDD or for unlabeled and  labeled  2,3,7,8-TCDD.

                     10.1.2.2   When the  IS  is present, no ion current'for  m/z
                                259, 320, or  322  is observed that  is  >2%  of  the
                                abundance of  m/z 332 within  5  scans  of  the  IS
                                peak maximum.

        "     10.1.3   Corrective  action for unacceptable  LRB.   Check  solvents,
 .:',,.'.               reagents,  apparatus, and glassware to locate  and  eliminate
   \                the source of contamination before any samples are extracted
                     and analyzed.   Purify or discard contaminated reagents  and
                     solvents.                             .

      10.2   Field  Blanks.    An acceptable field  blank must meet criteria  in
             Section  10.1.2.2.   When results  for a field  blank are acceptable,
             analysis of an LRB is not  needed  with that sample batch.  When field
             blank results  are  not  acceptable,  analysis^of an LRB  is  needed;  if
             LRB  results  are acceptable, data  for samples associated with  the
             field blank must be accompanied  by pertinent  information  about  the
             nature and amount  of contamination observed in the field  blank.

      10.3   Corrective action for unacceptable performance check  solution data.
             When the MS  sensitivity requirement (Section 9.2.2)   is not  met  at
             the end of a 12-h period in which sample extracts  were analyzed,  all
             related  sample extracts must be reanalyzed  after  criteria have been
             met.  When other criteria  (ion ratios or GC resolution) are not met,
; ,.           all  sample  extracts that produced positive  results  or  potential
                                      47

-------
            positive results must be reanalyzed after calibration criteria have
            been met.
11.    PROCEDURE
      11.1  Sample Extraction ~ Liquid-Liquid Extraction

            11.1.1  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.  A continuous liquid-liquid  extractor  may be used
                    instead of a separatory funnel.

            11.1.2  Add 1.5  mL  of the sample fortification  solution (Section
                    7.14) to the sample in the separatory funnel.

            11.1.3  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.  Allow
                    the organic layer to  separate from  the water  phase  for a
                    minimum of 10  min.  If an emulsion  interface between layers
                    exists,  the  analyst  may use  mechanical  techniques  to
                    complete  the  phase  separation.   Collect  the  methylene
                    chloride layer directly into a 500-mL Kuderna Danish (K-D)
                    concentrator  (mounted with  a 10-mL  concentrator tube)  by
                    passing the sample extract through a filter funnel packed
                    with a glass wool plug and 5-g of anhydrous sodium sulfate.
                    Repeat the extraction  with two additional 60-mL portions of
                    methylene chloride, filtering each extract before adding it
                    to the K-D concentrator.   After the third extraction, rinse
                    the sodium  sulfate with an additional 30 mL of methylene
                    chloride to ensure quantitative transfer, and add rinse to
                    composite extract.

            11.1.4  Add one or two clean boiling  chips  to the evaporative flask
                    and attach  a  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 (60-65°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.   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 min.  Remove
                    the Snyder  column,  add  50 mL of  hexane  and  a new boiling
                    chip  and reattach the Snyder column.   Increase the water
                    bath temperature  to 85-90°C and concentrate the  extract to
                    approximately 1  mL.   Rinse the  flask and  the lower joint
                    with  1-2 mL hexane.  Concentrate the extract to  1 mL under
                    a gentle stream of nitrogen.   If  further extract  processing
                    is  to  be  delayed, the  extract  should  be quantitatively
                                      48

-------
              transferred to a Teflon-sealed, amber, screw-cap  vial  and
              stored refrigerated and protected  from light.

      11.1.5  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.

11.2  Sample Extraction --  Liquid-Solid Extraction

      11.2.1  Preparation of disks

              11.2.1.1    Insert the disk into the 47mm  filter apparatus.
                         Wash the  disk  with about  .10  mL of benzene  by
                         adding the solvent  to  the disk, pulling  about
                         half through the disk and allowing it to soak the
                         disk  for  about   a  minute,  then  pulling  the
                         remaining rinse solvent through the disk.  With
                         the vacuum  on,  pull air through  the disk  for
                         about one minute.

              11.2.1.2    Pre-wet  the disk with 10 ml methanol  {MeOH)  by
                         adding the MeOH to the  disk, pulling about half
                         through  the disk  and allowing  it  to soak  for
                         about a minute, then pulling  MOST of the  MeOH
                         through. A layer of MeOH should be  left  on  the
                         surface  of the  disk, which should not be allowed
                         to  go dry from this point until the end  of  the
                         sample extraction. This  is  an  important  step to
                         ensure uniform flow  and good analyte recoveries.

              11.2.1.3    Rinse the  disk with 10 ml reagent water by adding
                         the water to the disk and  pulling most  through,
                         again leaving a layer on the surface of the disk.

      11.2.2  Mark the  water miniscus on the side  of the 1-L sample bottle
              for later determination of the exact  sample volume.

      11.2.3  Add the  water sample,  to which  all  necessary  surrogate
              compounds and  internal  standards have  been  added  according
              to Section 11.1.2,  to the reservoir and turn on the vacuum
              to begin  the extraction. Aspirator vacuum should be adjusted
              to  allow  the   sample   to   pass   through  the  disk   in
              approximately    20  minutes.   Extract  the  entire  sample,
              draining   as   much   water  as  possible   from   the  sample
              container. After all  the  sample has  passed through, draw
              air through the  disk  for about  10 minutes  to remove some of
              the residual water.

      11.2.4  Remove the filtration top  from the apparatus, but  do  not
              disassemble the reservoir and fritted  base.  Empty  the
                                49

-------
              water from the  flask  and  insert  a suitable sample tube to
              contain the eluate. The only constraint on the sample tube
              is  that  it  fit around  the  drip  tip of  the  fritted base.
    ;          Reassemble the apparatus.

      11.2.5  Add 5 ml benzene to the sample bottle and rinse the inside
              of the container. Transfer this benzene to the disk with a
              dispo- pipet or other suitable vessel, rinsing the sides of
              the filtration reservoir in  the process.  Pull about half of
              the benzene through the disk,  release the vacuum, and allow
              the disk  to soak for about a minute.  Pull  the remaining
              benzene through the disk.

      11.2.6  Repeat  the  above  step twice.  Pour the  combined eluates
              through a small  funnel containing about 3 grams of anhydrous
    \       '   sodium sulfate. The  sodium  sulfate may  be  contained in a
              prerinsed filter paper, or by  a plug of prerinsed glass wool
           -   in the stem of  the funnel. Rinse the sodium sulfate with a
              5 ml aliquot of benzene.

      11.2.7  Quantitatively  transfer the  combined eluate  to a suitable
              graduated concentrator tube,  and  rinse  the test tube with
     ^   '      benzene. Using  micro-Kuderna-Danish or  nitrogen blowdown,
              concentrate the eluate almost to dryness,  then add hexane
              to bring the volume to  1  ml for sample extract cleanup.

      11.2.8  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  Sample  Extract Cleanup

      11.3.1  Chromatography  columns   1  and  2,  described   below,  are
              recommended  for  every sample  extract.    A third  column
              containing silica gel  and carbon may be useful for removal
 "•""' -         of  interferences from some sample  extracts and may be used
              at the analyst's discretion.  Because each cleanup procedure
           •   ' increases  the  chances  of  analyte loss,  such  procedures
              should  be minimized.   Criteria  for predicting  when the
              carbon  column  will  be  needed  are  not  available,  but that
              column  is probably  not needed for finished drinking water
              samples that have been  filtered through granular activated
              carbon.

^      11.3.2  Column Preparation

              11.3.2.1   Column 1.   Place 1.0 g  of silica gel (See NOTE)
                          into a 1.0 cm  X 20  cm column  and tap the column
                         gently  to  settle   the  silica  gel.    Add  2  g
                         potassium  hydroxide impregnated silica gel, 1 g
                          silica gel,  4.0  g of sulfuric acid  impregnated

                                50

-------
                         silica gel,  and 2  g silica  gel.    Tap  column
                         gently after each addition.   NOTE:   The  silica
                         gel for this application is partially deactivated
                         with  1%  water  immediately  before  packing  the
                         column.

              11.3.2.2   Column 2.  Place 6.0 g of alumina into a 1.0 cm
                         X  30  cm column  and tap the  column  gently to
                         settle the alumina.  Add a 1-cm layer of purified
                         sodium sulfate to the top of the alumina.

              11.3.2.3   Add hexane to each  column until  the packing is
                         free  of  channels  and  air  bubbles.   A  small
                         positive pressure (5 psi) of clean nitrogen can
                         be used if needed.

      11.3.3  Quantitatively transfer the  sample extract to the top of the
              silica gel in column  1.  Rinse the concentrator tube with
              two 0.5 ml portions  of hexane;  transfer rinses to Column 1.
              With  90  ml  of  hexane,  elute  the  extract  from Column  1
              directly into Column 2.

      11.3.4  Add an  additional  20 ml of hexane to  Column  2  and elute
              until  the hexane level is just below the top of the sodium
              sulfate; discard the eluted hexane.

      11.3.5  Add 20  ml  of 20% methylene chloride/80% hexane  (v/v)  to
              Column 2 and collect the eluate.

      11.3.6  If carbon column cleanup is selected,  proceed with Section
              11.3.7.  If not, proceed with  Section  11.3.8.

      11.3.7  Optional cleanup with  Column 3.  Reduce the volume of eluate
              from Column 2 to about 1 ml in a  K-D apparatus.  Transfer
              the concentrated eluate from  Column 2  to a 4  mm X  200 mm
              column (2 ml disposable pipette) containing  200 mg  silica
              gel/carbon.  Elute with 15  ml  methylene chloride and 15 mL
              80%  methylene  chloride/20%   benzene   (v/v).  in  forward
              direction of  flow.   Discard these  fractions.  .Elute TCDD
              with 15 ml toluene  in a reverse direction  flow.   Collect
              this eluate.

      11.3.8  Concentrate the  eluate  (either the toluene  fraction from
             'Section 11.3.7  or  the methylene chloride/hexane fraction
              from  Section 11.3.5)  to   a small  volume  {<0.5 ml)  and
              transfer to a 1-mL minivial.  Store  the extract in the dark
              at 4°C  until  just before analysis.  Note:  The final  volume
              is adjusted to 10 /uL  immediately before 6C/MS analysis.

li.4  GC/MS Analysis of Extracts
                                51

-------
      11.4.1   Remove the sample or blank extract from  storage  and  allow
              it to warm to  ambient  laboratory  temperature.   Add  a  10-jiL
              aliquot of the RS solution (Section 7.15)  to the extract and
              reduce the extract volume to  10  ML  with a stream  of dry,
              purified nitrogen.

      11.4.2   Inject a 2-/iL aliquot  of the extract  into the GC, operated
              under  conditions  previously  used to  produce  acceptable
              results with the performance check solution.

      11.4.3   Acquire  SIM data  using  the same  analytical  conditions
              previously used to determine RFs.

11.5  Identification Criteria

      11.5.1   Obtain SICPs for each  ion monitored.

      11.5.2   The abundance of m/z 332 relative to m/z 334 produced by the
              IS must  be  >0.67  and  <0.87, and these ions  must maximize
              within  1  scan  of  each  other.   Retention  time  should  be
              within ±5  scans of that  observed during the  most recent
              acceptable calibration.  For good  performance, the retention
              time of  the IS  must be reproducible to  ±5 scans from one
              injection to  the  next.   Over the course  of a  12-h work
              period, the IS  retention  time should be reproducible within
              +10 scans.  Less reproducible IS retention times indicate
              serious  chromatography  problems  that should  be  corrected
              before further sample  analyses.

      11.5.3   The sample  component  must produce a signal  for  both ions
              monitored to detect and measure unlabeled 2,3,7,8-TCDO, and
              the abundance of m/z 320 relative to  m/z  322  must be >0.67
              and <0.87.  All  ions  must maximize  within 1  scan  of each
              other and within 3 sec of the IS.

      11.5.4  The S/N  ratio  for  each unlabeled TCDD and SC  ion  must be
              >2.5 and must not have saturated  the detector; the S/N ratio
              for  each IS and  RS  ion  must be  >10 and  must not  have
              saturated the detector.

      11.5.5  Additional  qualitative  confirmation  can  be obtained  by
              monitoring m/z 257 and 259 (fragment  ions produced by loss
              of  COC1  from   the  analyte)  along  with ions  previously
              specified or by reanalysis of an aliquot  of the extract to
              monitor m/z 257 and 259 along with m/z 268 and 270, fragment
              ions produced  by  loss of COC1 from the  IS.   The relative
              abundance of m/z 257 to 259 and m/z 268 to 270 should be 0.9
              to 1.1, and the abundance of 259 to  270 should be the same
              as  the  ratio  of  322  to 334  measured   in  the  previous
              injection.  Although variable with instrumental conditions,
              the abundance of fragment ions  relative  to molecular ions
                                52

-------
                    is approximately 30-45% for  each  compound;  therefore, the
                    detection limit  for these ions will  be greater  than for
                    molecular ions.
12.    CALCULATIONS
      12.1  From appropriate SICPs of nominal  m/z 259, 320 and 322, obtain and
            record the spectrum number of the apex of the chromatographic peak
            produced  by  unlabeled TCDD  and  the  area of  the entire  chroma-
            tographic peak.

      12.2  Calculate the concentration using the formular

                      Cx   «  
-------
            Ar.  .=  sum of areas for m/z  332  and  m/z  334  produced by the
                   RS,

            Qrs  =  quantity  (picograms)  of RS added  to the sample,

             RF  =  mean RF measured for the RS relative to the IS during
                   initial calibration,  and

              V =  Volume (liters) of water  extracted.

12.5  Report calculated concentrations with three  significant figures when
      measured concentration  is  >100 pg/L and with two significant figure
      when value is <100 pg/L.   The recovery  of the IS should be >40% and
IS
12.6  Estimated Maximum Possible Concentration  (EMPC)  --  For samples in
      which no  unlabeled  2,3,7,8-TCDD is detected,  calculate  the EMPC,
      which is  the  concentration  required to produce  a  signal  with S/N
      ratio of 2.5.   The background signal level (area or height) within
      ±5 scans of the  IS  peak  is  determined  as  previously described and
      is multiplied by 2.5.  With  the following formula,  the product is
      related to the estimated  unlabeled  TCDD  concentration required to
      produce a signal  equivalent of 2.5 S/N.

              EMPC   =  2.5 '  Bx • Qis / Ais •  RF  ' V

                 Bx   =  background (height or area)  for  either nominal
                        m/z 320 or 322 within ±5 scans of the IS peak,

                        peak height  or area  (depending on selection for
                        Bx) for nominal m/z 332  when m/z  320 is used to
                        determine Bx or nominal  m/z 334 when m/z 322 is
                        used to determine Bx> and

                 Qis, RF,  and V  retain previous  definitions.

12.7  An  interference  results  when  sample  a  component  elutes  in the
      retention time window for 2,3,7,8-TCDD and produces both monitored
      TCDD   ions  but   measured  relative   abundances   do   not  meet
      identification  criteria.    Any  ion with  S/N of  <2.5  should  be
      ignored.  Ions producing  S/N of >2.5 but with unacceptable relative
      abundance should be  treated as  an interference,  and a quantitative
      estimate  of  that  interference should  be  calculated  using the
      equation in Section 12.2.  Interferences observed in blanks and also
      present in samples  should not  be reported as a sample interference
      but should be reported as a blank interference.

12.8  Table 5 lists  results of analyses of fortified reagent  water samples
      carried out using the Empore disk extraction method  according to the
      procedure  detailed  in Section  11.2.  Even though  this method was
      developed for only 2,3,7,8-TCDD, since the other  dioxins  and furans

                                 54

-------
            had been studied,  the results were included.  The fortifying levels
            were 0.16 ng/L for the tetra  isomers, 0.8 ng/L for the penta, hexa,
            and hepta isomers, and 1.6 ng/L for the octa isomers.  The average
            recovery for all isomers in  all replicate analyses  is  80% with an
            11% relative  standard  deviation.  No  clean  up  was  done  on  these
            samples.
13.   REFERENCES
      1.   "Water Solubility of 2,3,7,8-Tetrachlorodibenzo-p-dioxin," L. Marple,
          R.  Brunck, and L. Throop, Environ;  Sci.  and Techno!. 1986, 20(2), 180-
          182.

      2.   "Carcinogens  -- Working  with Carcinogens,"   Department of Health,
          Education,  and Welfare,  Public  Health  Service, Centers  for Disease
          Control,  NIOSH,  Pub.  #77-206,  August  1977.

      3.   "Safety in Academic Chemistry Laboratories," American Chemical Society
          Publication,  Committee  on Chemical  Safety,  3rd  Ed.,  1979.

      4.   Statement of  Work, Dixoin Analysis, Soil/Sediment and Water Matrices,
          IFB WA86-K357,  September 1986.
                                      55

-------
TABLE 1.  POTENTIAL INTERFERENCES
Compound
Heptachlorobiphenyl
Nonachlorobiphenyl

Tetrachl oromethoxy-
biphenyl
Tetrachl orobenzyl -
phenyl ether
DDT

DDE

Pentachl orobenzyl -
phenyl ether
Tetrachl oroxanthene

Hydroxytetrachl oro-
dibenzofuran
Tetrachl orophenyl-
benzoqulnone
Interfering
Formul a
M+ - 2 35C1
M* - 4 35C1
M+ - 3 35C137C1
M*
M+
M*
M*
M+ - H35C1
M* - H35C1
M*
M*
M+ - H35C1
M+ - H35C1
M+
M+
M*
M+
M*
M+
Ion
m/z
321.867
319.8521
321.8491
319.9329
321.9299
319.9329
321.9300
319.9321
321.9292
319.9321
321.9292
319.9143
321.9ll4
319.9143
321.9114
319.8966
321.8936
319.8966
321.8936
Required
Resolution
12,476
7,189
7,233
8,805
8,848
8,813
8,843
9,006
9,050
9,011
9,050
18,043
18, 104
18,043
18,104
—
-
                56

-------
                         TABLE 2.  IONS TO BE MONITORED
Accurate
 Mass
                                Elemental
                                Composition
                                Compound
258.9298


319.8965


321.8936


327.8847
331.9368
and
333.9339
and
                                C12H435C1402
C12H435C1337C102


C12H437C1402
                                "C12H435C1337C102
                                                          Unlabeled 2,3,7,8-TCDD
                           Unlabeled 2,3,7,8-TCDD
                                                          Unlabeled 2,3,7,8-TCDD
                                                          37
                                                            Cl4-2,3,7,8-TCDD(SC)
                                                          13
                            C12-2,3,7,8-TCDD (IS)


                           13C12-1,2,3,4-TCDD (RS)
                           13
                            Cl2-2,3,7,8-TCDD (IS)
                                                          13,
                                                            C12-1,2,3,4-TCDD (RS)
                                       57

-------
                    TABLE 3.  GC OPERATING CONDITION GUIDELINES
Column coating
Film thickness
   i
Column dimensions
Helium* linear
 velocity
Initial temperature
Initial time
Temperature program
Retention time of
 2,3,7,8-TCDD
SP-2330
0.2 urn
60 m X 0.24 mm
28-29 cm/sec
at 240°C
70°C
4 min
Rapid increase to 200°C;
200°C to 240°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
*Hydrogen is an acceptable  carrier gas.
                                         58

-------
         TABLE 4.  COMPOSITION OF CONCENTRATION CALIBRATION SOLUTIONS
               Analyte
Surrogate Cmpd.    Internal Std.    Recovery Std.
             Unlabeled
CAL f       2,3,7,8-TCDD
37C1 -              13C
2,3,47,8-TCDD     2,3,^8-TCDD
                                                                      13
                                        c,,-
               2 pg//iL
  0.6 pg//iL
                   50
30
2
3
4
5
10
50
100
200
1.2
1.8
0
0
50
50
50
50
30
30
30
30
                                      59

-------
           TABLE 5.   RECOVERY  OF  CHLORINATED DIOXINS AND FURANS FROM
                     FORTIFIED* REAGENT WATER USING EMPORE DISK EXTRACTION
Compound
TCDF
TCDD
PCDF
PCDD
HxCDF
HxCDD
HpCDF
HpCDD
OCDF
OCDD
No. Samples
2
2
4
2
8
6
4
2
2
2
% Recovery
72
75
78
86
83
80
77
80
91
82
% RSD
6
.0
11
5
16
11
23
10
15
11
 * Fortifying levels were 0.16 ng/L for the tetra isomers,  0.8 ng/L
for the penta and hexa Isomers, and 1.6 ng/L for the octa isomers.
Analyses were carried out using the procedure described in  Section  11.2
of this method.
                                     60

-------
                                  I
                                 i
61

-------

-------