oo  a
        VOLATILE ORGANIC COMPOUNDS IN WHOLE BLOOD - DETERMINATION BY
                  HEATED DYNAMIC HEADSPACE PURGE AND TRAP
                          ISOTOPE DILUTION GC/MS
                                    •by

                              Paul H. Cramer
                             Kathy E. Boggess
                             John M. Hosenfeld
                        EPA Contract No.  68-02-4252
                         MRI Project No.  8822-A(01)
                               July 24,  1987
                                    For

                     National  Human Monitoring  Program
                   U.S.  Environmental  Protection  Agency
                           Field Studies  Branch
                       Design  and Development Branch
                Office of Pesticides and  Toxic  Substances
                            401 M Street,  S.W.
                           Washington,  DC  20460

Attn:   Ms.  Janet Remmers  and Mr.  Philip Robinson, Work Assignment Managers
       Dr.  Joseph Breen  and Ms.  Cindy  Stroup, Program Managers

-------
                                 DISCLAIMER


This document has been reviewed and approved for publication by the Office of
Toxic Substances, Office of Pesticides and Toxic Substances, U.S. Environmental
Protection Agency.  The use of trade names of commercial products does not
constitute Agency endorsement or recommendation for use.

-------
                                   PREFACE
          This method was developed under EPA Contract Nos.  68-02-3938  and
68-02-4252 for the Field Studies Branch  in the Office of Toxic  Substances.
The method is primarily intended to be used  in the National  Blood Network
to establish the baseline levels of volatile organics in the general popu-
lation of the United States.

          This method was developed by Mr. Paul Cramer and Ms.  Kathy Boggess.
Mr. John M. Hosenfeld was the MRI Work Assignment Leader.  We gratefully
acknowledge the valuable assistance given by Ms. Janet Remmers, EPA Work
Assignment Manager and Dr.  Joseph Breen, EPA Project Officer.

                                        MIDWEST RESEARCH INSTITUTE
                                        Paul H. Cramer
                                        Assistant Work Assignment Leader
                                        ]ohn M. Hosenfeld
                                        Work Assignment Leader
Approved:
Jack Balsinger
Quality Assurance Coordinator
Paul C.  Constant
Program Manager

-------

-------
1.0  Scope and Application

     1.1  This method covers the determination of 30 volatile organic
          compounds in whole human blood.   The following compounds may be
          determined by this method:
                    Compound                        CAS no.


               Benzene                               71-43-2
               Bromodichloromethane                  75-27-4
               Bromoform                             75-25-2
               Carbon tetrachloride                  56-23-5
               Chlorobenzene                        108-90-7
               Chloroform                            67-66-3
               Dibromochloromethane                 124-48-1
               1,2-Oichlorobenzene                   95-50-1
               1,3-Dichlorobenzene                  541-73-1
               1,4-Dichlorobenzene                  106-46-7
               1,1-Oichloroethane           ,          75-34-3
               1,2-Dichloroethane                    107-06-2
               1,1-Oichloroethene                     75-35-4
               trans-l,2-Dichloroethene   .           156-60-5
               1,2-Dichloropropane                   78-87-5
               cis-l,3-Dichloropropene        •     10061-01-5
               trans-l,3-Dichloropropene    ,       10061-02-6
               Ethyl benzene         •                 100-41-4
               Dichloromethane                        75-09-2
               Styrene                              100-42-5
               1,1,2,2-Tetrachloroethane              79-34-5
               Tetrachloroethene                     127-18-4
               Toluene                              108-88-3
               1,1,1-Trichloroethane                  71-55-6
               1,1,2-Trichloroethane                  79-00-5
               Trichloroethene                        79-01-6
               Trichlorof1uoromethane                 75-69-4
               1,2-Xylene                             95-47-6
               1,3-Xylene          .                  108-38-3
               1,4-Xylene                            106-42-3


    1.2  This  is a  heated dynamic  headspace purge and trap  gas chromato-
         • graphic method using mass spectrometry in  the limited mass scan
         •monitoring  mode.  The method is applicable  to the  determination of
         the compounds  listed above in an approximate concentration range of
         50 parts per trillion (ppt) to 5 parts per  billion (ppb) in 35-mL
         of whole human blood.

    1.3  Quantisation is achieved  by isotope dilution, where possible, or
         versus an appropriate internal standard with purging characteristics
         similar to  the native compound.

-------
     1.4  At this time, the estimated method detection limit (MDL) or  limit
          of detection (LOD) for each analyte has not been fully evaluated.
          Preliminary investigations have determined the compound LODs to
          range from 50 to 300 ppt in a 35-mL blood sample, depending  upon
          the compound.


2.0  Summary

     2.1  Volatile organic compounds in whole blood are determined by  a heated
          dynamic headspace purge and trap GC/MS method.   Stable isotopically
          labeled analogs of compounds of interest are added to a 35-mL blood/
          10-mL water sample contained in a specially designed purge vessel.
          Prepurified nitrogen is passed over the surface of the heated (50°C)
          and stirred blood/water mixture, removing volatile organics  from
          the sample into the gas stream.   The purged sample components pass
          through a two-way, six-port valve and are adsorbed on a polymeric
          trap.   After a known volume of purge gas has passed over the sample,
          the adsorbent trap is heated and backflushed with prepurified helium,
          desorbing the purgeable components onto the head of a wide-bore
          fused silica GC column interfaced to a mass spectrometer.   The mass
          spectrometer is operated in the limited mass scan (LMS) mode where
          only selected ions are scanned.   Quantitation of a detected analyte
          is determined from specific ion responses from standards of the ana-
          lytes and their corresponding labeled analogs.   The responses of
          the labeled compounds are used to correct the variability of the
          analytical  technique through the use of an isotope dilution calcu-
          lation procedure (EPA 1985) or an internal standard calculation
          procedure.


3.0  Interferences

     3.1  Compounds that have similar chromatographic properties and charac-
          teristic mass spectral  ions as the target compounds listed in Sec-
          tion 1.1 may interfere with the determinations.   Different ions may
          need to be  monitored to analyze samples that contain interfering
          compounds.

     3.2  Impurities  in the purge gas and organic compounds out-gassing from
          the plumbing ahead of the trap may account for  some of the contami-
          nation problems.   The analytical  system must be  demonstrated to be
          free from contamination under the conditions of  the analysis by
          running laboratory reagent blanks as  described  in Section  9.1.   The
          use of non-TFE plastic tubing,  non-TFE thread se_alants,  or flow
          controllers with rubber components in the purging device should be
          avoided.

     3.3  The water used to dilute the  whole blood samples  is of particular
          concern.  Due to the extremely low levels  achieved  by this  method,
          the water used for sample dilution must be shown  to be acceptably
          free of the compounds of interest.

-------
     3.4  Preparation of volatile-free water and handling of the blood sam-
          ples should be conducted in a laboratory room dedicated to this
          purpose or in some other designated volatile-free area.  All
          organic solvents, with the exception of methanol, should be pro-
          hibited from use or storage in this laboratory.

     3.5  Contamination by carryover can occur whenever high level and low
          level samples are sequentially analyzed.   To reduce carryover, the
          purge vessel must be changed between sample analyses.  Whenever a
          sample with a high concentration of volatile organics is encoun-
          tered, it should be followed by an analysis of reagent water to
          check for cross contamination.   The trap and .other parts of the
          system are also subject to contamination; therefore, frequent bake-
          out and purging of the entire system may be required.
4.0  Safety
     4.1  The toxicity or carcinogenicity of each reagent used in this method
          has not been precisely defined; however, each chemical compound
          should be treated as a potential health hazard.   From this view-
          point, exposure to these chemicals must be reduced to the lowest
          possible level by whatever means available.   The laboratory is re-
          sponsible for maintaining a current awareness file of OSHA regula-
          tions 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 chem-
          ical analysis.  Additional references to laboratory'safety are
          available and have been identified (NIOSH 1977,  OSHA 1976, ACS 1979)
          for the information of the analyst.

     4.2  The following parameters covered by this method have been tenta-
          tively classified as "known or suspected,, human or mammalian carcin-
          ogens:  benzene,  carbon tetrachloride,  chloroform,  dichloromethane,
          and 1,4-dichlorobenzene.   Primary standards  of these toxic compounds
          should be prepared in  a hood.   A NIOSH/MESA  approved toxic gas res-
          pirator should be worn when the analyst handles  high concentrations
          of these toxic compounds.

     4.3  Analytes detected in the blood samples  by this method are at trace
          levels with minimal  exposure risks to laboratory personnel.   How-
          ever,  the blood samples should be handled with adequate safety pre-
          cautions due to unknown microbiological  activity in the blood.
          Biosafety Level (BSL)  2 procedures'(NIH 1984) should be followed
          when handling blood samples during the  compositing  and analysis
          procedures.   These procedures  include handling blood samples with
          protective gloves in a hood with an  average  face velocity of 100
          linear feet per minute.   After a sample has  been analyzed,  it should
          be decontaminated with a chemical  disinfectant,  autoclaved,  and dis-
          posed  of as biological  waste.   Specific handling instructions  are
          given  in Appendix A of this method.

-------
5.0  Apparatus and Equipment

     5.1  Sample Collection Vial - Glass vial (17 x 60 mm, ~ 8.5 ml) with a
          Teflon-lined screw cap.  Special cleaning described in Section 7.1
          is required prior to use.  Detergent wash, rinse with tap and
          distilled water, rinse with pesticide-grade methanol, and dry at
          105°C in the designated volatile-free area before use.  An approxi-
          mate 5-mL graduation line is to be made on the vial.

     5.2  Purge and Trap Device - The purge and trap device consists of a
          dynamic headspace purging vessel, a stainless steel adsorbent trap,
          and a desorber heating element.  Figure 1 shows the arrangement of
          these components.

          5.2.1   A purge vessel such as that shown in Figure 2 is used in
                  the determination of volatile organic content in the blood.
                  A vessel volume of 60 ml as measured to the tip of the purge
                  inlet is required.  Any glass vessel of similar design and
                  function is adequate for this analysis.  The vessel is con-
                  nected to the purge gas supply and purge/desorb valve via
                  Teflon tubing to allow flexibility and chemical inertness.

          5.2.2   The stainless steel trap (10 cm x 2.64 mm ID) is packed with
                  ~ 100 mg of Tenax TA® (35/60 mesh) adsorbent.  Alternatively,
                  the trap can be purchased commercially as the required trap
                  for the EPA "Method 624-for purgeables.  However, the silica
                  gel in this trap adsorbs water which subsequently is passed
                  into the GC/MS system and can cause degeneration of the ion
                  source after repeated analyses.  The ion source may need to
                  be cleaned or replaced more frequently if silica gel" is used
                  in the trap.

          5.2.3   Gas flow through the trap is directed with a two-way, six-
                  port valve.  All transfer lines are heated to 80°C.

          5.2.4   The Tenax trap is contained in a desorption unit activated
                  by a temperature sensing and controlling device.  The de-
                  sorber heats the trap from 0°C to 220°C in less than 1 min
                  and maintains the temperature within a 5°C range.

     5.3  GC/MS System

          5.3.1   The gas chromatographic system includes a 30 m x 0.53 mm ID
                  DB-624 fused silica column (J&W Scientific, Folsom, CA).
                  The recommended GC parameters are:

                  Temperature Program:  30°C (2-min initial hold) to 225°C
                    at 10°C/min (10-min final hold)
                  Column Flow:  20 mL/min He
                  Purge Flow:  35 mL/min N2

                  Equivalent GC parameters may be used.

-------
                                    Temperature
                                    Controller
         Two - way
         Six-Port
         Valve
      Stainless
      Steel (1/16")
      Heated
      Transfer
      Line

             Union
                     Antiseptic
                     Solution
                     (or Class A Hood)
Water
Temperature
Controller
Teflon
Tubing
(1/8"
O.D.)

 Outlet - to
  Tenax Trap
                                  Water Jacket
                         Teflon Stopper
24/40 Joint
   J/4" O.D.
    Purge  Gas Inlet
 Flow
•Controller
 for N2
• • -r
t 	 -"'""'
~L_ i"~^' ,. .

	 	 IT
-aa,

/
                                                  Magnetic
                                                  Stirrer
                                              	
                               Return Water Line
             Figure  1.   Dynamic headspace  purge and trap equipment.

-------
         1/4" O.D.
         Outlet -
         to Tenax Trap
               Water Jacket
Water (50°C) jn
                                                    Teflon Stopper
                                                    24/40 Joint
                                       Blood/Water
                                       Sample
1/4" O.D.
Purge Gas Inlet
                                                         ,	^
         Water Out
                                                                Stir Bar
                     Figure 2.   Volatile organic purge vessel.

-------
     5.3.2   The recommended MS parameters are:

             Electron Energy:  70 eV
             Electron Multiplier Voltage:  -1950 V
             Emission Current:  0.3 mA
             Scan Parameters:  Limited mass ranges given below:
             Total Scan Time:  1.1 s
             Interval no.
                             Time request
Low mass       High mass         (s)
1
2
3
4
5
6
7
8
9
10
11
49 .,515
60.518
74.522,
80.524
90.527
110.534
115.535
126.538
145.544
163.549
168.550
56.517
68.520
79.523
88.526
106.532
113.534
120.536
131. 540
152. 544
166.550
174.552
0.100
0.100
0.100
0.100
0.200
0.050
0.100
0.100
0.100
0..050
0.100
             Equivalent MS parameters may be used with the exception of
             the mass ranges.   Different mass ranges may be used if
             interferences are encountered in the analysis of selected
             samples.  Changing the mass range scanned, however, would
             require recalibration of the GC/MS system.

     5.3.3   A data system is interfaced to the mass spectrometer for
             acquisition of a data file for each chromatographic analy-
             sis.   The raw data are stored on accessible discs for manual
             interpretation.   Ion abundances of a specific mass are
             plotted versus scan number.  The computer measured peak
             areas of the characteristic mass fragmentation ions of a
             particular analyte and its corresponding labeled analog are
             used to determine the concentration of the analyte in the
             blood.

5.4  Glassware - Fixed needle glass syringes with ± 1% accuracy are re-
     quired for addition of internal  standard to the solution.  Syringes
     of 10, 50, 100, and 250 uL should be available for dilution of stock
     solution for standard preparation.

5.5  Analytical Balance - Analytical  balance accurate to the nearest
     0.1 mg (for standards preparation) and to the nearest O.Ol.g (for
     sample weight determination).

-------
6.0  Reagents, Materials, and Standards

     6.1  Methanol - Pesticide quality or equivalent for dilution of stock
          standards and glassware preparation.

     6.2  Reagent Water - The water used in the dilution of the blood sample
          is obtained from a Millipore® reagent water purification system.
          This water is then passed through a carbon filter bed into a glass
          holding tower where the water is held at 50°C and continuously
          purged with prepurified nitrogen.   It is recommended that this
          volatile-free water be generated and stored in the designated
          volatile-free area.

     6.3  Stock Standard Solution - Stock standard solutions can be prepared
          from neat compounds of 99% purity on the target list (Section 1.1)
          Neat materials can be obtained from commercial suppliers (e.g.,
          Aldrich Chemical Company).   Individual  stock standards in a water
          miscible solvent can be obtained from the EPA Repository for Toxic
          and Hazardous Materials (EPA/EMSL-Cincinnati).

          Stock standard solutions are prepared from neat compounds by deter-
          mining the weight of the neat compound  added to a known amount of
          methanol.   Specifically, place approximately 9.8 ml of methanol
          into a 10-mL ground glass stoppered volumetric flask.   Allow the
          flask to stand unstoppered  for about 10 min or until  all alcohol-
          wetted surfaces have dried.   Weigh the-flask to the nearest 0.1 mg.
         .Then,  using a 100-uL glass  syringe, add two or more drops of neat
          compound to the flask and reweigh.   Alternatively,  using the den-
          sity of the compound,  determine the volume needed for a specific
          mass,  deliver that amount to the flask,  then reweigh  the flask.
          In either case, the difference in  weight is the amount of neat
          compound added.

          The liquid must fall  directly into the  solution without contacting
          the neck of the flask.   After the  weight of the added compound has
          been determined, dilute the  standard  to volume with methanol,  stop-
          per,  and invert several  times before  diluting to secondary standard
          levels.

          The stock standard solutions are stored at -20°C in Teflon-lined
          screw-cap vials.   The  recommended  maximum storage time for'concen-
          trated stocks in methanol  is 1 month  under these conditions.

     6.4  Secondary Dilution Standards - Mixed  secondary standards are  pre-
          pared  by diluting the  stock  standard  solutions into one solution  of
          methanol.   Dilutions  of this mixed secondary standard  are made to
          generate working solutions for the calibration curve.   These  working
          solutions  should be freshly  prepared  on  a weekly basis.   Secondary
          standards  can be stored at -20°C for  up  to 2 weeks.

-------
     6.5  Labeled Analog Solution - Stock solutions of the labeled analogs
          are prepared from the neat compounds following the same procedures
          as described in Section 6.3.  Neat materials can be obtained from
          commercial suppliers (e.g., MSD Isotopes).  A mixed analog spiking
          solution is prepared by taking an aliquot of each stock solution to
          an exact volume of methanol.  Dilutions of this solution are made'
          to yield an analog concentration of 3.5 ng/uL for each of the labeled
          compounds.  With each sample analysis, 5 uL of this solution is added
          to the purge vessel.  Stock solutions can be kept for 1 month at
          -20°C.  Working solutions should be freshly prepared on a weekly
          basis.

     6.6  Mixed Analyte/Analoq Solutions - Aliquots of the mixed secondary
          analyte solution (Section 6.4) and the mixed labeled analog solution
          (Section 6.5) are combined to generate working solutions for the
          calibration curve.   These solutions are prepared such that the con-
          centration of the labeled analogs remains constant and the analyte
          concentrations are varied.  These calibration solutions should be
          kept at -20°C for storage and should be freshly prepared on a weekly
          bas is.

     6.7  Anticoagulant Solutions - An aqueous anticoagulant solution is pre-
          pared by dissolving 500 mg of EDTA (ethylenediaminetetraacetic acid
          disodium salt, Fisher Scientific Company) in 10 ml of organic-free
          reagent water for a final concentration of 50 mg/mL   Two hundred
     . •    microliters of this solution is used for each 5 ml of blood col-
         lected.   This solution should be freshly prepared on a monthly basis.

     6.8  Charcoal - Activated charcoal  (cocoanut,  activated,  8-12 mesh) is
          used for the storage of samples.

     6.9  Trap Materials - 2,6-Diphenyloxide polymer - Tenax® TA (35/60 mesh).

     6.10 Disinfectant Solutions - Household bleach containing at least 5%
          NaOCl  (e.g., Clorox®) diluted 1:10 with tap water.


7.0  Sample Preservation,  Collection,  and Storage

     7.1  Sample Preservation - Prior to collection of the blood samples,  the
          sample vials are rinsed with reagent water and pesticide quality
          methanol.   The vials should then be dried at least 4 h at 105°C.
          After the vials  have cooled in a volatile-free environment (e.g.,
          sealed chamber containing charcoal),  200  uL of the EDTA anticoagu-
          lant solution at 50 mg/mL is added to each vial. _  The vials  are'then
          capped with Teflon-lined screw caps and stored fn  containers  with
          charcoal in the  designated volatile-free  area until  used for  blood
          collection.

-------
     7.2  Collection - The blood sample  is collected by the blood  center  per-
          sonnel after the collection of the donated unit of blood and  any
          pilot tubes.  To collect the blood sample, disconnect the collec-
          tion bag from the blood collection tubing by clamping the tubing to
          stop blood flow.  Then cut the tubing between the clamp.and the
          blood bag, process the unit as normal, and fill any pilot tubes as
          necessary.  Open the two collection vials.  Unclamp the  tubing  and
          allow the blood to flow from the donor through the tubing into  two
          collection vials.  Fill each vial to the 5-mL graduated  mark.   Cap
          the vials.  Remove the needle  and tubing from the donor.  Invert
          the vials at least 10 times to mix the anticoagulant with the blood
          sample.

     7.3  Storage - Following collection of the blood sample at the blood col-
          lection center, the sample is  transferred to the shipping container
          as soon as possible.   The shipping container will be fitted with
          vial racks, which will hold the samples.  Wet ice will be added to
          cool the samples to about 5 to 10°C.   After receipt and  log-in  at
          the analytical laboratory, the samples should be grouped according
          to the sample-compositing design and placed in a wide-mouth jar
          half-filled with cocoanut charcoal.   The jar should be sealed and
          placed in a refrigerator at 4°C.   The blood samples for  volatile
          analysis should never be frozen.


8.0  Calibration

     Mass calibration of the mass spectrometer is conducted according to  manu-
     facturer specifications.   The isotope dilution method is based upon  the
     response and chemical behavior relationship between the analyte and  its
     labeled analog.   Method calibration of the GC/MS system can be conducted
     by direct injection of varying levels of the analytes with a constant
     amount of the corresponding labeled analog.   That is, standards need not
     be purged from water to calibrate the purge and trap-GC/MS system.

     8.1  The gas chromatograph must be operated using temperature and flow
          rate parameters equivalent to those in Section 5.3.1.

     8.2  Mass calibrate the GC/MS system daily with an instrument manufac-
          turer's specified calibration gas (e.g., perfluorotributylamine) to
          ensure proper mass identification.   Since this is a low resolution
          method, only unit resolution is required.

     8.3  Calibrate the GC/MS  system using isotopic dilution and internal
          standard techniques  as described below.

          8.3.1   Mixed analyte/analog working  solutions (Section  6.6) are
                  used to generate a calibration curve for each analyte  in
                  the range of  2 ng to 200 ng.   This corresponds to ~  60-ppt
                  to ~ 6-ppb levels for a 35-mL (35-g)  composited  blood  sam-
                  ple.
                                      10

-------
             Analyze the standards by direct injection into the GC/MS
             system.  Additional higher and lower level standards may
             also be analyzed until nondetection or nonlinearity occurs.

     8.3.2   Keep the labeled analog amount in each injection constant.
             An amount of 17.5 ng (equivalent to 500 ppt) is recommended.

     8.3.3   The analyte concentration range should consist of at least
             five separate concentration levels (e.g., 2-10-50-100-200 ng).

    •8.3.4   If analytes are detected in the blood/water mixture at levels
             greater than 200 ng, the calibration curve is extended to
             bracket those analytes, if possible.   If the extended cali-
             bration curve becomes nonlinear, then a duplicate aliquot
             of the original blood sample is diluted and reanalyzed within
             the range of the established calibration curve.  If no dup-
             licate is available, the concentration of the analyte is
             qualified by indicating it is outside the calibration curve
             (e.g., > x, where x is the highest level of the analyte
             standard).

8.4  Isotope Dilution Calculations - The labeled analogs chosen for quan-
     titation of the standard analytes in the blood matrix are shown in-
     Table 1.  A labeled analog, deuterium (d) or carbon-13 (13C), is
     used to quantitate the corresponding analyte by-the isotope dilution
     method.  The amount of labeled isotope remains constant during cali-
     bration 'and sample analysis.  The amount of analyte is varied from
     2 to 2.00 ng to generate relative response factors.  The isotope
     dilution method is different from the internal standard method only
     in cases where there is contribution to the response of the analyte
     quantitation ion by the analog, or vice versa.  In these cases, the
     calculations explained in the following section are used to'determine
     the RRF (relative response factor) for the analyte.

     8.4.1   The relative response of the quantitation ion of the analyte
             (Table 2) to the quantitation ion of its labeled analog
             (Table 3) is determined from isotope ratio values computed
             from GC/MS analyses of the analytes only, the analogs only,
             and a mixture of analytes and analogs.  Three isotope ratios
             are determined:  Rx = the isotope ratio measured for the
             pure analyte; Ry = the isotope ratio measured for the pure
             labeled compound; and Rm = the isotope ratio of the analyt-
             ical mixture of analyte and labeled compounds (as in the
             calibration curve solutions).   Quantitation ions have been
             selected so that Rx > Ry and Rm is between 2Ry and O.SRx.

             8.4.1.1  To begin calibration of the GC/MS system, analyze
                      a solution containing only the analytes at a known
                      concentration (e.g.,  20 ng).   Then analyze a solu-
                      tion of only the labeled analogs at the same con-
                      centration (e.g., 20 ng).   From these data,  the Rx
                      and Ry isotope ratios are determined as given below.
                                 11

-------
        Table 1.  Target Compounds and Corresponding  Labeled Analogs
       Compounds
         Labeled analog
Benzene
Bromodichloromethane
Bromoform
Carbon tetrachloride
Chlorobenzene'
Chloroform
Dibromochloromethane
1,2-Dichlorobenzene
1,3-Dichlorobenzene
1,4-Dichlorobenzene
1,1-Dichloroethane
1,2-Oi ch1oroethane
trans-1,2-Dichloroethene
1,2-Oichloropropane
cis-l.S-Oichloropropene
trans-1,3-Oichloropropene
Ethyl benzene
Styrene
1,1,2,2-Tetrach1oroethane
Tetrachloroethene
Toluene
1,1,1-Tri chloroethane
1,1,2-Trichloroethane
Tri chloroethene
1,2-Xylene
l,3-Xylene/l,4-Xylene

Dichloromethane >c
1,1-Dichloroethane0
Trichlorof1uoromethane
 Benzene-d6
 l,2-Dichloropropane-d6
 Bromoform-13C
 Carbon  tetrachloride-13C
 Chlorobenzene-d5
 Chloroform-d
 Chlorobenzene-dsa
 l,3-Dichlorobenzene-d4
 1,3-Dichlorobenzene-d4
 l,3-Dichlorobenzene-d4
•l,l-Dichloroethane-d3
 1,2-0ichloroethane-d4
 trans-1,2-Di chloroethene-d?.
 1,2-Di cnloropropane-dg
 c_j_s-l,3-Dichloropropene-d4
 trans-1,3-D ichloropropene-dd.
 Ethylbenzene-d10
 S.tyrene-dg
 1,1,2,2-Tetrach1oroethane-d2
 Tetrachloroethene-13C2
 Toluene-dg
 1,1,1-Tri chloroethane-d3
 1,1,2-Tri chloroethane-ds
 Trichloroethene-13C
 l,2-Xylene-d10
 l,2-Xylene-d10

 Dichloromethane-d2
b$uggested internal standard.
 High levels of native dichloromethane contribute significantly to dichloro-
Cmethane-d2 and may preclude quantisation.
 These compounds did not meet the accuracy and precision criteria established
 for validation of the method (50-150% recovery).  These compounds should be
 reported as qualitatively detected.
                                    12

-------
Table 2.   Quantitation and Confirmation Ions for Target Compounds
Compound
Benzene
Bromodichloromethane
Bromoform
Carbon tetrachloride
Chlorobenzene
Chloroform
Dibromochloromethane
1, 2- Di chlorobenzene
1, 3-D i chlorobenzene
1,4-Di chlorobenzene
1 , 1-Di chl oroethane
1,2-Di chl oroethane
1,1-Dichloroethene
trans-l,2-Di chl oroethene
Dichloromethane
1, 2-Di chl oropropane
cis-l,3-Dichloropropene
trans- 1,3-Dichloropropene
Ethyl benzene
Styrene
1 , 1 , 2 , 2-Tetrach 1 oroethane
Tetrach 1 oroethene
Toluene
1,1, 1-Tri chl oroethane
1,1,2-Tri chl oroethane
Tri chl oroethene
Trichlorofluoromethane
1,2-Xylene
1,3-Xylene
1,4-Xylene
Quantitation
ion
78
83
173
117
112
83
129
146
146
146
65
64
96
96.
84
76
75
75
9.1
104
83
164
91 .
97
97
130
101
91
91
91
Confirmation
ion
so :
85
171
119
77
85
127
148
148
148
63
62
61
61
86
62
77
77
106
103'
85
129
92
99
83
95
103
106
106
106
                               13

-------
 Table 3.   Quantitation and Confirmation Ions for Labeled Analogs
Compound
Quantitation
     ion
Confirmation
     ion
Benzene-d6
Bromoform-13C
Carbon tetrach1oride-13C
Chi orobenzene-ds
Chloroform-d
l,3-Dichlorobenzene-d4
l,l-Dichloroethane-d3
l,2-Dichloroethane-d4
trans-1, 2-Dich1oroethene-d2
Dichloromethane-d2
l,2-Dichloropropane-d6
cj_s-l,3-Dichloropropene-d4
trans-1 ,3-Di chl oropropene-d4
Ethyl benzene-d10
Styrene-d8
1,1,2, 2-Tetrachl oroethane- d2
Tetrachl oroethene- 13C2
Toluene-d8
1,1, 1-Tri chl oroethane-d3
l,l;2-Trichloroethane-d3
Trichloroethene-13C
l,2-Xylene-d10
84
172
118
117
86
152
68
67
65
88
67
79
79
98
112
86
172
98
100
102
99
116
56
174
120
119
84
150
66
65
100
51
81
81
81
116
84
84
170
100
102
100
,101
98
                                 14

-------
         For Rx,

         n  _ 	area quantitation ion of analyte	
              area analyte contribution to the quantitation
                           ion of labeled analog

         at the retention time (RT) of the analyte during
         the analyte-only analysis.  For Ry,
         Ry =
area labeled analog contribution to the
	quantitation ion of analyte	
area quantitation ion of labeled analog
         at the retention time of the labeled analog during
         the labeled analog-only analysis.

         If no area is detected for a given ion as deter-
         mined by the qualitative criteria given in Section
         11.0 (i.e., there is no contribution), a value of 1
         is assigned in the equations given above.

8.4.1.2  To determine the value of Rm and subsequently RR
         (relative response), analyze a 1.0-uL aliquot of
         each calibration standard (Section 8.3).   Rm is
         calculated using the following equation:

         Rm = area quantitation ion of analyte at RT of analyte
               area quantitation ion of analog at RT of analog

         Rm is then used to determine the RR (relative
         response) of the mixture in each of the three
         cases given in the following sections.

         8.4.1.2.1  If the analyte-analog pair is  chromato-
                    graphically separated such that the
                    height of the valley between the two
                    peaks at the same m/z is less  than 10%
                    of the height of the shorter of the two
                    peaks, then the relative response (RR) =
                    Rm and no correction is needed for over-
                    lap contributions.

         8.4.1.2.2  If there is no mass contribution to the
                    labeled quantitation ion from  the Rx
                    determination (Section 8.4.1.1) and
                    there is no contribution to the analyte
                    quantitation ion from the Ry determina-
                    tion (Section 8.4.1.1), then the relative
                    response (RR) = Rm.

         8.4.1.2.3  If the analyte-analog pair is  not chro-
                    matographically separated and  if respec-
                    tive mass contributions were found from
                    15

-------
                            Rx and Ry determinations (Section 8.4.1.1),
                            then the relative response (RR) is calcu-
                            lated using the following equation:

                            RR _ (Ry - Rm)(Rx + 1)
                            KK   (Rm - Rx)(Ry + 1)

                            where Rx and Ry are values determined
                            in Section 8.4.1.1 and Rm is the analyte/
                            analog peak area ratio in the mixture
                            being analyzed (Section 8.4.1.2).  Rm
                            must be between 2Ry and 0.5Rx (2Ry < Rm
                            < .5Rx).  The calibration curve becomes
                            nonlinear at the point where Rm falls
                            outside this limit.   Significant mass
                            contribution has been observed for
                            dichloromethane and trichloroethene.
                            Alternatively, the analyte could be
                            quantitated by the internal standard
                            method relative to a different labeled
                            analog (Section 8.5).

8.4.2   After the relative response (RR) has been determined (Sec-
        tion 8.4.1.2), the relative response factor (RRF) is then
        calculated as follows.

        DOC - DD ~ ng of labeled compound
        t\l\r — KK X —       ,-    -i .
                       ng of analyte

        for each level of the analyte in the calibration curve.

8.4.3   Consider the following example for clarification of Sections
        8.4.1 and 8.4.2 calculations.

        •  Analyze 1.0 uL of the 20 ng/uL solution containing only
           the anaTytes ' (analyte response = 10,000 area units;
           contribution to the labeled analog response = 1,000 area
           units).

              _ 10,000 _ ,n
                       - 10
           Analyze 1.0 uL of the 20 ng/uL solution containing only
           the labeled analogs (contribution to the analyte response
           = 500 area units; labeled analog response = 15,000 area
           units).
           Analyze a 1.0-uL calibration standard containing 100 ng/|jL
           analytes and 20 ng/uL labeled analogs.   The analyte-analog
                            16

-------
              .  pair is not chromatographically separated (analyte
                response =40,000 area units; labeled analog response =
                10,000 area units).

                Rm = 40,000 = 4
                Km   10,000   4

             '  RR = (Ry - Rm)(Rx + 1)
                     (Rm - Rx)(Ry + 1)

                   _ (0.03 - 4)(10 + 1)
                     (4 - 10)(0.03 + 1)

                RR = 7.1

             •  RRF _ RR x ng labeled analog
                              ng analyte

                    = 7.1 x 20 ng/100 ng

                RRF = 1.4

8.5  Internal Standard Calculations - The internal standard method is
     used to quantitate those compounds where a corresponding labeled
     analog of the compound is not available.  The internal standards
     recommended for quantisation of the analytes in the blood matrix
     are shown in Table 1,  The area responses of the quantitation ion
     for each analyte relative to the area response of the quantitation
     ion for the specified internal standard are tabulated to determine
     a relative response factor (RRF).   The amount of internal standard
     remains constant.   The amount of standard will be varied from 2 to
     200 ng to generate relative response factors using' the equation below.

           AS x ng IS

     RRF = AJS x ng S

   -  where AS = area of the quantitation ion for the analyte measured;

        ng IS = total nanograms of the specified internal standard in-
                  jected into the GC/MS system;
          AT_ = area of the quantitation ion for the internal standard;
     and   Xi
         ng S = total nanograms of the analyte injected into the GC/MS
                  system.

8.6  Ideally, the relative response factors are constant over .the entire
     concentration range of interest.   If the RRF value is constant over
     the working range (^ 30% RSD), the average RRF will  be. used for
     determining the concentration of a given analyte in a blood sample
     (Section 13.0).   If not, analyte concentrations must be determined
     from the actual  calibration curve by plotting RR versus analyte
     concentration for those compounds determined by isotope dilution
     (Section 8.4).   For those compounds quantitated by the internal

                                 17

-------
          standard method (Section 8.5), the concentration must be determined
          by plotting analyte/analog response versus analyte concentration.

     8.7  A dally check of the instrument sensitivity will be performed at
          the beginning and end of each day's analysis with a low level and
          mid level calibration standard, respectively.  The daily relative
          response factor (Sections 8.4 and 8.5) for each analyte must be
          ± 30% of the cumulative average relative response factor from the
          calibration curve and succeeding standard analyses.  The daily
          relative response factor is then incorporated into the average
          relative response factor from the calibration curve, and a cumu-
          lative response factor average is used to quantitate each day's
          analysis.

          If the daily response factor is outside the 30% limit, the daily
          standard analysis must be repeated using a fresh calibration
          standard.  If the criterion is still not met, a new calibration
          curve and relative response factors must be established.

     8.8  External Standard Calibration (Analogs Only) - The external standard
          method is used to establish a calibration curve for determining
          percent recovery of the labeled analogs from the blood matrix.   The
          recoveries of the added analogs are used to determine the stability
          of the system (Section 9.5).   Dilutions of the labeled analog solu-
          tion (Section 6.5) are made to yield standards corresponding to
          20, 50, and 100% recovery.   The. standards are analyzed by direct
          injection (e.g., 3.5, 8.75, 17.5 ng) to form a three-point analog
          calibration curve of area units versus percent recovery.   Percent
          recovery of the labeled analog in a blood analysis is determined
          directly from the linear regression curve.   These standards are
          analyzed throughout the day to monitor changes in instrument sensi-
          tivity.  The suggested sequence of analysis is given in Table 4.


9-0  Quality Control

     9.1  A reagent blank consisting of volatile-free water and labeled ana-
          logs will be analyzed with each day's sample set.   Forty-five mini-
          liters of reagent water is added-to a purge vessel with 17.5 ng of
          each labeled analog and analyzed as an actual sample.   Contaminants
          are determined and quantitated relative to their corresponding
          labeled analog.

          9.1.1   Reagent water (Section 6.2) is required for dilution of the
                  blood samples.   A sufficient volume to complete the day's
                  analyses is collected from the holding tower (Section 6.2)
                  and transferred to a precleaned glass container.   This
                  entire volume of water is prepurged again at the  site of
                  analysis (GC/MS laboratory) and then, transferred  'back to
                  the precleaned glass  container.   This water is used for
                  dilution of the blood during the day's analyses,  and for
                  the reagent water blank.
                                      18

-------
     9.1.2   A water blank should be analyzed  at  the  beginning  of each
             day's analysis, after the daily standard,  to  determine
             potential contamination from diluting  the  samples  with  water.
             If unacceptable levels of target  compounds  are  found in the
             reagent water (e.g., > 50 ppt), the  analysis  is  repeated.
             If contamination still exists, new reagent  water should be
             prepared.

     9.1.3   Field blanks consisting of EDTA and  reagent water  in a  col-
             lection vial will be shipped to and 'from a  collection site
             with the field samples and will be analyzed for  each set of
             samples from a site.

9.2  After generation of the calibration curve, a calibration standard
     should be analyzed daily to ensure reproducibility.   A  low level
     standard is run at the beginning of the day, and a  mid  level  stan-
     dard is run at the end of the day.   The relative response  factors
     for each analyte should be checked against the average relative
     response factor from the established calibration curve.  Agreement
     should be within 30% of the established RRFs.  If they are  not,  a
     new standard is prepared and analyzed or a new calibration  completed.

9.3  A duplicate sample will  be analyzed for each 10 samples analyzed.
     If there are less than 10 samples in a specific batch, at  least  one
     duplicate will be analyzed.   Sample collection includes collecting
     two vials of blood (Section 7.2).   For duplicate analyses,   the  second
     vial collected will  be used to form the duplicate composite  sample.
     Half of the duplicates will  be analyzed on different analysis days
     than the corresponding original  so that between day precision can
     be determined.

9.4  A spiked blood sample will  be prepared from duplicate samples col-
     lected and will  be analyzed for each 10 samples analyzed.   If there
     are less than 10 samples in a specific batch, at least one  spiked
     sample will  be analyzed.

9.5.  Since labeled analogs are spiked into each blood sample, recovery
     of these compounds can be calculated versus their response' in direct
    -injection by the external  standard method.  Initially, a default
     control  limit of ± 30% from the  average absolute recovery will be
     used.   After a database  of  labeled analog recoveries (R) has been
     established and  upper and lower  control limits  defined based on the
     standard deviation of the recoveries,  these criteria can be used to
     determine whether or not a  sample  needs to be reanalyzed.   Initially,
     if 70% of the compound recoveries  are in  control-; analyses  can con-
     tinue,  otherwise the analysis must  stop until the problem is cor-
     rected.   The upper and lower control  limits are calculated  using
     the formula:

     Upper Control  Limit  (UCL) =  R +  3s
     Lower Control  Limit  (LCL) =  R -  3s

     where R  = average  labeled analog recovery
           s  = standard deviation of  recoveries

                                 19

-------
          It is recommended that the UCL and  LCL  "limits based on a 40-point
          "float" technique.  The UCL and  LCL criteria are  first established
          after 20 analyses.  These control limits should be used to  evaluate
          the results from the next 20 analyses.  New UCL and LCL criteria
          are then calculated for the first 40 data points  and are used  to
          consider the quality of the next 20 QC  analyses.  The first 20 data
          points are dropped and the final 20 data points are included to
          determine the 40-point "float" average.

     9.6  External Quality Controls Checks -  Performance audit standards will
          be prepared by an external quality  control coordinator (QCC).  The
          purpose of this standard is to check the performance of the instru-
          ment and the accuracy of the calibration curve.   The concentration
          of the standard must be in the working  range of the calibration curve.
          Labeled analogs must be included in this solution at the same  concen-
          tration as in the calibration standards.  The performance audit
          standard will be analyzed by direct injection.   Total nanograms of
          analytes detected will be calculated according to the calibration
          established in Section 8 and the results reported to the QCC.   Ac-
          curacy will be determined by the QCC.    If the accuracy is 70 to 130%,
          the performance objectives will be  considered acceptable.    If  the
          limits of accuracy are not reached, another performance standard is
          analyzed.   If the limits of accuracy are still  not reached, a  new
          calibration curve will be generated from freshly prepared standards.

          Table 4 shows the sequential analyses on a. typical day.


10.0 Procedure

     10.1 This analysis procedure covers the analysis of single samples
          (~ 5 mL) and composited (seven 5-mL samples) samples.

          10.1.1  Prior to the analysis of each sample,  whether analyzed
                  singularly or in a composite,  determine the combined weight
                  of the sample and the collection vial.   After transference
                  of the sample(s) to the purge vessel without rinsing,   the
                  empty vial will be reweighed and weight of blood determined
                  and recorded.

          10.1.2  Single Sample Method - Place 40 mL of organic-free  reagent
                  water in the purge vessel.   Add a Teflon-coated stir bar to
                  the vessel.   Transfer the blood sample  (approximately  5 mL;
                  actual  weight will  be determined as described in Section
                  10.1.1) to the purge vessel  in a hood.   Immediately spike
                  the blood/water mixture with 5 uL of the-'labeled analog
                  (3.5 ng/uL)  working solution (Section  6.5).   Close the
                  vessel  and connect the vessel  to purge  and trap  system.
                  Recap the sample vial  for weight determination and subse-
                  quent disposal  by autoclaving.
                                      20

-------
                 Table 4.  Scheme for Typical Day's Analysis
1.   Mass calibrate the GC/MS system (Section 8.2).

2.   Inject a low level mixed standard.  Check response factors (Section 8.7),

3.   Inject an analog-only standard (Section 8.8).

4.   Analyze 45 ml of the reagent water (Sections 9.1.1 and 9.1.2).

5.   Analyze blood samples (Section 10.0).

6.   Analyze a second level analog-only standard (Section 8.8).  Check for
     sensitivity changes in the instrument.

7.   Analyze blood samples (Section 10.0).

8.   Analyze the third level  analog-only standard (Section 8.8).

9.   Analyze a mid-level mixed standard (Section 8.7).
                                      21

-------
     10.1.3  Composite  Sample Method  -  The  seven  blood  samples  are  com-
             posited  in a  hood  and  in an  environment  that  is  as  volatile-
             free as  possible.  This  is done  in close proximity  to  the
             GC/MS  instrument to minimize contamination during  transpor-
             tation of  the  sample to  the  instrument.  To composite  the
             samples, place 10  ml of  volatile-free water in a clean purge
             vessel with a  Teflon stirring  bar.   Immediately  add, with
             no stirring,  each  of the individual  blood  samples  to be
             composited to  the  vessel (approximately  5  ml  each;  actual
             weight will be determined  as described in  Section  10.1.1).
             Immediately spike  the  blood/water mixture  with 5 uL of the
             labeled  analog (3.5 ng/uL) working solution (Section 6.5).
             Quickly  close  the  vessel and connect it  to the purge and
             trap system without delay.   Recap the sample  vials  for
             weight determination and disposal after  autoclaving.

     10.1.4  A constant temperature recirculating water bath  is  equili-
             brated at 50 ± 5°C.  After the blood has been transferred
             to the purge vessel, immediately connect the  recirculating
             bath hoses to the  jacketed purge vessel.    Begin  stirring
             and purge the sample as  directed in  Section 10.2.

10.2 Purge - With the adsorbent trap  at ice water temperature (0°C),
     ensure that the  valve is in the  purge position,  begin the stirring
     of the blood/water solution, start the purge flow,  and purge the
     sample for 20 min.   Check  the purge flow to  ensure "a  35  ± 2 mL/m.i.n
     gas flow is being maintained.

10.3 Desorption and Data Acquisition  - After the  20-min  purge to the trap
     is complete,  turn the valve to the desorb mode, activate the trap
    . temperature controller to  heat the trap, begin the GC temperature
     program, and begin the GC/MS data aquisition.  Leave the trap  at
     220 ± 5°C and the valve in the desorb mode until  the GC  temperature
     program and data acquisition are complete.   This  procedure will
     ensure that the trap and tubing  are cleaned  between samples and
     carryover is  minimized.

10.4 Vessel Change - During the desorption and data acquisition, discon-
     nect the purge vessel  from the system.   Connect a clean vessel  to
     the system and allow the purge flow to pass through the vessel  to
     the transfer  lines and out the vent line to clean the system and
     minimize carryover between samples.  When data acquisition from the
     previous analysis is complete,  cool the trap to 0°C.  Prepare the
     next sample for analysis as described in Section  10.1.2 or Section
     10.1.3.   Dispose of the sample that has been analyzed into a plastic
     gallon jug containing Clorox®  (1:10 dilution),  "immerse the used
     purge'vessel  in a Clorox® bath (1:10  dilution)  for 5 min  and wash
     the vessel  with.detergent,  rinse with water and pesticide quality
     methanol,  then water,  and dry  in an oven at 105°C.
                                 22

-------
11.0 Qualitative Identification

     11.1 Obtain the extracted ion plots (EICPs) for the quantisation and
          confirmation ions for each compound of interest and its  labeled
          analog (Tables 2 and 3).  The following criteria must be met to
          make a qualitative identification.

          11.1.1  The characteristic ions (quantisation and confirmation) of
                  each compound of interest must maximize in the same or
                  within one scan of each other.

          11.1.2  The retention time of the analyte must fall within ± 5 scans
                  of the retention time of the corresponding labeled analog.

          11.1.3  The peak area ratio of the two characteristic ions in the
                  EICPs must fall within ± 30% of the peak area ratio of these
                  ions in the daily calibration standard.

          11.1.4  The quantitation ion must consist of at least two chromato-
                  graphic data points.

          Figure 3 shows the qualitative identification procedure in flowchart
          form.

     11.2 Structural isomers that have very similar mass spectra and less than
          30 s difference in retention time can be explicitly identified only
          if the resolution between authentic isomers in a standard mix is
         , acceptable.   Acceptable resolution is achieved if the baseline to
          valley height between the isomers is less than 25% of the sum of
          the two peak heights.   Otherwise,  structural  isomers are. identified
          as isomeric pairs.

     11.3 After qualitative identification of an analyte,  determine its con-
          centration in the blood according to the procedures in Section 13.0.


12.0 Compounds to be Reported as Qualitatively Identified

     Three of the 30 analytes included in this method are reported on a quali-
     tative basis.   These compounds are 1,1-dichloroethene,  dichloromethane,
     and trichlorofluoromethane.   Preliminary spiked recovery data for these
     compounds were outside the accuracy range established for the method
     validation work (Appendix B,  Table B-l).

     The data for these compounds will  be qualified as  qualitatively detected
     (QD) in the data reporting form (Section 16.0).


13.0 Calculations

     13.1 Labeled analogs are added to  each  sample in the  purge  vessel.   The
          amount of each labeled analog added is 17.5 ng (5  uL  of  a mixed
          solution  containing 3.5 ng/uL of each labeled analog).


                                      23

-------
                   Qualitative
                   Identification of
                   Volatile Organic:
                   in Blood
1
f
Examine EICP for
Analyte and Analog
- Determine RT of
Analog
                       Analyte
                     Quant.  Ion
                    within 5  Scans
                     of Analog ?
Analyte  Not
Detected - Report
MDLas Calculated
from Daily Spika
                      Analyte
                    Quant.  Ion
                    Consist of at
                     Least Two
                    Chrom. Data
                     Points ?
                      Analyte
                    Characteristic
                    Ions Maximize
                     within one
                      Scan?
 Analyte Not
 Detected - Calculate
 MDLas if Detected
 Compound were
 Analyte
                   Characteristic
                  Ion Ratios within
                   £30% of Daily
                      Standard?
                           YES
                 Analyte Detected-
                 Quantitate as per
                 Protocol
Figure 3.    Qualitative  identification procedure.
                             24

-------
13.2 After a peak has been qualitatively identified, quantisation will
     be performed relative to the appropriate labeled analog shown in
     Table 1.  The area of the quantisation ion for a given compound will
     be used for calculations (Table 2).

13.3 The total amount of analyte present in the sample is determined using
     the average relative response factor for that analyte from the up-
     dated calibration curve (Section 8.7).   Use the same ions for quan-
     ti tat ion of the samples and standards as given in Tables 2 and 3.

     13.3.1  For the isotope dilution calculation:
             Total  analyte (ng) = RR x amount 1abgjgd ana1°9 (nc°
                                                 RRF

             where   RR = relative response of analyte to labeled analog,
                            where Rm and subsequent RR values for the
                            analyte in the sample are determined from
                            the equations given in Section 8.4.1.2; and

                   RRF = average relative response factor for the analyte
                           versus the 'labeled analog as determined from
                           the updated calibration curve (Section 8.7).

     13.3.2  For the internal  standard method,  when a corresponding
             labeled analog is not available for isotope dilution,  the
             following formula is used to determine the amount of an
             analyte in a blood sample:

                                  A- x ng IS
             Total  analyte (ng) = — --
                                  A,.- x RRF

             where     AS = area of the quantisation ion for the compound
                            to be measured;
                   ng IS = total  nanograms of the internal  standard added
                            to the sample;
                     A~ = area of the quantisation ion for the
                            internal  standard;  and
                     RRF = average relative response  factor for the
                            compound versus  the internal  standard  as
                            determined!  from  the updated calibration
                            curve (Section 8.7).

             Figure  4 shows  the procedure  for quantitating  an  analyte  in
             flow chart form.   In  the  flow chart calculated levels  of
             analytes  greater  than or  equal  to  three  times  the  historical
             MDL for that analyte  (Section 14.0)  are  labeled as  "positive
             quantifiable."  Analytes  with calculated  levels that are
             less than  three times  the MDL but  greater  than the  historical
             MDL are labeled as  "trace."   If  the  calculated analyte  level
             does not  meet the  qualitative criteria (Section 11.0),  the
             analyte is  labeled "not  detected."
                                25

-------
                         Analyte
                         Quantitation
                         Procedure
                           Response
                           Meets Al
                          Qualitative
                           Criteria?
Report as ND
-Calculate MDL
                     Calculate Analyte
                     Level as per Protocol
                           Analyte
                           Level Less
                           than 3x
Label as Positive
Quantifiable
      Figure 4.   Analyte quantitation procedure.
                                 26

-------
     13.4 After the total nanograms of an analyte  from  the  blood  sample  are
          determined, the weight of the blood  sample as  determined  in  Section
          10 will be used to calculate the concentration of the analyte  in
          the blood in ng/g using the equation below.

               Total ng analyte          ,      ,.,..,,
          g blood - g anticoagulant = ng/g analyte  ™ blood.

     13.5 Corrections must be made for any contaminants  from the  daily water
          blank (Section 9.1.1).  The area response of the  quantisation  ion
          for a detected analyte in the blank will be adjusted by the appro-
          priate labeled analog (Table 1) response.  Total  nanograms of  ana-
          lyte detected in the water blank (45 mL) will  be  determined from
          Section 13.3 calculations.   The total nanograms of analyte per
          milliliter of water will then be determined (total analyte detected/
          45 ml = ng/mL).   This ng/mL analyte concentration will  be used to
          determine total  nanograms of analyte in the 10 mL of water used to
          dilute the blood (total ng = 10 ml. x ng/mL).    Total nanograms  deter-
          mined in the 35-mL blood/10 mL water mixture will be adjusted  by
          subtraction of the amount of analyte attributed to the  10 mL of
          water in the mixture.

     13.6 Field blank values will be reported with the corresponding field
          sample results.


14.0 Precision and Accuracy

     14.1 Precision -  Duplicate samples  analyzed in Section 9.3 can be used
          to calculate a range percent,  P (%) ,  to estimate the precision of
          the method using the equation:
                          x 10o
          where C^ = high concentration of given analyte in sample;
                C2 = low concentration of given analyte in corresponding
                 _     duplicate sample;  and
                 C = average concentration of analyte.

     14.2 Accuracy - The accuracy of the  method can be  assessed from the analy-
          sis  of spikes generated in Section 9.4.   The  measurement for percent
          accuracy,  A (%),  will  be:

          A  (0,-v  _ Spiked sample  value - Unspiked sample value
                                Amount of spike               x 10°

     14.3 A  preliminary evaluation of the precision and accuracy of  this method
          has  been performed from a  4 level  x 4 replicate  study.   The accuracy
          and  precision results  of this study are  given in Appendix  B,  Table  B-l.
          Selected analytes  were chosen for  additional  evaluation in a 3 level  x
          3  replicate study.   The results of this  study are presented in Appen-
          dix  B,  Table B-2.
                                     27

-------
15.0 Estimated Method Detection Limit (MDL)

     The estimated MDLs given in Appendix B, Table B-l, were determined from
     spiked sample data and are based on 35-mL samples.  These MDLs may be
     initially adopted by the analyst but are subject to change as the data-
     base of spiked sample information increases.  Initially, the analyst must
     demonstrate the ability to detect these levels in whole, blood by perform-
     ing duplicate analyses of blood samples spiked at the estimated MDLs for
     each compound.  The analytes must meet the qualitative criteria set forth
     in Section 11.0.  If 70% of the analytes are not detected in both of the
     spiked samples, the analyst must increase the GC/MS sensitivity (e.g., by
     increasing the EM voltage of the MS) until the MDLs given in Appendix 8,
     Table B-l, are achieved.  If the estimated MDLs in Table B-l are not
     achieved, the EPA Project Officer must be notified of the higher operat-
     ing MDLs.  If lower MDLs are achieved than those specified in this method,
     then new MDLs are established by the method given in Section 15.1.1 or
     by using a scientifically acceptable statistical method (Long 1985,
     Keith 1983).

     There are several methods currently in use to determine estimated method
     detection limits.  Among these are determinations from noise levels, from
     regression analyses, and from spike sample data.  The method presented
     in Section 15.1.1 determines MDLs from the daily sample spikes.
     15.1 Esti-mated method detection limits can be calculated In'situations   ."
          where (1) no response is observed for an analyte; (2) a response is
          observed but ion ratios are incorrect; and (3) where a response is
          quantitated as a trace value (i.e., where the value.is between the
          MDL and 3 x MDL).  Alternatively, the MDLs given in Appendix B,
          Table B-l may be used provided that those MDLs have been achieved
          and demonstrated-

          15.1.1  For samples in which the quantitation ion for a given analyte
                  is not observed, calculate the estimated MDL by extrapolat-
                  ing from the analyte's quantitation ion peak area in the
                  unfortified or fortified sample (where the analyte is first
                  observed) from that day's analyses to the peak area where
                  peak area approaches peak height.  This corresponds to the
                  response region where the minimum qualitative criterion
                  (Section 11.0) can be met.  The estimated MDL is then cal-
                  culated as below:

                                  Peak Area
                  Estimated MDL = -=—r--r—      x Analyte Level (in unfortified
                                   Peak Areass                    Qr fortified
                                                                  sample)

                  where Peak Areap,_p,, is the approximate peak area where the

                  analyte's quantitation peak area approaches the peak height;
                  and Peak Area^ is the quantitation ion peak area in the
                               OO                  '                 ;
                  daily unfortified or fortified sample.
                                      28

-------
                  For example, if the analyte quantiation  ion  for  the  100-ppt
                  spike level has a peak area of 5,000 area  units  and  the
                  point where the peak area approaches the peak  height is
                  approximately 1,000 units (as approximated from  the  cali-
                  bration data), the estimated MDL  is calculated as  shown
                  below:
                  Estimated MDL =  ^    x 100 ppt = 20 ppt
                  Figure 5 shows the MDL determination process in flowchart
                  form.

          15.1.2  For samples in which a response at the retention time of a
                  given analyte is observed, but the ion ratios differ from
                  these ratios observed in the daily standard by more than
                  30%, the estimated method detection limit is calculated as
                  if the peak observed were the actual analyte using the equa-
                  tions in Section 13.0.   The values are qualified as not de-
                  tected, ND, and the concentration is reported in parentheses.

          15.1.3  If a response for a given analyte is qualified as a trace,
                  TR, value (between the estimated MDL and three times the
                  estimated MDL),  the analyst will report the estimated MDL
                  that was used to classify the value as a trace.   This value
                  will come from the analysts'  previous determinations of the
                  estimated MDL for that analyte or from Table B-l.


16.0 Reporting and Documentation

     All  data should be reported on a composite or single sample basis,  as
     required, using a data report format shown in -Figure 6.   The analyst, is
     required to maintain all raw data,  calculations,  and control  charts  in a
     format as to allow a complete external  data review.
                                      29

-------
              MDL
          Determination
      Manual Examination of
      Analyte Ion Plot for
      Unfortified Sample
                                         Go to 100 ppt Spike  Level
                                         or Next Highest Spike
                                         Level (If no higher spike
                                         level, MDL cannot be
                                         determined)
    Primary and Secondary Ions
   Coelure with Correct Ratio
             &RT?
       Analyte Not
       Detected
    Primary and Secondary ions
Contain at  Least Two Data Points?
       Calculate Ratio of*
       (Peak Area where
       Analyte Peak Area =
       Peak Height)/(Peak
       Area in Sample)
(Approximate Peak
Area where Peak
Area = Peak Height
for Analyte from
Calibration Data
      Multiply Ratio Times
      Amount Determined In
      Unfortified or Fortified
      Sample to Determine
      MDL
    Figure 5.  MDL  determination procedure.
                            30

-------
U.S. Environmental Protection Agency
Office of Toxic Substances
Exposure Evaluation Division
Comoosite Samnla IO»
Laboratory*
Batcn Numoar*
Analvsls Data-
Ravlaumd hy , ,

Analytes
Benzene
Eromodichlorometnane
Sromoform
Carbon tetracnicride
Chlorobenzene
Chloroform
Oibromocnloromamane
1 .2-Olcnlorobenzen9
1 ,3-Olchlorooenzene
1 ,4-Olcnlorobenzene
1.1-Oichloroethane
1 ,2-Olchtaroetfiane
1,1-Oichloroetnene
irans-l .2-Olcnioroetnene
Oichloromethane
1 ,2-Oichloroorooane
eia-l ,3-Oichloroorooene
lrans-1 ,3-OicfiloroproDene
Ethylbenzeno
Styrene
1.1.2.2-1 atracnloroethane
Tetrachloroetriene
Toluene
1,1.1-Trichloroetnane'
1,1.2-Trichloroemane
Tricnioroetnane
Trichlorofluorometnana
1.2-Xylene
1 .3-Xylene/1 .4-Xylene (3)
Data
Qualifier (1 )















.













MOL
(P9S)





























National Stood Network
Analysis flegort Form

Concentration
(P9g)(2)





























Sample Weignts
Individual Sample I0# Weignt (g)






Total CompositB Sample Weight (g)
Labeled Analog
of internal Std
Senzene-d,
E!ramoform-!:!C
Carbon tetracnloride-13C
Chlorobenzene-
-------
References

ACS.  1979.  American Chemical Society.  Safety  in academic chemistry  labora-
tories, 3rd ed.  American Chemical Society Publication, Committee  on Safety.

Keith LH, Crummett W, Deegan J Jr, Libby RA, Taylor JK, Wentler  G.  1983.
Analytical Chemistry 55:2210-2218.

Long GS, Winefordner JD.  1983.  Analytical Chemistry 55:712A-724A.

NIH.  1984 (March).  Natl. Inst. of Health.  Biosafety  in microbiological  and
biomedical laboratories.  Washington, DC:  NIH,  U.S. Dept. of  Health and  Human
Services, Public Health Service,  Center for Disease Control.  HHS Pub. (CDC)
84-8395.

NIOSH.  1977 (August).  Natl. Inst. Occupational Safety and Health.  Carcinogens
working with carcinogens.  Washington, DC:  NIOSH, U.S. Dept.  Health,  Education,
and Welfare, Public Health Service, Center for Disease  Control.  DHEW  Pub.
NIOSH 77-206.

OSHA.  1976 (January).  Occupational Safety and  Health  Admin.  OSHA safety
and health standards, general industry.  Washington, DC:  OSHA,  U.S. Dept.
Labor.  OSHA Pub. 2206.  (29 CFR 1910).

USEPA.  1985 (January)r  U. S." EnvTrdhmentaT'Pr'btection  Agency.   Volatile  organic
compounds by isotope 'dilution GC/MS.  Method 1624 Revision B.  Cincinnati,
OH:  USEPA.
                                      32

-------
                APPENDIX A
SAFETY PROCEDURES FOR HANDLING BLOOD IN THE
          NATIONAL BLOOD NETWORK
                    A-l

-------
               SAFETY PROCEDURES FOR HANDLING BLOOD IN THE NBN


          This document is intended to be a stand-alone guideline for labora-
tory personnel working on the National Blood Network (NBN).  The following
handling procedures are intended to provide maximum safety to those personnel
handling blood for the NBN.  These procedures are based on procedures given
in "Biosafety in Microbiological and Biomedical Laboratories," HHS Publica-
tion No. (CDC) 84-8395,1 and in the Agent Summary Statement for the HTLV-III
virus.2  General handling procedures, receipt, log-in, storage, analysis, and
disposal of the blood samples are discussed in the following sections.


GENERAL HANDLING PROCEDURES (Biosafety Level 2)

          The following practices are particularly pertinent to the handling
of blood anticipated in the NBN.

          *    Use of syringes, needles, and other sharp instruments should
               be avoided if possible.

          *    Gloves should be worn by all personnel engaged in activities
               that may involve skin contact with the blood.

          *    Generation of aerosols, splashes, and spills of blood should
               be avoided.

          *    All laboratory'glassware, equipment, disposable materials, and
               wastes must be decontaminated before washing,  discarding, etc.

          *    If laboratory clothing becomes contaminated with blood, it
               must be decontaminated before being laundered or discarded.

          *    Work surfaces should be decontamianted at the end of each day
               or when overtly contaminated. .


RECEIPT OF BLOOD SAMPLES

          Blood samples provided by the blood centers will be received by a
designated sample receiver in the specially designed sample kits.   The kits/
sampling containers will consist of vial racks which will hold the blood sam-
ple vials and bagged ice which will keep the samples cold.  These will be sur-
rounded by a Styrofoam box which is further surrounded by a metal  shipping
container.


LOG-IN AND SAMPLE HANDLING

          Upon receipt of a shipping container, the designated receiver of
the samples should complete the following activities:
                                     A-2

-------
           1.  Place the sample container in a Class I faiosafety  hood  (or
 average face velocity of 100 linear feet per minute).

           2.  Put on disposable gloves and laboratory coat.

           3.  Open the shipping container.

           4.  Check for sample leakage or breakage of the sample vials.

                a.  If leakage or broken vials are discovered, unbroken vials
 should^be retrieved and the exterior of each vial decontaminated with a chem-
 ical disinfectant.  Wiping the surface of or immersing the vial  racks in a
 1:10 solution of household bleach for 5 min will decontaminate the surfaces
 in question.  The remainder of the interior of the sampling container, including
 broken vials, racks, documents, etc.,  will  also be decontaminated before being
 discarded.

                b.  If no leakage or breakage is discovered, the receiver
 should continue with sample storage.


 SAMPLE STORAGE

_.	After log-in,  samples should be stored according to the designated
 storage"condvt.To'ns"'for each sample type.   Those conditions are:

           1.  Semivolatiles:   -10°C

           2.  Volatiles:   Seal  in wide-mouth  jar with activated charcoal,
                           then  store  at 4°C

           3.  Trace elements:   -10-°C

           Each sample  will  be  stored,  under the conditions given, in the
 sample prep  laboratory for that particular  sample type.   After storage of  the
 samples,  the receiver  should  remove his/her disposable gloves and disinfect
 and discard  them according to  the disposal  procedures  described  subsequently
 in this document.


 ANALYSIS

           1.  General  considerations.

           *     Personnel working  with  the blood  should follow BSL 2  safety
                procedures  described in  the  general  handling" procedures sec-
                tion  of this document.

           *     All  glassware that  has  come  into  contact with  the  blood sam-
                ples  should  be  immersed  in a 1:10  solution of  household bleach
                for 5 min before being washed or  discarded.
                                     A-3

-------
          *    Compositing of individual blood samples into the composite sam-
               ple for analysis should be conducted in a Class A hood follow-
               ing BSL 2 procedures.

          2.  Semivolatile analysis.

          *    Since the blood is pH-adjusted to very low and then very high
               pHs in this method, no biological activity is expected in the
               samples after pH adjustment.

          3.  Volatile analysis.

          *    During the purge step, the purge gas flow from the vent port
               of the two-way six-port valve should be vented to a Class A
               hood or allowed to bubble through a 3% H202 solution before
               being vented into the room.

          4.  Trace elements analysis.

          *    Because of the acid digestion step in this analysis, no bio-
               logical activity is expected in these samples after acid
               digestion.


DISPOSAL

          The procedures listed below will be followed in matters regarding
disposal.
               Disposable pipettes, gloves, sample vials, etc., will be im-
               mersed in a 1:10 solution of commercial bleach (e.g., Clorox®)
               and placed in an autoclavable container, autoclaved, and dis-
               posed of as solid waste according to MRI safety procedures.

               Blood from the volatile analysis will be mixed in equal volume
               with a 1:10 solution of household bleach, the solution and con-
               tainer autoclaved, and then disposed of as nonflammable waste
               according to MRI safety procedures.

               Blood from the semivolatile analysis will be disposed of as
               alkaline waste according to MRI safety procedures.

               Blood from the trace elements analysis will be disposed of as
               acid waste according to MRI safety procedures.

               Laboratory coats contaminated with blood must be soaked in
               a 1:10 solution of household bleach before being laundered
               or discarded.
                                     A-4

-------
               Solution of household bleach (dilute 1:10)  used  for  the  decon-
               tamination of glassware will be disposed of by draining  into
               an autoclavable container, autoclaving the  container and solu-
               tion, and disposing of them according to MRI safety  procedures.

               Organic solvents will be disposed of according to MRI safety
               procedures.
REFERENCES
1.  CDC-NIH.  1984.  U.S. Department of Health and Human Services.  Public
    Health Service.  Centers for Disease Control and National Institute of
    Health.   Biosafety in microbiological and biomedical laboratories.  HHS
    Publication No. (CDC) 84-8395.   U.S. Government Printing Office.

2.  CDC.   1986.   Centers for Disease Control.  Human T-lymphotropic virus  '
    type Ill/lymphadenopathy associated virus:  agent summary statement.
    Journal  of the American Medical Association 256(14):1857, 1861, 1868,
    1873.
                                    A-5

-------

-------
        APPENDIX B





PRECISION AND ACCURACY DATA
           B-l

-------
                 Table  8-1.   Estimated MDL and Method PerfOi-mance (Accuracy and Precision)
                                  From Four Level  by Four Replicate Study3
     Compounds
Estimated
   MOL
 (pg/mL)
Percent"
accuracy
                                                                                      Precision
	RSD (%) at spike level	
100 pg/mL   500 pg/mL   1000 pg/mL
Benzene                          100
Sronad'ichloromethane             100
Brofflofona                        100
Carbon tatrachloridn              50
Chlorobenzene                    100
Chloroform           .             50
01broraochloromethane              50
1,2-Oichlorobenzene              100
1,3-Oichlorobenzane              100
1,4-Oichlorobenzena              100
1,1-Qfchloroethane                SO
1,2-OichloroeEhane                50
trans-1.2-0ichloroethana         100
1,2-Qichloropropane     .         100
£is-l,3-0ichloropropene   *       200
trans-l,3-Qichloropropene'       300
Sthylbenzene                     100
Styrene                           50
1,1,2,2-Tetrachloroethana       100
!,I,i-Trichloroethane            100
1,1,2-Trichtoroathane            100
Trichloroethene                  100
l,2;Xylene                       100
l,3-Xylene/l,4-Xylene            100
1,1-Oicnloroethene9               50'
Oichloromethana
rrichlorofluoromethane9           25
                           88
                          - 90
                          105
                           96
                           91
                           91
                          120
                          109
                          107
                          116
                          103
                          108
                          107
                           81
                         .  79
                           54
                           79
                           72
                          134
                           71
                           82
                           95
                           72
                           96
                           29

                           10
                        14
                        26
                        35
                         9
                        10
                        13
                        33
                        12
                         9
                        17
                        36
                        50
                         6

                        41e
                        NO8
                        NO
                        15
                        87
                        12
                        15
                        12'
                         9
                        16
                        13
                        SO

                        58
                13
                31
                10
                14
                 6
                13
                36
                22
                 4
                14
                26
                22
                10
                17
                20
                30
                13
                 8
                31
                21
                 8
                17
                10
                23
                23

                47
 7
44
10
 0
 3
 5
37
13
16
12
13
 9
 7
19
 7
24
 5
27
18
16
10
22
 3
 6
20

54
^Analyses performed on a Finnigan OWA GC/MS system.
fiAs .measured by the slope of the regression equation.
 The upper 95% confidence limit on the RSD is approximately  2  x  RSD  (4 replicates  at each spike level,
 .four replicate nonfortified blood samples were  the  fourth level).
 Labeled analog not available; quantitated relative  to  internal  standard  method.
 -NO = not detected (not included in calculation  of accuracy).
 C_is_-l,3-0ichloropropane spiked at 39 pg/mL, 195 pg/mL, and  390  pg/mL.                            ,
 trans-1.3-0ich1oropropane spiked at 61 pg/mL, 305 pg/mL, and  610  pg/mL.
3These compounds, calculated by the internal standard method,  do not meet the  accuracy critarian of
^50-150%.  Report as qualitatively present.
 High background levels of dich'loromethane precluded determinations  of accuracy, and .precision  from
 spiked blood report as qualitatively present.
                                                      .  B-2

-------
CO

to
                       Table B-2.  Estimated MDL and Method  Performance (Accuracy and Precision)
                                      From Three Level by  Three Replicate Study3


Compound
Bromodichloromethane .
Dibromochloromethane
Styrene
Tetrachl oroethene
Toluene
1,1,1-Trichloroethane
.Analyses performed on
Estimated
MDL
(pg/mL)
100
50
50
100
100
100
a Finnigan OWA GC/MS
. Fortification levels
Percent
accuracy
82
78
102
96
81
99
system.
Spike 1
(pg/mL)
200
200
1,000
1,500
20,000
2,500

Spike 2
(pg/mL)
500
500
2,500
3,750
50,000
•6,250

Precision0
RSD (%) at
Level 1
29
17
14
5
17
12

spike level
Level 2
33
61
10
11
42
16

       • *~  L.^VA w vi i wv* §^jf  fc-nv. «a i up^ ui  1*1 ic i cy i troo I ut
       The upper 95% confidence limit on the RSD is approximately 2 x  RSD  (triplicates at each spike  level,
      dtriplicate nonfortified blood samples were the third level.)
       Labeled analog  not available; quantitated relative to internal  standard method.

-------

-------
                                    TECHNICAL REPORT DATA
                            (Please read Inaructions on the reverse before completing)
 , REPORT NO.
  EPA-560/5-87-008
                                                            3. RECIPIENT'S ACCESSIONING.
4. TITLE AND SUBTITLE
  Volatile Organic Compounds  in Whole Blood - Determina-
  tion by Heated Dynamic  Headspace Purge and Trap  Isotope
  Dilution 6C/MS
               5. REPORT DATE
                July 1987
               6. PERFORMING ORGANIZATION CODE     |
                8822-A01
7. AUTHOR(S)
  Paul  H. Cramer, Kathy  E.  Boggess, John
                                                            8. PERFORMING ORGANIZATION REPORT NO.
Hosenfeld
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Midwest Research  Institute
  425 Volker Boulevard
  Kansas City, Missouri   64110
               10. PROGRAM ELEMENT NO.
                 Work  Assignment No. 22
               11. CONTRACT/GRANT NO.

                 68-02-4252
12. SPONSORING AGENCY NAME AND ADDRESS
  Field Studies Branch  (TS-798), Office of Toxic:  Substance
  U.S.  Environmental  Protection Agency
  401  M Street, S.W.
  Washington, DC   20460                              '
                                                            13. TYPE OF REPORT,AND PERIOD COVERED
               13. TYPE O.F REPORT AND PERIQI
               s  Special  (11/84-8/87)
               14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
  The EPA Work  Assignment  Manager is Janet C. Remmers  (202)  382-3582.
  The EPA Project Officer  is  Joseph J. Breen  (202)  382-3569.
16. ABSTRACT           ••                                                '
       The method described here was developed for  the determination of a  selected list
  of 30 organic compounds  in whole human blood.   The  method is a heated dynamic  head-
  space purge and trap  gas chromatographic method using mass spectrometry  in  the lim-
  ited mass scan mode.   A  whole blood sample is  diluted with organic free  water  and the
  mixture fortified  with isotopically labeled compounds.   The mixture is subsequently
  heated to 50°C while  stirred and the volatile  components are purged from the mixture
  and collected on an adsorbent trap.  The volatile compounds are thermally desorbed
  from the trap onto a  wide-bore fused silica capillary column.  Detection of the com-
  pounds is accomplished using mass spectrometry in the.limited mass scan  mode where
  only selected ions are scanned.
       Quantitation  of.  the analytes is accomplished by using the specific  ion responses
  from standards of  the analytes and their corresponding labeled analogs or internal
  standards.  The response of the labeled compounds are used to correct the variability
  of the analytical  technique through use of an  isotope dilution calculation  procedure
  or an internal standard  calculation procedure.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.lOENT!F!ERS/OPEN'ENDEO TERMS
                             c. COSATI Fieid/GrouD
  Volatile organics
  Determination
  Whole blood
  Isotope dilution
  GC/MS
  Purge and trap
 3. DISTRIBUTION STATEMENT

  Unlimited
  19. SECURITY CLASS /This Report)
    Unclassified
21. NO. OF PAGES
      44
                                               20. SECURITY CLASS (Thispage)
                                                  Unclassified
                                                                          22. PRICE
EPA Form 2220-1 (R«». 4_77)   PREVIOUS EDITION is OBSOV.ETE

-------

-------