United States
         Environmental Protection
         Agency
Office of Water
Engineering and Analysis Division (4303)
Washington. D.C. 20460
EPA-821-B-94-001
February 1995
oEPA  Analytical Methods for the
         Determination of Pollutants
         in Pharmaceutical Manufacturing
         Industry Wastewater

                                         . Printed on Recycled paper

-------
S4EPA Analytical Methods for the
       Determination of Pollutants
       in Pharmaceutical Manufacturing
       Industry Wastewater

-------
                                      Acknowledgments

This methods compendium was prepared under the direction of William A. Telliard of the Engineering and
Analysis Division within EPA's Office of Water.  This document was prepared under EPA Contract No. 68-
                  C3-0337 by the Environmental Programs Division of DynCorp.

                                          Disclaimer

 This methods compendium has been reviewed by the Engineering and Analysis Division, U.S. Environ-
mental Protection Agency, and approved for publication. Mention of trade names or commercial products
                  does not constitute and endorsement or recommendation for use.

-------
                                                  Table of Contents
Introduction	v

Method 1665
    Semi-Volatile Organic Compounds Specific to the Pharmaceutical
    Manufacturing Industry by Isotope Dilution QC/MS	1

Method 1666
    Volatile Organic Compounds Specific to the Pharmaceutical Manufacturing
    Industry by Isotope Dilution GC/MS	37

Method 1667
    Formaldehyde, Isobutyraldehyde, and Furfural by Derivatization
     and High-Pressure Liquid Chromatography	79

Method 1671
    Volatile Organic Compounds Specific to the Pharmaceutical Manufacturing
    Industry by GC/FID	97

Method 1673
    Polyethylene glycol)-600 by Derivatization and High-Pressure
    Liquid Chromatography	115

-------
                                                                     Introduction
The U.S. Environmental Protection Agency (EPA or the Agency) is proposing effluent limitations guide-
lines and standards for promulgation at 40 CFR Part 439 for the Pharmaceutical Manufacturing Industrial
category to control the discharge of pollutants into surface waters of the United States. This compendium
of test procedures (methods) supports the proposal. The purpose of publishing the compendium is to pro-
vide a single source of methods that are unique to the proposed rule. These methods or the referenced
methods must be used for filing permit applications and for compliance monitoring under the National Pol-
lutant Discharge Elimination System (NPDES) program.
Two of the methods included in this compendium are based on isotope dilution methods: "Method 1624:
Volatile Organic Compounds by Isotope Dilution GC/MS" and "Method 1625: Semivolatile Organic Com-
pounds by Isotope Dilution GC/MS.   Two additional methods included in this compendium are based
upon draft Method 8315, "Formaldehyde by High-Pressure Liquid Chromatography" (January 1990).
This compendium includes only those methods that are unique to the Pharmaceutical Manufacturing
Industry rulemaking. Other methods allowed under the proposed rule have been promulgated at 40 CFR
Part 136.
Questions concerning the methods in this compendium should be directed to:
                                        W.A. Telliard
                                         U.S. EPA
                              Engineering and Analysis Division
                               Office of Science and Technology
                                     401 M Street, SW
                                   Washington, DC 20460
1.    United States Environmental Protection Agency, Office of Water Regulations and Standards, Industrial
     Technology Division, Office of Water, June 1989.

-------
            Method 1665

Semi-Volatile Organic Compounds
   Specific to the Pharmaceutical
       Manufacturing Industry by
         Isotope Dilution GC/MS
                      January 1995

-------
                                 Method  1665

              Semi-Volatile Organic Compounds Specific to the

                   Pharmaceutical Manufacturing Industry by

                              Isotope Dilution GC/MS

1.   Scope and Application

1.1   This method is designed to meet the survey and monitoring requirements of the U.S. EPA Engineer-
     ing and Analysis Division (EAD). The method is used to determine the semivolatile  organic pollut-
     ants specific to the pharmaceutical manufacturing industry (PMI) that are amenable to extraction and
     analysis by capillary column gas chromatography-mass spectrometry (GC/MS).
1.2   The chemical compounds listed in Table 1 may be determined in waters, soils, and municipal sludges
     by this method or the method referenced.
1.3   The detection limits of the method are usually dependent on the level of interferences rather than
     instrumental limitations. The limits provided in Table 2 are the minimum amounts that can be reli-
     ably quantified with no interferences present.
1.4   The GC/MS portions of the method are for use only by analysts experienced with GC/MS or under
     the close supervision of such qualified persons. Laboratories unfamiliar with analysis of environ-
     mental samples by GC/MS should run the performance tests in Reference 1 before beginning.
1.5   This method is "performance-based." The analyst is permitted to modify the method to overcome
     interferences or to lower the cost of measurements, provided that all performance criteria in this
     method are met The requirements for establishing method equivalency are given in Section 9.1.2.

2.   Summary of Method

2.1   The percent solids content of a sample is determined. Stable, isotopically labeled analogs of the com-
     pounds of interest are added to the sample. If the solids content is less than 1%, a 1-L sample is
     extracted at pH 12-13 with methylene chloride using continuous extraction techniques. If the solids
     content is 30% or less, the sample is diluted to 1% solids with reagent water, homogenized ultrasoni-
     cally, and extracted at pH 12-13 with methylene chloride using continuous extraction techniques. If
     the solids content is greater than 30%, the sample is extracted using ultrasonic techniques. Each
     extract is dried over sodium sulfate, concentrated to a volume of S mL, cleaned up using gel-perme-
     ation chromatography (GPC), if necessary, and concentrated. Extracts are concentrated to 1 mL if
     GPC is not performed, and to 0.5 mL if GPC is performed. An internal standard is added to the
     extract, and a 1-uL aliquot of the extract is injected into the gas chromatograph (GC).  The com-
     pounds are separated by GC and detected by a mass spectrometer (MS).  The labeled compounds
     serve to correct the variability of the analytical technique.
2.2   Identification of a pollutant (qualitative analysis) is performed as follows: The GC/MS system is cal-
     ibrated and the mass spectrum and retention time for each standard are stored in a user-created
     library.  A compound is identified when its retention time and mass spectrum agree with the library
     retention time and spectrum.

2.3   Quantitative analysis is performed in one of two ways by GC/MS using extracted-ion current profile
     (EICP) areas: (1) The GC/MS system is calibrated and the analyte concentration is determined using

-------
Method 1665
            an isotope dilution technique; or (2) For analytes lacking a labeled analog, the GC/MS system is cal-
            ibrated and the compound concentration is determined using an internal standard technique.
       2.4  The quality of the analysis is assured through reproducible calibration and testing of the extraction
            and GC/MS systems.

       3.   Definitions

       There are no method-specific definitions to be noted for this document.

       4.   Interferences

       4.1   Solvents, reagents, glassware, and other sample processing hardware may yield artifacts and/or ele-
            vated baselines causing misinterpretation of chromatograms and spectra. All materials used in the
            analysis shall be demonstrated to be free from interferences under the conditions of analysis by run-
            ning method blanks initially and with each sample batch (samples started through the extraction pro-
            cess on a given 8-hour shift, to a maximum of 20). Specific selection of reagents and purification of
            solvents by distillation in all-glass systems may be required.  Glassware and, where possible, other
            equipment are cleaned by solvent rinse and baking at 450°C for a minimum of 1 hour.
       4.2   Interferences co-extracted from samples will vary considerably from source to source, depending on
            the diversity of the site being sampled.

       5.   Safety

       The toxicity or carcinogenicity of each compound or reagent used in this method has not been precisely
       determined; however, each chemical compound should be treated as a potential health hazard. Exposure to
       these compounds should be reduced to the lowest possible level. The laboratory is responsible for main-
       taining a current awareness file of OSHA regulations regarding the safe handling of the chemicals speci-
       fied in this method.  A reference file of material safety data sheets should also be made available to all
       personnel involved in these analyses. Additional information on laboratory safety can be found in Refer-
       ences 3-5.

       6.   Equipment and Supplies

       6.1   Sampling equipment for discrete or composite sampling.
            6.1.1   Sample bottles and caps.
                   6.1.1.1    Liquid samples (waters, sludges, and similar materials that contain less than five
                            percent solids)—Sample bottle, amber glass, 1.1-L minimum, with screw-cap.
                   6.1.1.2   Solid samples (soils, sediments, sludges, filter cake, compost, and similar materials
                            that contain more than 5% solids)—Sample bottle, wide mouth, amber glass, 500-
                            mL minimum.

                   6.1.1.3   If amber bottles are not available, samples shall be protected from light.
                   6.1.1.4   Bottle caps—Threaded to fit sample bottles. Caps shall be lined with polytetraflu-
                            oroethylene (FTFE).

-------
                                                                                             Method 1665
             6.1.1.5   Cleaning.
                     6.1.1.5.1    Bottles are washed with detergent and water, rinsed with reagent water,
                                then solvent rinsed or baked at 450°C for a minimum of 1 hour before
                                use.
                     6.1.1.5.2    Cap liners are washed with detergent and water, rinsed with reagent
                                water (see Section 7.5.1) and then solvent, and then baked for at least
                                1 hour at approximately 200°C.
     6.1.2    Compositing equipment—Automatic or manual compositing system incorporating glass
             containers cleaned per bottle cleaning procedure above. Sample containers are kept at 0^40C
             during sampling.  Only glass or PTFE tubing shall be used. If the sampler uses a peristaltic
             pump, a minimum length of compressible silicone rubber tubing may be used only in the
             pump. Before use, the tubing shall be thoroughly rinsed with methanol, followed by
             repeated rinsings with reagent water (Section 7.5.1) to minimize sample contamination.  An
             integrating flow meter is used to collect proportional composite samples.
6.2  Equipment for determining percent moisture.
     6.2.1    Oven, capable of maintaining a temperature of 110°C (±5°C).
     6.2.2    Desiccator.
6.3  Sonic disrupter—375-watt with pulsing capability and %" disrupter hom (Ultrasonics, Inc, Model
     375C, or equivalent).
6.4  Extraction apparatus.
     6.4.1    Continuous liquid-liquid extractor—PTFE or glass connecting joints and stopcocks without
             lubrication, 1.5- to 2-L capacity (Hershberg-Wolf Extractor, Ace Glass 6841-10, or equiva-
             lent).
     6.4.2    Beakers.
             6.4.2.1   1.5- to 2-L borosilicate glass beakers calibrated to 1 L.
             6.4.2.2   400- to 500-mL borosilicate glass beakers.
             6.4.2.3   Spatulas—Stainless steel.
     6.4.3    Filtration apparatus.
             6.4.3.1   Glass funnel—125- to  250-mL.
             6.4.3.2   Filter paper for above (Whatman 41, or equivalent).
6.5  Drying column—15- to 20-mm inside diameter (i.d.) Pyrex chromatographic column equipped with
     coarse-glass frit or glass-wool plug.
6.6  Concentration apparatus.
     6.6.1    Concentrator tube—Kuderna-Danish (K-D) 10-mL, graduated (Kontes K-570050-1025, or
             equivalent) with calibration verified. Ground-glass stopper (size 19/22 joint) is used to pre-
             vent evaporation of extracts.
     6.6.2    Evaporation flask—K-D 500-mL (Kontes K-570001-0500, or equivalent), attached to con-
             centrator tube with springs (Kontes K-662750-0012, or equivalent).
     6.6.3    Snyder column—K-D three-ball macro (Kontes K-503000-0232, or equivalent).

-------
Method 1665
            6.6.4   Snyder column—K-D two-ball micro (Kontes K-469002-0219, or equivalent).
            6.6.5   Boiling chips—Approximately 10/40 mesh, extracted with methylene chloride and baked at
                   450°C for a minimum of 1 hour.
            6.6.6   Nitrogen evaporation device—Equipped with a water bath that can be maintained at 35-
                   40°C.  The N-Evap by Organomation Associates, Inc., South Berlin, MA (or equivalent) is
                   suitable.
       6.7  Water bath—Heated, with concentric ring cover, capable of temperature control (±2 *C), installed in a
            fume hood.
       6.8  Sample vials—Amber glass, 2- to 5-mL with PTFE-lined screw-cap.
       6.9  Balances.
            6.9.1   Analytical—Capable of weighing 0.1 mg.
            6.9.2   Top loading—Capable of weighing 10 mg.
       6.10 Automated gel-permeation chromatograph (Analytical Biochemical Labs, Inc., Columbia, MO,
            Model GPC Autoprep 1002, or equivalent).
            6.10.1  Column—600 to 700 mm long x 25 mm i.d., packed with 70 g of SX-3 Bio-beads (Bio-Rad
                   Laboratories,  Richmond, CA, or equivalent).
            6.10.2  UV detectors—254-nanometers (nm), preparative or semi-prep flow cell.
                   6.10.2.1  Schimadzu, 5-mm path length.
                   6.102.2  Beckman-Altex 152W, 8-uL micro-prep flow cell, 2-mm path length.
                   6.10.2.3  Pharmacia UV-1,3-mm flow cell.
                   6.10.2.4  LDC Milton-Roy UV-3, monitor #1203.
       6.11  Gas chromatograph—Shall have splitless or on-column injection port for capillary column, tempera-
            ture program with 30°C hold, and shall meet all of the performance specifications in Section 13.
            6.11.1  Column—30 m (±5 m) long x 0.32 mm (±0.02 mm) i.d., with a 1-jjm coating of 5% phenyl,
                   94% methyl, 1% vinyl silicone bonded-phase on a fused-silica capillary column (Restek Rtx-
                   5,  or equivalent). Temperature program: 5 minutes at 30°C; 30-280°C at 8°C per minute;
                   isothermal at 280°C until N,N-dimethylaniline elutes. Gas Velocity: 30 cm/sec (±5 cm/sec)
                   at  30°C.

       6.12 Mass spectrometer—70 eV electron impact ionization, shall repetitively scan from 35 to 450 dalton
            in 0.95-1.00 second, and shall produce a unit resolution (valleys between mass over charge [m/z]
            441-442 less than 10% of the height of the 441 peak), background corrected mass spectrum from
            50 ng decafluorotriphenylphosphine (DFTPP) introduced through the GC inlet. The spectrum shall
            meet the mass-intensity criteria in Table 3 (Reference 6).  The mass spectrometer shall be interfaced
            to the GC such that the end of the capillary column terminates within 1 cm of the ion source but does
            not intercept the electron or ion beams. All portions of the column that connect the GC to the ion
            source shall remain at or above the column temperature during analysis to preclude condensation of
            less volatile compounds.
       6.13 Data system—Shall collect and record MS data, store mass-intensity data in spectral libraries, pro-
            cess GC/MS data, generate reports, and shall compute and record response factors.
            6.13.1  Data acquisition—Mass spectra shall be collected continuously throughout the analysis and
                   stored on a mass-storage device.

-------
                                                                                             Method 1665
     6.13.2  Mass spectral libraries—User-created libraries containing mass spectra obtained from analy-
            sis of authentic standards shall be prepared.
     6.13.3  Data processing—The data system shall be used to search, locate, identify, and quantify the
            compounds of interest in each GC/MS analysis.  Software routines shall be employed to
            compute retention times and peak areas. Displays of spectra, mass chromatograms, and
            library comparisons are required to verify results.
     6.13.4  Response factors and multipoint calibrations—The data system shall be used to record and
            maintain lists of response factors (response ratios for isotope dilution)and multi-point cali-
            bration curves (Section 10).  Computations of relative standard deviation (coefficient of vari-
            ation) are used for testing calibration linearity. Statistics on initial (Section 9.2) and on-going
            (Section 13.7) performance shall be computed and maintained.

7.   Reagents and Standards

7.1   Reagents for adjusting sample pH.
     7.1.1   Sodium hydroxide—Reagent grade, 6 normal (N) in reagent water.
     7.1.2   Sulfuric acid—Reagent grade, 6 N in reagent water.
7.2   Sodium sulfate—Reagent grade, powdered anhydrous, rinsed with methylene chloride (20 mL/g),
     baked at 450°C for a minimum of 1 hour, cooled in a desiccator, and stored in a pre-cleaned glass
     bottle with screw-cap that prevents moisture from entering.
7.3   Methylene chloride—Distilled in glass (Burdick and Jackson, or equivalent).
7.4   GPC calibration solution—Containing 300 mg/mL com oil, 15 mg/mL bis(2-ethylhexyl) phthalate,
     1.4 mg/mL pentachlorophenol, 0.1 mg/mL perylene, and 0.5 mg/mL sulfur.
7.5   Reference matrices.
     7.5.1   Reagent water—Water in which the compounds of interest and interfering compounds are
            not detected by this method.
     7.5.2   High-solids reference matrix—Playground sand or similar material in which the compounds
            of interest and interfering compounds are not detected by this method.
7.6   Standard solutions—Purchased as solutions or mixtures with certification to their purity, concentra-
     tion, and authenticity, or prepared from materials of known purity and composition.  If compound
     purity is 96% or greater, the weight may be used without correction to compute the concentration of
     the standard.  When not being used, standards are stored in the dark at -20 to -10°C in screw-capped
     vials with PTFE-lined lids. A mark is placed on the vial at the level of the solution so that solvent
     evaporation loss can be detected. The vials are brought to room temperature prior to use. Any pre-
     cipitate is redissolved and solvent is added if solvent loss has occurred.

7.7   Preparation of stock solutions—Prepare in methylene chloride, benzene, p-dioxane, or a mixture of
     these solvents per the steps below. Observe the safety precautions in Section 5.  The working range
     for most compounds is 10-200 ug/mL. Compounds with a reduced MS response may be prepared at
     higher concentrations.

     7.7.1   Dissolve an appropriate amount of assayed reference material in a suitable solvent. For
            example, weigh 400 mg N,N-dimethylaniline in a 10-mL ground-glass-stoppered volumetric
            flask and fill to the mark with benzene. After the N,N-dimethylaniline is completely dis-
            solved, transfer the solution to a 15-mL vial with PTFE-lined cap.

-------
Method 1665
            7.7.2   Stock standard solutions should be checked for signs of degradation prior to the preparation
                    of calibration or performance test standards. Quality control check samples may be available
                    from the National Institute of Standards and Technology, Bethesda, Maryland.

            7.7.3   Stock standard solutions shall be replaced after six months, or sooner if comparison with
                    quality control check standards indicates a change in concentration.
       7.8  Labeled compound spiking solution—From stock standard solutions prepared as above, or from mix-
            tures, prepare the spiking solution at a concentration of 200 ug/mL, or at a concentration appropriate
            to the MS response of each compound.
       7.9  Secondary standard—Using stock solutions (Section 7.7), prepare a secondary standard containing
            all of the compounds hi Table 1 at a concentration of 400 ug/mL, or higher concentration appropriate
            to the MS response of the compound.
       7.10 Internal standard solution—Prepare 2,2'-difluorobiphenyl (DFB) at a concentration of 10 mg/mL in
            benzene.

       7.11 DFTPP solution—Prepare at 50 jag/mL in acetone.
       7.12 Solutions for obtaining authentic mass spectra (Section 7.6)—Prepare mixtures of compounds at
            concentrations that will assure authentic spectra are obtained for storage in libraries.
       7.13 Calibration solutions—Combine 0.5 mL of the solution from Section 7.8 with 25,50,125,250, and
            500 |jL of the solution from Section 7.9 and bring each to a total volume of 1.00 mL.  This will pro-
            duce calibration solutions of nominal 10,20,50,100 and 200 ug/mL of the pollutants and a constant
            nominal 100 ug/mL of the labeled compounds. Spike each solution with 10 uL of the internal stan-
            dard solution (Section 7.10). These solutions permit the relative response (labeled to unlabeled) to
            be measured as a function of concentration (Section 10.4).

       7.14 Precision and recovery standard—Used for determination of initial (Section 9.2) and ongoing (Sec-
            tion 13.7) precision and recovery. This solution shall contain the pollutants and labeled compounds
            at a concentration 1000 times the spike level shown in Table 4 for each native and labeled analyte.

       7.15 Stability of solutions—All standard solutions (Sections 7.8-7.14) shall be analyzed within 48 hours
            of preparation and on a monthly basis thereafter for signs of degradation. Standards will remain
            acceptable if the peak area at the quantitation mass relative to the DFB internal standard remains
            within ±15% of the area obtained in the initial analysis of the standard.

       8.   Sample Collection, Preservation, and Handling

       8.1  Collect samples in glass containers following conventional sampling practices (Reference 8).  Aque-
            ous samples that flow freely are collected in refrigerated bottles using automatic sampling equip-
            ment.  Solid samples are collected as grab samples using wide-mouth jars.

       8.2  Maintain samples at 0-4°C from the time of collection until extraction. If residual chlorine is present
            in aqueous samples, add 80 mg sodium thiosulfate per liter of water. EPA Methods 330.4 and 330.5
            may be used to measure residual chlorine (Reference 9).

       8.3  Begin sample extraction within seven days of collection, and analyze all extracts within 40 days of
            extraction.
    8

-------
                                                                                             Method 1665
9.   Quality Assurance/Quality Control

9.1   Each laboratory that uses this method is required to operate a formal quality assurance program (Ref-
     erence 7).  The minimum requirements of this program consist of an initial demonstration of labora-
     tory capability, analysis of samples spiked with labeled compounds to evaluate and document data
     quality, and analysis of standards and blanks as tests of continued performance. Laboratory perfor-
     mance is compared to established performance criteria to determine if the results of analyses meet
     the performance characteristics of the method. If the method is to be applied routinely to samples
     containing high-solids with very little moisture (e.g., soils, filter cake, compost), the high-solids ref-
     erence matrix (Section 7.5.2) is substituted for the reagent water (Section 7.5.1) in all performance
     tests, and the nigh-solids method (Section 11) is used for these tests.
     9.1.1    The analyst shall make an initial demonstration of the ability to generate acceptable accuracy
            and precision with this method. This ability is established as described in Section 9.2.
     9.1.2    In recognition of advances that are occurring in analytical technology, and to allow the ana-
            lyst to overcome sample matrix interferences, the analyst is permitted certain options to
            improve separations or lower the costs of measurements.  These options include alternative
            extraction, concentration, cleanup procedures,  and changes in columns and detectors. Alter-
            native techniques, such as the substitution of spectroscopy or immunoassay, and changes that
            degrade method performance are not allowed.  If an analytical technique other than the tech-
            niques specified in this method is used, that technique must have a specificity equal to or bet-
            ter than the specificity of the techniques in this method for the analytes of interest.
            9.1.2.1   Each time a modification is made to this method, the analyst is required to repeat
                     the procedure  in Section 9.2. If the detection limit of the method will be affected
                     by the change, the laboratory is required to demonstrate that the MDL (40 CFR
                     Part 136, Appendix B) is lower than  one-third the regulatory compliance level.  If
                     calibration will be affected by the change, the analyst must recalibrate the instru-
                     ment per Section  10.
            9.1.2.2   The laboratory is  required to maintain records of modifications made to this
                     method. These records include the information below, at a minimum.
                     9.1.2.2.1    The names, titles, addresses, and telephone numbers of the analyst(s)
                               who performed the analyses and modification, and of the quality con-
                               trol officer who witnessed and will verify the analyses and modifica-
                               tion.
                     9.1.2.2.2   A listing of pollutant(s) measured, by name and CASRN.
                     9.1.2.2.3   A narrative stating the reason(s) for the modification.
                     9.1.2.2.4   Results from all quality control (QC) tests comparing the modified
                               method to this method, including:
                               (a)  Calibration (Section 10.5 or 10.6)
                               (b)  Calibration verification (Section 13.5)
                               (c)  Initial precision and accuracy (Section 9.2)
                               (d)  Labeled compound recovery (Section 9.3)
                               (e)  Analysis of blanks (Section 9.5)
                               (f)   Accuracy assessment (Section 9.4)

-------
Method 1665
                             9.1.2.2.5    Data that will allow an independent reviewer to validate each determi-
                                        nation by tracing the instrument output (peak height, area, or other sig-
                                        nal) to the final result, including:
                                        (a) Sample numbers and other identifiers
                                        (b) Extraction dates
                                        (c) Analysis dates and times
                                        (d) Analysis sequence/run chronology
                                        (e) Sample weight or volume (Section 11)
                                        (f) Extract volume prior to each cleanup step (Section 11.9)
                                        (g) Extract volume after each cleanup step (Section 11.9)
                                        (h) Final extract volume prior to injection (Section 12)
                                        (i) Injection volume (Section 12)
                                        (j) Dilution data, differentiating between dilution of a sample or an
                                           extract (Section 15.4)
                                        (k) Instrument and operating conditions
                                        (I) Column (dimensions, liquid phase, solid support, film thickness,
                                           etc.)
                                        (m) Operating conditions (temperature, temperature program, flow
                                           rates, etc.)
                                        (n) Detector (type, operating condition, etc.)
                                        (o) Chromatograms, printer tapes, and other recording of raw data
                                        (p) Quantitation reports, data system outputs, and other data necessary
                                           to link raw data to the results reported
             9.1.3   Analyses of blanks are required to demonstrate freedom from contamination. The proce-
                    dures and criteria for analysis of a blank are described in Section 9.5.
             9.1.4   The laboratory shall spike all samples with labeled compounds to monitor method perfor-
                    mance. This test is described in Section 9.3. When results of these spikes indicate atypical
                    method performance for samples, the samples are diluted to bring method performance
                    within acceptable limits (Section 15).
             9.1.5   The laboratory shall, on an ongoing basis, demonstrate through calibration verification and
                    the analysis of the precision and recovery standard (Section 7.14) that the analysis system is
                    in control. These procedures are described in Sections 13.1,13.5, and 13.7.
             9.1.6   The laboratory shall maintain records to define  the quality of data that is generated.  Devel-
                    opment of accuracy statements is described in Section 9.4.
       9.2   Initial precision and accuracy—To establish the ability to generate acceptable precision and accuracy,
             the analyst shall perform the operations described in this subsection.
             9.2.1   For low-solids (aqueous) samples, extract, concentrate, and analyze four 1-L aliquots pre-
                    pared by spiking 1 mL of the precision and recovery standard (Section 7.14) into 1 L of
                    reagent water, according to the procedure in Section 11.  For high-solids samples, 30-g ali-
                    quots of the high-solids reference matrix, spiked with 1 mL of the precision and recovery
                    standard, are used.

             9.2.2   Compute the average recovery (X) in ug/L and the standard deviation of the recovery (s) in
                    Hg/L for each compound, by  isotope dilution for pollutants with a labeled analog, and by
                    internal standard for labeled compounds and pollutants with no labeled analog.
    10

-------
                                                                                            Method 1665
     9.2.3   For each compound, compare s and X with the corresponding limits for initial precision and
            accuracy in Table 4. If s and X for all compounds meet the acceptance criteria, system per-
            formance is acceptable and analysis of blanks and samples may begin. If, however, any of
            the compounds fail, the analysis system is not performing properly. In this event, correct the
            problem and repeat the test (Section 9.2.1) or follow the procedure in Section 9.2.3.1
            9.2.3.1   The test may be repeated without adjustment if the analyst believes that it is war-
                     ranted. In this event, compute s and X for only those PMI analytes that failed the
                     test the first time (Sections 9.2.2 and 9.2.3). If these PMI analytes now pass, sys-
                     tem performance is acceptable for all compounds and analysis of samples and
                     blanks may begin. If, however, any of the same PMI analytes fail again, the
                     extraction/concentration processes are not being performed properly for these
                     compounds.  In this event, correct the problem, and begin again.
9.3   The laboratory shall spike all samples with labeled compounds to assess method performance on the
     sample matrix.
     9.3.1   Analyze each sample according to the method beginning in Section 11.
     9.3.2   Compute the percent recovery (P) of the labeled compounds using the internal standard
            method (Section 10.5).
     9.3.3   Compare the labeled compound recovery for each compound with the corresponding limits
            in Table 4. If the recovery of any compound falls outside these limits, method performance
            is unacceptable for that compound in that sample.  To overcome such difficulties, water sam-
            ples are diluted, and smaller amounts of soils, sludges, and sediments are reanalyzed per Sec-
            tion 15.
9.4   As part of the QA program for the laboratory, method accuracy for samples shall be assessed and
     records shall be maintained. After the analysis of five samples of a given matrix type (e.g., water,
     soil, sludge, sediment) for which the labeled compounds pass the tests in Section 9.3, compute the
     average percent recovery (P) and the standard deviation of the percent recovery (Sp) for the labeled
     compounds only. Express the accuracy assessment as a percent recovery interval from P - 2spto
     P + 2sp for each matrix.

     For example, if P = 90% and sp = 10% for five analyses of compost, the accuracy interval is
     expressed as 70-110%. Update the accuracy assessment for each compound in each matrix on a reg-
     ular basis (e.g., after each five to ten new accuracy measurements).
9.5   Blanks—Reagent water and high-solids reference matrix blanks are analyzed to demonstrate free-
     dom from contamination.

     9.5.1   Extract and concentrate a 1-L reagent water blank or a high-solids reference matrix blank
            with each sample lot (samples started through the extraction process on the same 8-hour
            shift, to a maximum of 20 samples). Analyze the blank immediately after analysis of the
            precision and recovery standard (Section 7.14) to demonstrate freedom from contamination.
     9.5.2   If any of the compounds of interest (Table 1) or any potentially interfering compound is
            found in an aqueous blank at greater than the ML (assuming a response factor of 1 relative to
            the internal standard for compounds not listed in Table 1), analysis of samples is halted until
            the source of contamination is eliminated and a blank shows no evidence of contamination at
            this level.
                                                                                                11

-------
Method 1665
       9.6  The specifications contained in this method can be met if the apparatus used is calibrated properly,
            then maintained in a calibrated state. The standards used for calibration (Section 10), calibration ver-
            ification (Section 13.5), and initial (Section 9.2) and ongoing (Section 13.7) precision and recovery
            should be identical, so that the most precise results will be obtained. The GC/MS instrument in par-
            ticular will provide the most reproducible results if dedicated to the settings and conditions required
            for the analyses of semivolatiles by this method.
       9.7  Depending on specific program requirements, field replicates may be collected to determine the pre-
            cision of the sampling technique, and spiked samples may be required to determine the accuracy of
            the analysis when the internal standard method is used.

       10.  Calibration

       10.1  Assemble the GC/MS system and establish the operating conditions in Section 6.11.1. Analyze stan-
            dards per the procedure in Section 12 to demonstrate that the analytical system meets the minimum
            levels in Table 2, and the mass-intensity criteria in Table  3 for 50 ng DFTPP.
       10.2  Mass-spectral libraries—Detection and identification of compounds of interest are dependent upon
            spectra stored in user created libraries.
            10.2.1   Obtain a mass spectrum of each pollutant, labeled compound, and the internal standard by
                    analyzing an authentic standard either singly or as part of a mixture in which there is no
                    interference between closely eluted components. Examine the spectrum to determine that
                    only a single compound is present.  Fragments not attributable to the compound under study
                    indicate the presence of an interfering compound.
            10.2.2   Adjust the analytical conditions and scan rate (for this test only) to produce an undistorted
                    spectrum at the GC peak maximum. An undistorted spectrum will usually be obtained if five
                    complete spectra are collected across the upper half of the GC peak. Software algorithms
                    designed to "enhance" the spectrum may eliminate distortion, but may also eliminate authen-
                    tic masses or introduce other distortion.
            10.2.3   The authentic reference spectrum is obtained under DFTPP tuning conditions (Section 10.1
                    and Table 3) to normalize it to spectra from other instruments.
            10.2.4   The spectrum is edited by saving the five most intense mass-spectral peaks and all other
                    mass spectral peaks greater than 10% of the  base peak.  The spectrum may be further edited
                    to remove the common interfering m/z's. If five  mass spectral peaks cannot be obtained
                    under the scan conditions given in Section 6.12, the mass spectrometer may be scanned to an
                    m/z lower than 35 to gain additional spectral information. The spectrum obtained is stored
                    for reverse search and for compound confirmation.
       10.3  Analytical range—Demonstrate that 20 ng anthracene or phenanthrene produces an area at m/z 178
            approximately one-tenth that required to exceed the linear range of the system. The exact value must
            be determined by experience for each instrument. It is used to match the calibration range of the
            instrument to the analytical range and detection limits required,  and to diagnose instrument sensitiv-
            ity problems (Section 16.3).  The 20 ug/mL calibration standard (Section 7.13) can be used to dem-
            onstrate this performance.
            10.3.1  Polar compound detection—Demonstrate that unlabeled pentachlorophenol and benzidine
                   are detectable at the 50 |ig/mL level (per all criteria in Section 14). The 50 ng/mL calibration
                   standard (Section 7.13) can be used to demonstrate this  performance.
   12

-------
                                                                                             Method 1665
10.4 Calibration with isotope dilution—Isotope dilution is used when (1) labeled compounds are avail-
     able, (2) interferences do not preclude its use, and (3) the quantitation m/z (Table 5) EICP area for the
     compound is in the calibration range. Alternative labeled compounds and quantitation m/z's may be
     used based on availability. If any of the above conditions preclude isotope dilution, the internal stan-
     dard method (Section 10.5) is used.


Note:   Because the composition and purity of commercially supplied isotopically labeled standards may
        vary, the primary m/z of the labeled analogs given in Table 5 should be used as guidance. The
	appropriate m/z of the  labeled analogs should be determined prior to use for sample analysis.


     10.4.1  A calibration curve encompassing the concentration range is prepared for each compound to
            be determined.  The relative response (pollutant-to-labeled) versus concentration in standard
            solutions is plotted or computed using a linear regression. The example in Figure 1 shows a
            calibration curve for aniline using aniline-d7 as the isotopic diluent. Also shown are the
            ±10% error limits (dotted lines). Relative response (RR) is determined according to the pro-
            cedures described below. A minimum of five data points are employed for calibration.
     10.4.2  The relative response of a pollutant to its labeled analog is determined from isotope ratio val-
            ues computed from acquired data. Three isotope ratios are used in this process:
                                area m,
                            R =
                             x

                           R =
  x          T
             1
                                aream2/z
                                 aream
,.      '^
 m
                                 area m^/z I at RT. 1
                      Where:

                      Rx = The isotope ratio measured for the pure pollutant

                      Ry = The isotope ratio measured for the labeled compound

                      Rftr The isotope ratio of an analytical mixture of pollutant and
                          labeled compounds

            The m/z's are selected such that R, > R^ If Rm is not between 2R, and 0.5R,, the method
            does not apply and the sample is analyzed by the internal standard method.
     10.4.3  Capillary columns usually separate the pollutant-labeled pair, with the labeled compound
            eluted first (Figure 2). This is indeed the case for the PMI analytes treated here.
     10.4.4  Special precautions are taken when the pollutant-labeled pair is not separated, or when
            another labeled compound with interfering spectral masses overlaps the pollutant (a case that
            can occur with isomeric compounds). In this case, it is necessary to determine the respective
                                                                                                 13

-------
Method 1665
                   contributions of the pollutant and labeled compounds to the respective EICP areas. If the
                   peaks are separated well enough to permit the data system or operator to remove the contri-
                   butions of the compounds to each other, the equations in Section 10.4.2 apply. This usually
                   occurs when the height of the valley between the two GC peaks at the same m/z is less than
                   10% of the height of the shorter of the two peaks. If significant GC and spectral overlap
                   occur, RR is computed using  the following equation:
                                          RR =
(R -R  }(R
\  y   mj\
(R  -R Y/?
^  m    *A
                   where R, is measured as shown in Figure 3A, Ry is measured as shown in Figure 3B, and R,,,
                   is measured as shown in Figure 3C.  For the example,
                                               46100
                                          V -4780


                                          Ry= 43600= °-06078



                                          Rm= 48300= L019

                                          RR= 1.115


                   The data from these analyses are reported to three significant figures (see Section 15.6).
                   Therefore, in order to prevent rounding errors from affecting the values to be reported, all
                   calculations performed prior to the final determination of concentrations should be carried
                   out using at least four significant figures.
            1 0.4.5  To calibrate the analytical system by isotope dilution, analyze a 1 .0-|jL aliquot of each of the
                   calibration standards (Section 7.13) using the procedure in Section 12. Compute the RR at
                   each concentration.
            10.4.6  Linearity — If the ratio of relative response to concentration for any compound is constant
                   (less than 20% coefficient of variation) over the five-point calibration range, an averaged rel-
                   ative response/concentration ratio may be used for that compound; otherwise, the complete
                   calibration curve for that compound shall be used over the five-point calibration range.
       10.5  Calibration by internal standard — Used when criteria for isotope dilution (Section 10.4) cannot be
            met  The internal standard to be used for both acid and base/neutral analyses is 2,2'-difluorobiphe-
            nyl. The internal standard method is also applied to determination of compounds having no labeled
            analog, and to measurement of labeled compounds for intra-laboratory statistics (Sections 9.4 and
            13.7.4).
   14

-------
                                                                                            Method 1665
     10.5.1  Response factors—Calibration requires the determination of response factors (RF) which are
            defined by the following equation:
                                              A  xC.
                                              -1	if
                                              A. xC
                                               IS    S
                      Where:
                      As = The area of the characteristic mass for the compound in the
                          daily standard
                      AX = The area of the characteristic mass for the internal standard

                      Qy = The concentration of the internal standard (/jg/mL)

                      Cs —The concentration of the compound in the daily standard
                          (Hg/mL)
            10.5.1.1   The response factor is determined for at least five concentrations appropriate to the
                     response of each compound (Section 7.13); nominally, 10,20,50,100, and 200 ug/
                     mL. The amount of internal standard added to each extract is the same (100 ug/
                     mL) so that Cu remains constant.  The RF is plotted versus concentration for each
                     compound in the standard (CJ to produce a calibration curve.

            10.5.1.2   Linearity—If the response factor (RF) for any compound is constant (less than
                     35% coefficient of variation) over the five-point calibration range, an averaged
                     response factor may be used for that compound; otherwise, the complete calibra-
                     tion curve for that compound shall be used over the five-point range.
10.6  Combined calibration—By using calibration solutions (Section 7.13) containing the pollutants,
     labeled compounds, and the internal standard, a single set of analyses can be used to produce calibra-
     tion curves for the isotope dilution and internal standard methods. These curves are verified each
     shift (Section 13.5) by analyzing the  100 ug/mL calibration standard (Section 7.13). Recalibration is
     required only if calibration verification (Section 13.5) criteria cannot be met.

11.  Sample Extraction, Concentration, and Cleanup

Samples containing less than 1% solids are extracted directly using continuous liquid/liquid extraction
techniques (Section 11.2.1 and Figure 4). Samples containing 1-30% solids are diluted to the 1% level
with reagent water (Section 11.2.2) and extracted using continuous liquid/liquid extraction techniques.
Samples containing greater than 30% solids are extracted using ultrasonic techniques (Section 11.2.5).
11.1  Determination of percent solids.
     11.1.1   Weigh 5-10 g of sample into a tared beaker.

     11.1.2  Dry overnight (12 hours minimum) at 110°C(±5°C), and cool in a desiccator.
                                                                                                15

-------
Metfjod 1665
             11.1.3   Determine percent solids as follows:
                                          ...     weight of dry sample
                                      % sollds =  weight of vet sample X 10°
       1 1 .2 Preparation of samples for extraction.

            11.2.1   Samples containing less than 1% solids — Extract sample directly using continuous liquid/
                    liquid extraction techniques.

                    11.2.1.1   Measure  1.00 L (±0.01 L) of sample into a clean 1.5- to 2.0-L beaker.

                    11.2.1 2   Dilute aliquot — For samples which are expected to be difficult to extract, concen-
                             trate, or clean up, measure an additional 100.0 mL (±1.0 mL) into a clean 1.5- to
                             2.0-L beaker and dilute to a final volume of 1.00 L (±0.1 L) with reagent water.

                    11 .2.1 .3   Spike 0.5 mL of the labeled compound spiking solution (Section 7.8) into the sam-
                             ple aliquots. Proceed to preparation of the QC aliquots for low-solids samples
                             (Section 11.2.3).

            112.2   Samples containing 1-30% solids.

                    1 1 .2.2.1   Mix sample thoroughly.

                             Using the percent solids found in 11.1.3, determine the weight of sample required
                             to produce 1 L of solution containing 1% solids as follows:
                                             ,    •  z.
                                        sample weight =  % solids grams

                    11 .22.3   Discard all sticks, rocks, leaves, and other foreign material prior to weighing.
                             Place the weight determined in 11.2.2.2 in a clean 1.5- to 2.0-L beaker.
                    11 .22.4   Dilute aliquot — For samples that are expected to be difficult to extract, concen-
                             trate, or clean up, weigh an amount of sample equal to one-tenth the amount deter-
                             mined in 11.2.2.2 into a second clean 1.5- to 2.0-L beaker. When diluted to 1.0 L,
                             this dilute aliquot will contain 0.1% solids.
                    11 .2.2.5   Bring the sample aliquot(s) above to a volume of 100-200 mL with reagent water.

                    112.2.6   Spike 0.5 mL of the labeled compound spiking solution (Section 7.8) into each
                             sample aliquot.
                    11 .2.2.7   Using a clean metal spatula, break any solid portions of the sample into small
                             pieces.

                    11.2.2.8   Place the %" horn on the ultrasonic probe approximately Vi" below the surface of
                             each sample aliquot and pulse at 50% for 3 minutes at full power.  If necessary,
                             remove the probe from the solution and break any large pieces using the metal
                             spatula or a stirring rod and repeat the sonication. Clean the probe with methylene
                             chloride:acetone (1:1) between samples to preclude cross-contamination.

                    11 22.9   Bring the sample volume to 1 .0 L (±0. 1 L) with reagent water.
   16

-------
                                                                                             Method 1665
     11.2.3  Preparation of QC aliquots for samples containing low solids (<30%).
            11.2.3.1   For each sample or sample lot (to a maximum of 20) to be extracted at the same
                     time, place three  1.0-L (±0.01 L) aliquots of reagent water in clean  1.5- to 2.0-L
                     beakers.
            11.2.3.2   Spike 0.5 mL of the labeled compound spiking solution (Section 7.8) into one
                     reagent water aliquot. This aliquot will serve as the blank.
            11.2.3.3   Spike 1.0 mL of the precision and recovery standard (Section 7.14)  into the two
                     remaining reagent water aliquots.
     11.2.4  Stir and equilibrate all sample and QC solutions for 1-2 hours. Extract the samples and QC
            aliquots per Section 11.3.
     11.2.5  Samples containing greater than 30% solids.
            11.2.5.1   Mix the sample thoroughly.
            11.2.5.2   Discard all sticks, rocks, leaves and other foreign material prior to weighing.
                     Weigh 30 g (±0.3 g) into a clean 400- to 500-mL beaker.
            11.2.5.3   Dilute aliquot—For samples which are expected to be difficult to extract, concen-
                     trate, or clean up, weigh 3 g (±0.03 g) into a clean 400- to 500-mL beaker.
            11.2.5.4   Spike 0.5 mL of the labeled compound spiking solution (Section 7.8) into each
                     sample aliquot.
            11.2.5.5   QC aliquots—For each sample or sample lot (to a maximum of 20) to be extracted
                     at the  same time, place three 30 g (±0.3 g) aliquots of the high-solids reference
                     matrix in clean 400- to 500-mL beakers.
            11.2.5.6   Spike 0.5 mL of the labeled compound spiking solution (Section 7.8) into one
                     nigh-solids reference matrix aliquot. This aliquot will serve as the blank.
            11.2.5.7   Spike 1.0 mL of the precision and recovery standard (Section 7.14)  into the two
                     remaining high-solids reference matrix aliquots. Extract, concentrate, and clean up
                     the high-solids samples and QC aliquots per Sections 11.4 through 11.8.
11.3  Continuous extraction of low-solids (aqueous) samples—Place 100-150 mL methylene chloride in
     each continuous extractor and 200-300 mL in each distilling flask.
     11.3.1  Pour the sample(s), blank, and QC aliquots into the extractors. Rinse the glass containers
            with 50-100 mL methylene chloride and add to the respective extractors. Include all solids
            in the extraction process.
     11.3.2  Base/neutral extraction—Adjust the pH of the waters in the extractors to 12 -13 with 6N
            NaOH while monitoring with a pH meter. Begin the extraction by heating the flask until the
            methylene chloride is boiling. When properly adjusted, one to two drops of methylene chlo-
            ride per second  will fall from the condenser tip into the water. Test and adjust the pH of the
            waters during the first to second hour and during the fifth to tenth hour of extraction. Extract
            for 24-48 hours.

     11.3.3  Remove the distilling flask, estimate and record the volume of extract (to the  nearest 100
            mL), and pour the contents through a drying column containing 7-10 cm anhydrous sodium
            sulfate.  Rinse the distilling flask with 30-50 mL of methylene chloride and pour through the
            drying column.  Collect the solution in a 500-mL K-D evaporator flask equipped with a 10-
                                                                                                 17

-------
Method 1665
                    mL concentrator tube. Seal, label as the base/neutral fraction, and concentrate per Sections
                    11.5 to 11.6.


       Note:   All of the PMI analytes listed in Table 1 will be extracted from basic media. Acid extraction is not
               necessary unless other analytes are being sought.


            11.3.4   Acid extraction—Adjust the pH of the waters in the extractors to 2 or less using 6N sulfuric
                    acid. Charge clean distilling flasks with 300-400 mL of methylene chloride. Test and adjust
                    the pH of the waters during the first 1-2 hours and during the fifth to tenth hour of extraction.
                    Extract for 24—48 hours. Repeat Section 11.3.3, except label as the acid fraction.
       11.4 Ultrasonic extraction of high-solids samples.
            11.4.1   Add 60 g of powdered anhydrous sodium sulfate to the sample and QC aliquot(s) and mix
                    thoroughly.
            11.4.2   Add 100 mL (±10 mL) of acetone:methylene chloride (1:1) to the sample and mix thor-
                    oughly.

            11.4.3   Place the %" horn on the ultrasonic probe approximately Vi" below the surface of the solvent
                    but above the solids layer and pulse at SO percent for three minutes at full power.  If neces-
                    sary, remove the probe from the solution and break any large pieces using the metal spatula
                    or a stirring rod and repeat the sonication.
            11.4.4   Decant the extracts through Whatman 41 filter paper (or equivalent) using glass funnels and
                    collect in 500- to 1000-mL graduated cylinders.
            11.4.5   Repeat the extraction steps (Section 11.4.2 to Section 11.4.4) twice more for each sample and
                    QC aliquot. On the final extraction, swirl the sample or QC aliquot, pour into its respective
                    glass funnel, and rinse with acetone:methylene chloride. Record the total extract volume.
            11.4.6   Pour each extract through a drying column containing 7-10 cm of anhydrous sodium sulfate.
                    Rinse the graduated cylinder with 30-50 mL of methylene chloride and pour through the dry-
                    ing column. Collect each extract in a 500-mL K-D evaporator flask equipped with a  10-mL
                    concentrator tube. Seal and label as the high-solids semivolatile fraction. Concentrate and
                    clean up the samples and QC aliquots per Sections 11.5 through 11.8.
       11.5 Macro concentration—Concentrate the extracts in separate 500-mL K-D flasks equipped with 10-mL
            concentrator tubes.
            11.5.1   Add one or two clean boiling chips to the flask and attach a three-ball macro Snyder column.
                    Prewet the column by adding approximately 1 mL of methylene chloride through the top.
                    Place the K-D apparatus in a hot water bath so that the entire lower rounded surface of the
                    flask is bathed with steam.  Adjust the vertical position of the apparatus and the water tem-
                    perature as required to complete the concentration in 15 to 20 minutes.  At the proper rate of
                    distillation, the balls of the column will actively chatter but the chambers will not flood.
                    When the liquid has reached an apparent volume of 1 mL, remove the K-D apparatus from
                    the bath and allow the solvent to drain and cool for at least 10 minutes.  Remove the Snyder
                    column and rinse the flask and its lower joint into the concentrator tube with 1-2 mL of
                    methylene chloride. A 5-mL syringe is recommended for this operation.
    18

-------
                                                                                            Method 1665
     11.5.2  For performance standards (Sections 9.2 and 13.7) and for blanks (Section 9.5), combine the
            acid and base/neutral extracts for each at this point.  Do not combine the acid and base/neu-
            tral extracts for aqueous samples.
11.6 Micro-concentration.
     11.6.1  Kuderna-Danish (K-D)—Add a clean boiling chip and attach a two-ball micro Snyder col-
            umn to the concentrator tube. Prewet the column by adding approximately 0.5 mL methyl-
            ene chloride through the top. Place the apparatus in the hot water bath. Adjust the vertical
            position and the water temperature as required to complete the concentration in 5-10 min-
            utes. At the proper rate of distillation, the balls of the column will actively chatter but the
            chambers will not flood.  When the liquid reaches an apparent volume of approximately
            0.5 mL, remove the apparatus from the water bath and allow to drain and cool for at least 10
            minutes. Remove the micro Snyder column and rinse its lower joint into the concentrator
            tube with approximately 0.2 mL of methylene chloride. Adjust the final volume to 5.0 mL if
            the extract is to be cleaned up by GPC, to 1.0 mL if it does not require clean-up, or to 0.5 mL
            if it has been cleaned up.
     11.6.2  Nitrogen blowdown—Place the concentrator tube in a warm water bath (35°C) and evapo-
            rate the solvent volume using a gentle stream of clean, dry nitrogen (filtered through a col-
            umn of activated carbon).


Caution: New plastic tubing must not be used between the carbon trap and  the sample, since it may intro-
        duce interferences.


            The internal wall of the tube must be rinsed down several times with methylene chloride dur-
            ing the operation. During evaporation, the tube solvent level  must be kept below the water
            level of the bath.  The extract must never be allowed to become dry.  Adjust the final volume
            to 5.0 mL if the extract is to be cleaned up by GPC, to 1.0 mL if it does not require clean-up,
            or to 0.5 mL if it has been cleaned up.
11.7 Transfer the concentrated extract to a clean screw-cap vial. Seal the vial with a PTFE-lined lid, and
     mark the level on the vial. Label with the sample number and fraction, and store in the dark at -20 to
     -10°C until ready for analysis.
11.8 GPC setup and calibration.
     11.8.1  Column packing.

            11.8.1.1  Place 75 g (±5 g) of SX-3 Bio-beads  in a 400- to 500-mL beaker.
            11.8.1.2  Cover the beads and allow to swell overnight (12 hours minimum).
            11.8.1.3  Transfer the swelled beads to the column and pump solvent through the column,
                    from bottom to top, at 4.5-5.5 mL/min prior to connecting the column to the detec-
                    tor.

            11.8.1.4  After purging the column with solvent for 1-2 hours, adjust the column head pres-
                    sure to 7-10 psig, and purge for 4-5 hours to remove air from the column. Main-
                    tain a head pressure of 7-10 psig. Connect the column to the detector.
     11.8.2  Column calibration.

            11.8.2.1  Load 5 mL of the calibration solution (Section 7.4)  into the sample loop.
                                                                                                19

-------
Method 1665
                    11.8.2.2   Inject the calibration solution and record the signal from the detector.  The elution
                             pattern will be com oil, bis(2-ethylhexyl)phthalate, pentachlorophenol, perylene,
                             and sulfur.
                    11.8.2.3   Set the "dump time" to allow >85% removal of the corn oil and >85% collection of
                             the phthalate.

                    11.8.2.4   Set the "collect time" to the peak minimum between perylene and sulfur.
                    11.8.2.5   Verify the calibration with the calibration solution after every 20 extracts. Calibra-
                             tion is verified if the recovery of the pentachlorophenol is greater than 85%. If cal-
                             ibration is not verified, the system shall be recalibrated using the calibration
                             solution, and the previous 20 samples shall be re-extracted and cleaned up using
                             the calibrated GPC system.
       11.9 Extract cleanup.

            11.9.1   Filter the extract or load through the filter holder to remove particulates. Load the 5.0-mL
                    extract onto the column.  The maximum capacity of the column is 0.5-1.0 g. If necessary,
                    split the extract into multiple aliquots to prevent column overload.
            11.9.2   Elute the extract using the calibration data determined in 11.8.2. Collect the eluate in a clean
                    400- to 500-mL beaker.
            11.9.3   Concentrate the cleaned up extract to 5.0-mL per Section 11.5.
            11.9.4   Rinse the sample loading tube thoroughly with methylene chloride between extracts to pre-
                    pare for the next sample.
            11.9.5   If a particularly dirty extract is encountered, a 5.0-mL methylene chloride blank shall be run
                    through the system to check for carry-over.
            11.9.6   Concentrate the extract to 0.5 mL and transfer to a screw-cap vial per Sections 11.6 and 11.7.
                    Concentrating these extracts to 0.5 mL will compensate for the loss of one-half of the extract
                    in the cleanup process.

       12.  GC/MS Analysis

       12.1  Establish the operating  conditions given in Section 6.11.1 for analysis of extracts.
       12.2  Bring the concentrated extract (Section 11.7) or standard (Sections 7.13-7.14) to room temperature
            and verify that any precipitate has redissolved. Verify the level on the extract (Sections 7.6 and 11.7)
            and bring to the mark with solvent if required.
       12.3  Add the internal standard solution (Section 7.10) to the extract (use 1.0 uL of solution per 0.1 mL of
            extract) immediately prior to injection to minimize the possibility of loss by evaporation, adsorption,
            or reaction.  Mix thoroughly.
       12.4  Inject a volume of the standard solution or extract such that 100 ng of the internal standard will be
            injected, using on-column or splitless injection. For 1-mL extracts, this volume will be 1.0 pL.  Start
            the GC column initial isothermal hold upon injection. Start MS data collection after the solvent peak
            elutes. Stop data collection after the internal standard (DFB) elutes. Return the column to the initial
            temperature for analysis of the next sample.
   20

-------
                                                                                            Method 1665
13.  System and Laboratory Performance

13.1  At the beginning of each 8-hour shift during which analyses are performed, GC/MS system perfor-
     mance and calibration are verified for all pollutants and labeled compounds. For these tests, analysis
     of the calibration standard (Section 7.14) shall be used to verify all performance criteria. Adjustment
     and/or recalibration (per Section 10) shall be performed until all performance criteria are met.  Only
     after all performance criteria are met may samples, blanks, and precision and recovery standards be
     analyzed.
13.2  DFTPP spectrum validity—Inject 1  jjL of the DFTPP solution (Section 7.11) either separately or
     within a few seconds of injection of the standard (Section 13.1) analyzed at the beginning of each
     shift. The criteria in Table 3 shall be met.
13.3  Retention times—The absolute retention time of 2,2-difluorobiphenyl shall be within the range of
     1092 to 1268 seconds and the relative retention times of all pollutants and labeled compounds shall
     fall within the limits given in Table  2 or within ±10% of that specified.
13.4  GC resolution—All native PMI analytes shall be fully resolved and the valley height between any
     two analytes shall not exceed 10% of the taller of the two peaks.
13.5  Calibration verification—Compute the concentration of each pollutant (Table 1) by isotope dilution
     (Section 10.4) for those compounds that have labeled analogs. Compute the concentration of each
     pollutant that has no labeled analog by  the internal standard method (Section 10.5). Compute the
     concentration of the labeled compounds by the internal standard method. These concentrations are
     computed based on the calibration data determined in Section 10.
     13.5.1  For each pollutant and labeled  compound being tested, compare the concentration with the
            calibration verification limit in Table 4.  If all compounds meet the acceptance criteria, cali-
            bration has been verified and analysis of blanks, samples, and precision and recovery stan-
            dards may proceed. If, however, any compound fails, the measurement system is not
            performing properly for that compound. In this event, prepare a fresh calibration standard  or
            correct the problem causing the failure and repeat the test (Section 13.1), or recalibrate (Sec-
            tion  10).
13.6  Multiple peaks—Each compound injected shall give a single, distinct GC peak.
13.7  Ongoing precision and accuracy.
     13.7.1  Analyze the extract of the precision and recovery standard (Section 9) prior to analysis of
            samples from the same lot.

     13.7.2  Compute the concentration  of each pollutant (Table 1) by isotope dilution (Section 10.4) for
            those compounds which have labeled analogs.  Compute the concentration of each pollutant
            that has no labeled analog by the internal standard method (Section 10.5). Compute the
            concentration of the labeled compounds by the internal standard method.

     13.7.3  For each pollutant and labeled  compound, compare the concentration with the limits for on-
            going accuracy in Table 4.  If all compounds meet the acceptance criteria, system perfor-
            mance is acceptable and analysis of blanks and samples may proceed.  If, however, any indi-
            vidual concentration falls outside of the range given, system performance is unacceptable for
            that compound.  In this event, correct the problem, re-extract the sample lot (Section 11) and
            repeat the ongoing precision and recovery test (Section 13.7).

     13.7.4  Add results that pass the specifications in Section 13.7.3 to initial and previous ongoing data
            for each compound in each  matrix.  Update QC charts to form a graphic representation of
            continued laboratory performance (Figure 5). Develop a statement of laboratory accuracy
                                                                                                21

-------
Method 1665
                    for each pollutant and labeled compound in each matrix type by calculating the average per-
                    cent recovery (R) and the standard deviation of percent recovery (sr).  Express the accuracy
                    as a recovery interval from R - 2sr to R + 2sr For example, if R = 95% and sr = 5%, the accu-
                    racy is 85-105%.

       14.  Qualitative Determination

       Identification is accomplished by comparison of data from analysis of a sample or blank with data stored in
       the mass-spectral libraries. For compounds for which the relative retention times and mass spectra are
       known, identification is confirmed per Sections 14.1 and 14.2.
       14.1  Labeled compounds and pollutants having no labeled analog.
            14.1.1   The signals for all characteristic m/z's stored in the spectral library (Section 10.2.4) shall be
                    present and shall maximize within the same two consecutive scans.
            14.1.2   Either (1) the background corrected EICP areas, or (2) the corrected relative intensities of the
                    mass-spectral peaks at the GC peak maximum shall agree within a factor of 2 (0.5 to 2 times)
                    for all masses stored in the library.
            14.1.3   For the compounds for which the system has been calibrated (Table 1), the retention  time
                    shall agree with those specified in Table 2, or have a relative retention time ±10% of that
                    specified.
       14.2  Pollutants having a labeled analog.
            14.2.1   The signals for all characteristic m/z's stored in the spectral library (Section 10.2.4) shall be
                    present and shall maximize within the same two consecutive scans.
            14.2.2   Either (1) the background corrected EICP areas, or (2) the corrected relative intensities of the
                    mass-spectral peaks at the GC peak maximum shall agree within a factor of two for all
                    masses stored in the spectral library.
            14.2.3   The relative retention time between the pollutant and its labeled analog shall be within the
                    windows specified in Table 2 or within ±10% of that specified.
       14.3  The m/z's present hi the experimental mass spectrum that are not present in the reference mass spec-
            trum shall be accounted for by contaminant or background ions. If the experimental mass spectrum
            is contaminated, or if identification is ambiguous, an experienced spectrometrist (Section 1.4) is to
            determine the presence  or absence of the compound.

       15.  Quantitative Determination

       15.1  Isotope dilution—Because the pollutant and its labeled analog exhibit the same effects upon  extrac-
            tion, concentration, and gas chromatography,  correction for recovery of the pollutant can be made by
            adding a known amount of a labeled compound to every sample prior to extraction. Relative
            response (RR) values for sample mixtures are used in conjunction with the calibration curves
            described in Section 10.4 to determine concentrations directly, as long as labeled compound spiking
            levels are constant. For the aniline example given in Figure 1 (Section 10.4.1), RR would be equal to
            1.114.  For this RR value, the aniline calibration curve given in Figure 1 indicates a concentration of
            27 ug/mL in the sample extract
   22

-------
                                                                                             Method 1665
15.2 Internal standard — compute the concentration in the extract using the response factor determined
     from calibration data (Section 10.5) and the following equation:

                                                   A  x C.
                                                    is


                      Where:
                      Cgx = The concentration of the compound in the extract, and the
                          other terms are as defined in Section 10.5.1.
15.3 The concentration of the pollutant in the solid phase of the sample is computed using the concentra-
     tion of the pollutant in the extract and the weight of the solids (Section 11), as follows:
                                                          C   xV
                           *                 ,. , ,    n •.     eX    eX
                           Concentration in solid ((ig/xg) = - ^ -
                                                               s
                      Where:
                      Vex = The extract volume in mL

                      Ws = The sample weight in kg

15.4 Dilution of samples — If the EICP area at the quantitation m/z for any compound exceeds the calibra-
     tion range of the system, the extract of the dilute aliquot (Section 11) is analyzed by isotope dilution.
     If further dilution is required and the sample holding time has not been exceeded, a smaller sample
     aliquot is extracted per Sections 15.4.1-15.4.3.  If the sample holding time has been exceeded, the
     sample extract is diluted by successive factors of 10, internal standard is added to give a concentra-
     tion of 100 ug/mL in the diluted extract, and the diluted extract is analyzed by the internal-standard
     method.
     15.4.1  For samples containing 1% solids or less for which the holding time has not been exceeded,
            dilute 10 mL, 1.0 mL, 0.1 mL, etc., of sample to 1 L with reagent water and extract per Sec-
            tion 11.2.1.
     15.4.2  For samples containing 1-30% solids for which the holding time has not been exceeded,
            extract an amount of sample equal to one-hundredth the amount determined in 1 1.2.2.2.
            Extract per Section 11.2.2.
     15.4.3  For samples containing 30% solids or greater for which the holding time has not been
            exceeded, extract 0.30 g (±0.003 g) of sample per Section 11.2.5.
15.5 Results are reported to three significant figures for all pollutants and labeled compounds found in all
     standards, blanks, and samples. For aqueous samples, the units used are ng/L. For samples containing
     1-30% solids, ug/L may be used if the actual sample volume before dilution to 1% solids is used in
     the calculation. In this latter case, the laboratory must report the percent solids so that conversion to
     other units may be made, if required. The units of fig/kg may be used for samples containing 1-30%
     solids, if based on dry weight, but must be used for samples containing greater than 30% solids,
     based on dry weight.

     15.5.1  Results for samples that have been diluted are reported at the least dilute level at which the
            area at the quantitation m/z is within the calibration range (Section 15.4), or at which no  m/z
            in the spectrum is saturated (Section 15.5). For compounds having a labeled analog, results
                                                                                                 23

-------
Method 1665
                    are reported at the least dilute level at which the area at the quantitation m/z is within the cal-
                    ibration range (Section 15.4) and the labeled compound recovery is within the normal range
                    for the method (Section 16.4).

       16.   Analysis of Complex Samples

       16.1   Some samples may contain high levels (>1000 jig/L) of the compounds of interest, interfering com-
             pounds, and/or polymeric materials. Some samples will not concentrate to 1 mL (Section 11.6); oth-
             ers will overload the GC column and/or mass spectrometer.
       16.2   Analyze the dilute aliquot (Section 11) when the sample will not concentrate to 1 mL. If a dilute ali-
             quot was not extracted, and the sample holding time (Section 8.3) has not been exceeded, dilute an
             aliquot of an aqueous sample with reagent water, or weigh a dilute aliquot of a high-solids sample
             and re-extract (Section 11); otherwise, dilute the extract (Section 1S.4) and analyze by the internal
             standard method (Section 15.2).
       16.3   Recovery of internal standard—The EICP area of the internal standard should be within a factor of
             two of the area in the shift standard (Section 13.1). If the absolute areas of the labeled compounds
             are within a factor of 2 of the respective areas in the shift  standard, and the internal standard area is
             less than one-half of its respective area, then loss of the internal standard in the extract has occurred.
             In this case, use one of the labeled compounds to compute the concentration of a pollutant with no
             labeled analog.
       16.4   Recovery of labeled compounds—In most samples, labeled compound recoveries will be similar to
             those from reagent water or from the high-solids reference matrix (Section 13.7). If the labeled com-
             pound recovery is outside the limits given in Table 4, the extract from the dilute aliquot (Section 11)
             is analyzed as in Section 15.4.  If the recoveries of all labeled compounds and the internal standard
             are low (per the criteria above), then a loss in instrument sensitivity is the most likely cause. In this
             case, the 100 ug/mL calibration standard (Section 13.1) shall be analyzed and calibration verified
             (Section 13.5). If a loss in sensitivity has occurred, the instrument shall be repaired, the performance
             specifications in Section 13 shall be met, and the extract reanalyzed. If a loss in instrument sensitivity
             has not occurred, the laboratory must use additional cleanup steps to remove interferences. Sugges-
             tions for additional cleanup are provided in Reference 10.  If all attempts to resolve interference
             problems fail, the method does not apply to the sample being analyzed, and the result may not be
             reported for regulatory compliance purposes.

       17.   Method Performance

       17.1   With the exception of o-picoline (2-methylpyridine), this  method was validated in a single labora-
             tory. a-Picoline was validated in the interlaboratory validation of EPA Method 1625 (Reference 6).
       17.2   A chromatogram of the 100 jjg/mL calibration standard (Section 7.13) is shown in Figure 6.

       18.   Waste Management

       18.1   It is the laboratory's responsibility to comply with all federal, State, and local regulations governing
             waste management, particularly the hazardous waste identification rules and land disposal restric-
            tions.  In addition, it is the laboratory's responsibility to protect air, water, and land resources by min-
            imizing and controlling all releases from fume hoods and bench operations. Also, compliance is
            required with any sewage discharge permits and regulations.
   24

-------
                                                                                           Method 1865
18.2  Samples containing acids at a pH of less than 2 are hazardous and must be neutralized before being
     poured down a drain or must be handled as hazardous waste.
18.3  For further information on waste management, consult "The Waste Management Manual for Labora-
     tory Personnel," available from the American Chemical Society's Department of Government Rela-
     tions and Science Policy, 1115 Sixteenth Street, NW, Washington, DC 20036.
                                                                                               25

-------
Method 1665
                                              References

       1.    "Performance Tests for the Evaluation of Computerized Gas Chromatography/Mass Spectrometry
            Equipment and Laboratories," U.S. EPA, EMSL Cincinnati, Ohio 45268, EPA-600/4-80-025 (April
            1980).

       2.    National Standard Reference Data System, "Mass Spectral Tape Format," U.S. National Bureau of
            Standards (1979 and later attachments).

       3.    "Working with Carcinogens," DREW, PHS, CDC, NIOSH, Publication 77-206 (1977).

       4.    "OSHA Safety and Health Standards, General Industry," 29 CFR 1910, OSHA 2206 (1976).

       5.    "Safety in Academic Chemistry Laboratories," American Chemical Society Publication, Committee
            on Chemical Safety (1979).

       6.    "Interlaboratory Validation of U. S. Environmental Protection Agency Method 1625A, Addendum
            Report,"  SRI International, Prepared for Analysis and Evaluation Division (WH-557), U.S. EPA,
            Washington DC 20460 (January 1985).

       7.    "Handbook of Analytical Quality Control in Water and Wastewater Laboratories," U.S. EPA, EMSL
            Cincinnati, OH 45268, EPA-600/4-79-019 (March 1979).

       8.    "Standard Practice for Sampling Water," ASTM Annual Book of Standards, ASTM, Philadelphia,
            PA, 76 (1982).

       9.    "Methods 330.4 and 330.5 for Total Residual Chlorine," U.S. EPA, EMSL, Cincinnati, OH 45268,
            EPA 600/4-70-020 (March 1979).

       10.   "Guidance Evaluation, Resolution, and Documentation of Analytical Problems Assoicated with
            Compliance Monitoring." EPA 821-B-93-001. Copies available from: Analytical Methods Staff,
            Engineering and Analysis Division, Office Science and Technology, U.S. Environmental Protection
            Agency, 401 M St., SW, Washington, DC 20460.
   26

-------
                                                                                                       Method 1665
Table 1. Base/Neutral Extractable PMI Analytes Determined by GC/MS Using Isotope Dilution or Internal
Standard Techniques
                                        Pollutant
Labeled Compound
CASRN1
62-53-3
127-19-5
121-6917
68-12-2
109-06-8
110-86-1
EPA-EGD
757
976
1329
785
703
1330
Analog
dr

di,
dr
dr
d
CASRN1
14545-23-4

Not Avail.
4472-41-7
93951-93-0
7291-22-7
EPA-EGD
657

1229
685
603
1230
  PMI Analyte
  Aniline
  N,N-Dimethylacetamide
  N,N-Dimethylaniline
  N,N-Dimethylforrnamide
  2-Methylpyridine (a-picoline)2
  Pyridine
  1 Chemical Abstracts Service Registry Number.
  2 This semivolatile compound may be analyzed by Method 1625B or the method described herein.
                                                                                                           27

-------
Method 1665
         Table 2. Gas Chromatographic Retention Times and Minimum Levels for Base/Neutral Extractable PMI
         Analytes
                                                                    Retention Time
          EGD
          No.1   PMI Analyte
           164   2,2'-Difluorobiphenyl (int std)
          1230   Pyridine-ds
          1330   Pyridine
           685   N,N-Dimethylformamide-d7
           785   N,N-Dimethylformamide
           603   2-Methylpyridine-d7 (a-picoline-d7)
           703   2-Methylpyridine (a-picoline)
           976   N,N-Dimethylacetamide
           657   Aniline-d7
           757   Aniline
          1229   N.N-Dimethylaniline-du
          1329   N,N-Dimethylaniline
          1 Three-digit EGD numbers beginning with 0,1,5, or 9 indicate a pollutant quantified by the internal standard method; begin-
           ning with 2 or 6 indicate a labeled compound quantified by the internal standard method; beginning with 3 or 7 indicate a pol-
           lutant quantified by isotope dilution. The initial "1° in four-digit EGD numbers is to be ignored in applying these rules.
          2 Single values in this column are based on single laboratory data.
          3This is a minimum level at which the entire analytical system shall give recognizable mass spectra (background-corrected) and
           acceptable calibration points, taking into account method-specific sample and injection volumes. The concentration in the
           aqueous or solid phase is determined using the equations in Section 15.
Mean
(sec)
1180
343
345
399
401
449
455
536
702
705
856
863
EGD
Ref
164
164
1230
164
685
164
603
164
164
657
164
1229
Relative2
1.000-1.000
0.291
1.006
0.338
1.005
0.326-0.393
1.006-1.028
0.454
0.595
1.004
0.725
1.008
MinimumLevel3
(M9/L)


5

5

5
50

2

10
    28

-------
                                                                                                    Method 1665
Table 3. DFTPP Mass-Intensity Specifications
                               Mass        Intensity Required
                               51           8-82% of m/z 198
                               68           less than 2% of m/z 69
                               69           11-91% of m/z 198
                               70           less than 2% of m/z 69
                               127         32-59% of m/z 198
                               197         less than 1% of m/z 198
                               198         base peak, 100% abundance
                               199         4-9% of m/z 198
                               275         11-30% of m/z 198
                               441          44-110% of m/z 443
                               442         30-86% of m/z 198
                               443         14-24% of m/z 442
                                                                                                         29

-------
Method 1665
Table 4. Acceptance Criteria for PMI Analytes
EGD
757
657
976
1329
1229
785
685
703
603
1330
1230
PMI Analyte
Aniline
Aniline-d7
N,N-Dimethylacetamide
N,N-Dimethylaniline
N.N-Dimethylaniline-dn
N,N-Dimethylformamide
N,N-Dimethylformamide-d7
2-Methylpyridine
(a-picoline)
2-Methylpyridine-d7
Pyridine
Pyridine-ds
Spike _
Level
(ug/L)
10
100
20
10
100
10
100
10
100
10
100
Initial Precision and
Accuracy ^M
Rnmnnnnri
(H9/L)
1
55
20
5
192
2
19
4
27
3
30
X Recovery
(ug/L) P(%)
8-10
75-184 70-188
d-39
6-15
47-431 31-446
9-13
65-103 64-104
7-15
55-109 53-112
7-12
54-113 52-116
Calibration
Verification
(M9/L)
8-10
75-184
d-39
6-15
47-431
9-13
65-103
7-15
55-109
7-12
54-113
Ongoing
Accuracy
(ug/L)
8-10
70-188
d-41
6-15
31-446
9-13
64-104
7-15
53-112
7-12
52-116
d = Detected (compound must be detected)
    30

-------
                                                                                                           Method 1665
Table 5. Characteristic m/z's and Response Factors for Base/Neutral Extractable PMI Analytes

                                                   Spike
                                                   Level          Labeled                       Response
  PMIAnalyte                                      (jig/L)         Analog       Primary m/z1        Factor2
  Aniline                                              2             d7            93/98             1.22
  N,N-Dimethylacetamide                              50                            44              0.33
  N,N-Dimethylaniline                                  10             dM           120/130           0.99
  N,N-Dmethylformamide                               5             d7            73/80             0.84
  2-Methylpyridine (a-picoline)                           5             d7           93/100            1.15
  Pyridine                                             5             ds            79/84             1.05
  1 Native/labeled
  2 Referenced to 2,2'-difluorobiphenyl
                                                                                                               31

-------
Method 1665
                                                                                 I	I
              10 —
             1.0 —
             0.1  —
                            T
                             2
 I         1          I
10       20        50

 Concentration (pg/mL)
 I        I
100     200
                                         The Dotted Lines Enclose a ±10% Error
                             Figure 1.  Relative Response Calibration Curve for Aniline
                                                                                                         52-025-2
    32

-------
                                                                                            Method 1665
Area at
rrig/z

Area at
m/z
                                                                  Area at
                                                                  rr^/z
  Figure 2.    Extracted Ion-Current Profiles for Chromatographically Resolved
              Labeled (m^z) and Unlabeled (m^z) Pairs
                                                                                    52-025-4
                                                                                                33

-------
Method 1665
                                   Area = 46100
                  (3A)
                          m2/z
                                                                 Area = 4780
                                 Area = 2650
                 (38)
m2/z

rr^/z
                                                                 Area = 43600
                                  Area = 49200
                 (3C)
                         IT12/Z
                                                                Area = 48300
                                                                                                   52-025-5
        Figure 3.    Extracted Ion-Current Profiles for (3A) Unlabeled Compound, (3B) Labeled Compound, and
                    (3C) Equal Mixture of Unlabeled and Labeled Compounds
   34

-------
                                                                                      Method 1665


CO
CO
"ni OR nnn _•
" OJ,UUU ^^
1
3* nnn -
1
0
•5"
0
Q. CO 1
SfS ^^
CC c 1.00 ^
II
i
A nrt — —
6/1
I 1 I 1 I I I I 1
Anthracene-dio
•
• .
• • •
•

I I I i I i I i I
2 3456789 10
Analysis Number
I I I I I I I I I
Anthracene
• • • •
• • • • •

- +3s
_ _ Qc
— —OS

- +3s
_ «3o
OS
I I I I I I I I I
6/1 6/1 6/1 6/2 6/2 6/3 6/3 6/4 6/5
                                        Date Analyzed
                                                                                   52-025-3
Figure 4.    Quality Control Charts Showing Area (top graph) and Relative Response of
            Anthracene to Anthracene-d10 (lower graph) Plotted as Function of Time or
            Analysis Number
                                                                                          35

-------
Method 1665
                                             f

ffl!t:4
c J? » » ~
|||| 3

2
2
F
r
2
F*








t
C
1
i i ' i ' i ' i
                             500                 1,000



                                                Seconds
1,500
2,000
                                                                                       52-025-1
                     Figure 5.  Chromatogram of PMI Analytes Specific to Method 1665
   36

-------
                   Method 1666

Volatile Organic Compounds Specific to the
 Pharmaceutical Manufacturing Industry by
                Isotope Dilution GC/MS
                              January 1995

-------
                                  Method 1666


                  Volatile Organic Compounds Specific to the

                   Pharmaceutical Manufacturing  Industry by

                              Isotope Dilution GC/MS

1.   Scope and Application

1.1   This method is designed to meet the survey and monitoring requirements of the EPA's Engineering
     and Analysis Division (EAD). The method is used to determine certain volatile organic pollutants
     specific to the pharmaceutical manufacturing industry (PMI) that are amenable to purge-and-trap gas
     chromatography/mass spectrometry (GC/MS) or direct aqueous injection GC/MS.
1.2   The PMI analytes listed in Tables 1 and 2 may be determined in waters, soils, and municipal sludges
     by this method or the method referenced.
1.3   The detection limits of the method are usually dependent on the level of interferences rather than
     instrumental limitations. The minimum levels (MLs) in Tables 3 and 4 are the level that can be
     attained with no interferences present.
1.4   The GC/MS portions of this method are for use only by analysts experienced with GC/MS or under
     the close supervision of such qualified persons. Laboratories unfamiliar with analysis of environmen-
     tal samples by GC/MS should run the performance tests in Reference 1 before beginning.
1.5   This method is performance-based. The analyst is permitted to modify the method to overcome
     interferences or to lower the cost of measurements, provided that all performance criteria in this
     method are met. The requirements for establishing method equivalency are given in Section 9.1.2.

2.   Summary of  Method

2.1   Purge-and-trap GC/MS.
     Stable, isotopically labeled analogs of the compounds of interest are added to the sample and the
     sample is purged with an inert gas at 45 °C in a chamber designed for soil or water samples, as appro-
     priate. In the purging process, the volatile compounds are transferred from the aqueous phase into
     the vapor phase, where they are passed into a sorbent column and trapped. After purging is com-
     pleted, the trap is backflushed and heated rapidly to desorb the compounds into a gas chromatograph
     (GC). The compounds are separated by the GC and detected by a mass spectrometer (MS) (Refer-
     ences 2 and 3).
2.2   Direct aqueous injection.
     Certain volatile, water-soluble organic compounds do not purge well from water and are analyzed by
     direct aqueous injection.
     2.2.1   The percent solids content of the sample is determined. If the solids content is known or
           determined to be less than 1%, stable, isotopically labeled analogs of the compounds of
           interest are added to a 5-mL sample. If the solids content of the sample is greater than 1 %, 5
           mL of reagent water and the labeled compounds are added to a S-g aliquot of sample. The
           mixture is sonicated in a centrifuge vial with little or no headspace for 5 minutes. During
           this period the native analytes and labeled analogs will equilibrate between the solid and

-------
Method 1666
                    aqueous phases. In some cases, additional sonication will be necessary to establish equilib-
                    rium.  The resulting suspension is centrifuged and the supernatant liquid analyzed.
            2.2.2    One |iL or more of the aqueous solution (or supernate) is injected into the GC/MS system.
                    The compounds are separated by the GC and detected by the mass spectrometer (References
                    2 and 3). The labeled compounds serve to correct the variability of the analytical technique.
       2.3  Identification of a pollutant (qualitative analysis) is performed by calibrating the GC/MS system with
            authentic standards and storing mass spectrum and retention time for each standard in a user-created
            library. A compound is identified when its retention time and mass spectrum agree with the library
            retention time and spectrum.

       2.4  Quantitative analysis is performed in one of two ways in this method by using extracted-ion current
            profile (EICP) areas. (1) For those compounds listed in Table 1 and Table 2, and for other com-
            pounds for which standards and labeled analogs are available, the GC/MS system is calibrated and
            the compound concentration is determined using an isotope dilution technique. (2) For those com-
            pounds listed in Table 1 and Table 2, and for other compounds for which authentic standards but no
            labeled compounds are available, the GC/MS system is calibrated and the compound concentration is
            determined using an internal standard technique.
       2.5  The quality of the analysis is assured through reproducible calibration of the GC/MS system.

       3.   Definitions

       There are no method-specific definitions to be noted for this document.

       4.   Interferences

       4.1  Impurities in the purge gas, organic compounds outgassing from the plumbing upstream of the trap,
            and solvent vapors in the laboratory account for the majority of contamination problems encountered
            with this method. The analytical system is demonstrated to be free from interferences under condi-
            tions of the analysis by analyzing reagent water blanks initially and with each sample batch (samples
            analyzed on the same 8-hour shift), as described in Section 9.5.
       4.2  Samples can be contaminated by diffusion of volatile organic compounds (particularly methylene
            chloride) through the bottle seal during shipment and storage. A field blank prepared from reagent
            water and carried through the sampling and handling protocol may serve as a check on such contam-
            ination.
       4.3  Contamination by carryover can occur when high-level and low-level samples are analyzed sequen-
            tially. To reduce carryover, the purging device (Figure 1 for samples containing less than 1% solids;
            Figure 2 for samples containing 1% solids or greater) in purge-and-trap analysis or the syringe in
            direct aqueous injection analysis is cleaned or replaced with a clean purging device or syringe after
            each sample is analyzed. When an unusually concentrated sample is encountered, it is followed by
            analysis of a reagent water blank to check for carryover. Purging devices and syringes are cleaned by
            washing with soap solution, rinsing with tap and distilled water, and drying in an oven at 100-125 °C.
            The trap  and other parts of the system are also subject to contamination; therefore, frequent bakeout
            and purging of the entire system may be required.
       4.4  Interferences resulting from samples will vary considerably from source to source, depending on the
            diversity of the site being sampled.
   40

-------
                                                                                            Method 1666
5.   Safety

5.1   The toxicity or carcinogenicity of each compound or reagent used in this method has not been pre-
     cisely determined; however, each chemical compound should be treated as a potential health hazard.
     Exposure to these compounds should be reduced to the lowest possible level. The laboratory is
     responsible for maintaining a current awareness file of OSHA regulations regarding the safe han-
     dling of the chemicals specified in this method. A reference file of material safety data sheets should
     also be made available to all personnel involved in these analyses. Additional information on labora-
     tory safety can be found in References 5 through 7.

6.   Equipment and Supplies

6.1   Sample bottles and septa.
     6.1.1    Bottle—25- to 40-mL with screw-cap (Pierce 13075, or equivalent). Detergent wash, rinse
            with tap and distilled water, and dry at > 105 °C for a minimum of 1 hour before use.
     6.1.2   Septum—Polytetrafluoroethylene (FTFE)-faced silicone (Pierce 12722, or equivalent),
            cleaned as above and baked at 100-200 °C for a minimum of 1 hour.
6.2   Purge-and-trap device—Consists of purging device, trap, and desorber.
     6.2.1    Purging devices for water and soil samples.
            6.2.1.1   Purging device for water samples—Designed to accept 5-mL samples with water
                     column at least 3  cm deep. The volume of the gaseous head space between the
                     water and trap shall be less than 15 mL. The purge gas shall be introduced less
                     than 5 mm from the base of the water column and shall pass through the water as
                     bubbles with a diameter less than 3 mm.  The purging device shown in Figure 1
                     meets these criteria.
            6.2.1.2   Purging device for solid samples—Designed to accept 5 g of solids plus 5 mL of
                     water. The volume of the gaseous head space between the water and trap shall be
                     less than 25 mL. The purge gas shall be introduced less than 5 mm from the base of
                     the sample and shall pass through the water as bubbles with a diameter less than 3
                     mm.  The purging device shall be capable of being controlled at a temperature  of
                     45 °C (±2°C) while the sample is being purged. The purging device shown in Fig-
                     ure 2 meets these criteria.
     6.2.2    Trap—25-30 cm long x 2.5 mm i.d. minimum, containing the following:
            6.2.2.1   Methyl silicone packing—1 cm (±0.2 cm), 3% OV-1 on 60/80 mesh Chromosorb
                     W, or equivalent.
            6.2.2.2   Porous polymer—15 cm (±1.0 cm), Tenax GC (2,6-diphenylene oxide polymer),
                     60/80 mesh, chromatographic grade, or equivalent.
            6.2.2.3   Silica gel—8 cm (±1.0 cm), Davison Chemical, 35/60 mesh, grade 15, or equiva-
                     lent.  The trap shown in Figure 3 meets these specifications.
     6.2.3    Desorber—Shall heat the trap to 175 °C (±5 °C) in 45 seconds or less. The polymer section of
            the trap shall not exceed a temperature of 180 °C and the remaining sections shall not exceed
            220 °C during desorb, and no portion of the trap shall exceed 225 °C during bakeout. The des-
            orber shown in Figure 3 meets these specifications.
                                                                                                41

-------
Method 1666
            6.2.4    The purge-and-trap device may be a separate unit or coupled to a GC, as shown in Figures 4
                    and 5.

       6.3  Gas chromatograph—Shall be linearly temperature-programmable with initial and final holds, and
            shall produce results that meet the calibration (Section 10), quality assurance (Section 9), and perfor-
            mance tests (Section 15) of this method.

            6.3.1    Column for purge-and-trap analyses—60 m long x 0.32 mm i.d. fused-silica microbore col-
                    umn coated with 1.5 pm of phenylmethyl polysiloxane (Restek RTX-Volatiles, or equiva-
                    lent).

            6.3.2    Column for direct aqueous injection analyses—30 m long x 0.32 mm i.d. fused-silica
                    microbore column coated with 1.5 um of 95% dimethyl- 5% diphenyl polysiloxene specially
                    passivated for chromatography of amines (Reslek RTX-5 Amine, or equivalent).
            6.3.3    GC operating conditions.

                    6.3.3.1   Purge-and-trap—4 minutes at 0 
-------
                                                                                            Method 1666
6.9   Syringe—5-mL, gas-tight, with shut-off valve.
6.10  Bottles—15-mL, screw-cap with PTFE liner.
6.11  Balances.
     6.11.1  Analytical, capable of weighing 0.1 mg.
     6.11.2  Top-loading, capable of weighing 10 mg.
6.12  Equipment for determining percent moisture.
     6.12.1  Oven, capable of being temperature-controlled at 110 °C (±5 °C).
     6.12.2  Desiccator.
     6.12.3  Beakers—50- to 100-mL.
6.13  Centrifuge apparatus.
     6.13.1  Centrifuge capable of rotating 10-mL centrifuge tubes at 5000 rpm.
     6.13.2  Centrifuge tubes, 10-mL, with screw-caps to fit centrifuge.
6.14  Sonication apparatus capable of sonicating 10 mL centrifuge tubes and thoroughly agitating contents.

7.   Reagents and Standards
7.1   Reagent water—Water in which the compounds of interest and interfering compounds are not
     detected by this method. It may be generated by any of the following methods.
     7.1.1    Activated carbon—Pass tap water through a carbon bed (Calgon Filtrasorb-300, or equiva-
            lent).
     7.12   Water purifier—Pass  tap water through a purifier (Millipore Super Q, or equivalent).
     7.1.3   Boil and purge—Heat tap water to 90-100 °C and bubble contaminant-free inert gas through
            it for approximately 1 hour. While still hot, transfer the water to screw-cap bottles and seal
            with a PTFE-lined cap.
7.2   Sodium thiosulfate—ACS granular.
7.3   Methanol—Pesticide-quality  or equivalent.
7.4   Standard solutions—Purchased as solutions or mixtures with certification to their purity, concentra-
     tion, and authenticity, or prepared from materials of known purity and composition. If compound
     purity is 96% or greater, the weight may be used without correction to calculate the concentration of
     the standard.
7.5   Preparation of stock solutions for purge-and-trap analysis.
     7.5.1    Place approximately 9.5 mL of methanol in a 10-mL ground-glass-stoppered volumetric
            flask. Allow the flask to stand unstoppered for approximately 10 minutes or until all metha-
            nol-wetted surfaces have dried. In each case, weigh the stoppered flask, add the compound,
            restopper, then immediately reweigh to prevent evaporation losses from affecting the mea-
            surement.
     7.5.2   Using a 100-uL syringe, permit two drops of liquid to fall into the methanol without contact-
            ing the neck of the flask. Alternatively, inject a known volume of the compound into the
            methanol in the flask using a microsyringe.
                                                                                                 43

-------
Method 1666
            7.5.3    Fill the flask to volume, stopper, then mix by inverting several times. Calculate the concen-
                    tration in milligrams per milliliter (mg/mL; equivalent to micrograms per microliter [ug/uL])
                    from the weight gain.

            7.5.4    Transfer the stock solution to a FTFE-sealed screw-cap bottle. Store, with minimal head-
                    space, in the dark at -10 to -20 °C.

            7.5.5    Replace standards after one month, or sooner if comparison with check standards indicate a
                    change in concentration.  Quality control check standards that can be used to determine the
                    accuracy of calibration standards may be available from the National Institute of Standards
                    and Technology, Gaithersburg, Maryland.

       7.6  Preparation of stock solutions for direct aqueous injection analysis.
            7.6.1    Place approximately 9.0 mL of reagent water in a 10-mL ground-glass-stoppered volumetric
                    flask. Allow the flask to stand unstoppered for approximately 10 minutes or until all wetted
                    surfaces have dried. In each case, weigh the stoppered flask, add the compound, restopper,
                    then immediately reweigh to prevent evaporation losses from affecting the measurement.
            7.6.2    Using a microsyringe, add sufficient liquid (about 100 mg) so that the final solution will
                    have a concentration of about 10 mg/mL.
            7.6.3    Fill the flask to volume, stopper, then mix by inverting several times. Calculate the concen-
                    tration in milligrams per milliliter (mg/mL; equivalent to micrograms per microliter [ug/uL])
                    from the weight gain.
            7.6.4    Transfer the stock solution to a PTFE-sealed screw-cap bottle. Store, with minimal head-
                    space, in the dark at approximately 4*C. Do not freeze.
            7.6.5    Replace standards after one month, or sooner if comparison with check standards indicate a
                    change in concentration.  Quality control check standards that can be used to determine the
                    accuracy of calibration standards may be available from the National Institute of Standards
                    and Technology, Gaithersburg, Maryland.
       7.7  Labeled compound spiking  solutions.
            7.7.1    For purge-and-trap analysis—From stock standard solutions (Section 7.5), or from mixtures,
                    prepare the spiking  solution to contain a concentration of labeled compound such that a 5- to
                    10-uL spike into each 5-mL sample, blank, or aqueous standard analyzed will result in a con-
                    centration of 50 ug/L of each compound with an ML of 20 jig/L or less, a concentration of
                    500 ug/L for each compound with an ML of 100 or 200 ug/L, and a concentration of 1 mg/L
                    for each compound with an ML of 500 ug/L (see Table 3).  Include the internal standards
                    (Section 10.4.2) in this solution, if appropriate, so that a concentration of 50 ug/L in each
                    sample, blank, or aqueous standard will be produced.
            7.7.2    For direct aqueous injection—From stock standard solutions (Section 7.6), or from mixtures,
                    prepare the spiking  solution to contain a concentration such that a 50- to 100-uL spike into
                    each sample, blank, or aqueous standard analyzed will result in a concentration of 1 mg/mL
                    of each labeled compound. Include the internal standard in this solution so that a concentra-
                    tion of 1 mg/mL will be produced.
       7.8   Secondary standards—Using stock solutions, prepare a secondary standard in methanol or water, as
            appropriate, to contain each pollutant at a concentration of 1 mg/mL, or 2.5 mg/mL for compounds
            with higher MLs.
            7.8.1    Aqueous calibration standards—Using a microsyringe, add sufficient secondary standard
                    (Section 7.8) to five reagent water aliquots to produce concentrations in the range of interest.
   44

-------
                                                                                            Method 1666
     7.8.2   Aqueous performance standard—An aqueous standard containing all pollutants, internal
            standards, labeled compounds, and BFB is prepared daily, and analyzed each shift to demon-
            strate performance (Section 15). This standard shall contain concentrations of pollutants,
            labeled compounds, BFB, and internal standards, as appropriate, within a factor of 1-5 times
            the MLs of the pollutants listed in Table 3 or 4. It may be one of the aqueous calibration
            standards described in Section 7.8.1.

     7.8.3   A methanolic standard containing all pollutants specific to this method (Table 1) and internal
            standards is prepared to demonstrate recovery of these compounds when syringe injection
            and purge-and-trap analyses are compared. This standard shall contain either 100 ng/mL or
            500 ug/mL of the PMI analytes, and 100 ug/mL of the internal standards (consistent with the
            amounts in the aqueous performance standard in Section 7.8.2).
     7.8.4   Other standards that may be needed are those for test of BFB performance (Section 10.1) and
            for collection of mass spectra for storage in spectral libraries (Section 10.1.1).

8.   Sample Collection, Preservation, and Handling

8.1   Grab samples are collected in glass containers having a total volume greater than 20 mL.  For aque-
     ous samples that pour freely, fill sample bottles so that no air bubbles pass through the sample as the
     bottle is filled and seal each bottle so that no air bubbles are entrapped.  Maintain the hermetic seal on
     the sample bottle until time of analysis.
8.2   Samples are maintained at 0-4 
-------
Method 1666
             9.1.2   In recognition of advances that are occurring in analytical technology, and to allow the ana-
                    lyst to overcome sample matrix interferences, the analyst is permitted certain options to
                    improve separations or lower the costs of measurements. These options include alternative
                    concentration and cleanup procedures, and changes in columns and detectors.  Alternative
                    techniques, such as the substitution of spectroscopy or immunoassay, and changes that
                    degrade method performance are not allowed.  If an analytical technique other than the tech-
                    niques specified in this method is used, that technique must have a specificity equal to or bet-
                    ter than the specificity of the techniques in this method for the analytes  of interest.
                    9.1.2.1    Each time a modification is made to this method, the analyst is required to repeat
                             the procedure in Section 9.2. If the detection limit of the method will be affected
                             by the change, the laboratory is required to demonstrate that the MDL (40 CFR
                             Part 136, Appendix B) is lower than one-third the regulatory compliance level.  If
                             calibration will be affected by the change, the analyst must recalibrate the instru-
                             ment per Section 10.
                    9.1.2.2    The laboratory is required to maintain records of modifications made to this
                             method. These records include the information below, at a minimum.
                             9.1.2.2.1   The names, titles, addresses, and telephone numbers of the analyst(s)
                                       who performed the analyses and modification, and of the quality con-
                                       trol officer who witnessed and will verify the analyses and modifica-
                                       tion.
                             9.1.2.2.2   A listing of pollutant(s) measured, by name and CAS Registry Number.
                             9.1.2.2.3   A narrative stating the reason(s) for the modification.
                             9.1.2.2.4   Results from all quality control (QC) tests comparing the modified
                                       method to this method,  including:
                                       (a)  Calibration (Section 10)
                                       (b)  Calibration verification (Section IS)
                                       (c)  Initial precision and accuracy (Section 9.2)
                                       (d)  Labeled compound recovery (Section 9.3)
                                       (e)  Analysis of blanks (Section 9.5)
                                       (f)  Accuracy assessment (Section 9.4)
                    9.1.2.3    Data that will allow an independent reviewer to validate each determination by
                             tracing the instrument output (peak height, area, or other signal) to the final result,
                             including:
                             (a)  Sample numbers and other identifiers
                             (b)  Analysis dates and times

                             (c)  Analysis sequence/run chronology
                             (d)  Injection logs
                             (e)  Sample weight or volume

                             (f)  Sample volume prior to each cleanup step, if applicable
                             (g) Sample volume after each cleanup step, if applicable
                             (h) Final sample volume prior to injection (Sections 11 and 12)
   46

-------
                                                                                              Method 1666
                      (i)  Injection volume (Sections 11 and 12)
                      (j)  Dilution data, differentiating between dilution of a sample or an extract (Sec-
                         tion 16.4)

                      (k) Instrument and operating conditions
                      (I)  Column (dimensions, liquid phase, solid support, film thickness, etc.)
                      (m) Operating conditions (temperature, temperature program, flow rates, etc.)
                      (n) Detector (type, operating condition, etc.)
                      (o) Chromatograms, printer tapes, and other recording of raw data
                      (p) Quantitation reports, data system outputs, and other data necessary to link raw
                         data to the results reported.
     9.1.3   Analyses of blanks are required to demonstrate freedom from contamination and that the
            compounds of interest and interfering compounds have not been carried over from a previ-
            ous analysis (Section 4.3). The procedures and criteria for analysis of a blank are  described
            in Section 9.5.

     9.1.4   The laboratory shall spike all samples with labeled compounds to monitor method perfor-
            mance. This test is described in Section 9.3.  When results of these spikes indicate atypical
            method performance for samples, the samples are diluted to bring method performance
            within acceptable limits (Section 16).
     9.1.5   The laboratory shall, on an ongoing basis, demonstrate through the analysis of the aqueous
            performance standard (Section 7.8.2) that the analysis system is in control. This procedure is
            described in Sections 15.1 and 15.5.
     9.1.6   The laboratory shall maintain records to define the quality of data that is generated. Devel-
            opment of accuracy statements is described in Sections 9.4 and 15.5.2.
9.2   Initial precision and accuracy—To establish the ability to generate acceptable precision and accuracy,
     the analyst shall perform the following operations for compounds to be calibrated:
     9.2.1   Analyze two sets of four 5-mL aliquots (eight aliquots total) of the aqueous performance
            standard (Section 7.8.2) containing Table 1 PMI analytes by purge-and-trap. Or, for Table 2
            PMI analytes, analyze two sets of four aliquots (eight aliquots total) by direct aqueous injec-
            tion.
     9.2.2   Using results of the first set of four analyses in Section 9.2.1, compute the average recovery
            (X) in percent of spike level and the standard deviation of the recovery (s) in percent of spike
            level, for each compound, by isotope dilution for pollutants with a labeled analog, and by
            internal standard for labeled compounds and pollutants with no labeled analog.
     9.2.3   For each compound, compare s and X with the corresponding limits for initial precision and
            accuracy found in Table 6. If s and X for all compounds meet the acceptance criteria, system
            performance is acceptable and analysis of blanks and samples may begin. If, however, any
            individual s exceeds the precision limit or any individual X falls outside the range for accu-
            racy, system performance is unacceptable for that compound.
                                                                                                  47

-------
Method 1666
       Note:   The large number of compounds in Table 6 presents a substantial probability that one or more will
               fail one of the acceptance criteria when all compounds are analyzed. To determine if the analyti-
               cal system is out of control, or if the failure can be attributed to probability, proceed as follows.


            9.2.4   Using the results of the second set of four analyses, compute s and X for only those com-
                   pounds that failed the test of the first set of four analyses (Section 9.2.3). If these compounds
                   now pass, system performance is acceptable for all compounds, and analysis of blanks and
                   samples may begin. If, however, any of the same compounds fail again, the analysis system
                   is not performing properly for the compound(s) in question. In this event, correct the prob-
                   lem and repeat the entire test (Section 9.2.1).
       9.3  The laboratory shall spike all samples with labeled compounds to assess method performance on the
            sample matrix.
            9.3.1   Spike and analyze each sample according to the appropriate method in Section 11 or 12.
            9.3.2   Compute the percent recovery (P) of the labeled compounds using the internal standard
                   method (Section 10.4.2).
            9.3.3   Compare the percent recovery for each compound with the corresponding labeled compound
                   recovery limit in Table 6. If the recovery of any compound falls outside its warning limit,
                   method performance is unacceptable for that compound in that sample. Therefore, the sam-
                   ple matrix is complex and the sample is to be diluted and reanalyzed, per Section 16.
       9.4  As part of the QA program for the laboratory, method accuracy for wastewater samples shall be
            assessed and records shall be maintained. After the analysis of five wastewater samples for which
            the labeled compounds pass the tests in Section 9.3.3, compute the average percent recovery (P) and
            the standard deviation of the percent recovery (sp) for the labeled compounds only. Express the accu-
            racy assessment as a percent recovery interval from P - 2sp to P + 2sp. For example, if P = 90% and
            sp = 10%, the accuracy interval is expressed as 70-110%. Update the accuracy assessment for each
            compound on a regular basis (e.g., after each five to ten new accuracy measurements).
       9.5  Blanks—Reagent water blanks are analyzed to demonstrate freedom from carryover and contamina-
            tion (Section 4).
            9.5.1   The level at which the purge-and-trap system will carry greater than the ML of a pollutant of
                   interest (Table 1) into a succeeding blank shall be determined by analyzing successively
                   larger concentrations of these compounds. When a sample contains this concentration or
                   more, a blank shall be analyzed immediately following this sample to demonstrate no carry-
                   over at the ML.
            9.5.2   With each sample batch (samples analyzed on the same 8-hour shift), a blank shall be ana-
                   lyzed immediately after analysis of the aqueous performance standard (Section 15.1) to dem-
                   onstrate freedom from contamination. If any of the compounds of interest (Table 1 or 2) or
                   any potentially interfering compound is found in  a blank at greater than the ML (assuming a
                   response factor of 1 relative to the nearest-eluted internal standard for compounds not listed
                   in Tables 1 and 2), analysis of samples is halted until the source of contamination is elimi-
                   nated and  a blank shows no evidence of contamination at this level.
       9.6  The specifications contained in this method can be met if the apparatus used is calibrated properly,
            then maintained in a calibrated state. The standards used for calibration (Section 7), calibration veri-
            fication (Section 15.5), and initial (Section 9.2) and ongoing (Section 15.5) precision and accuracy
            should be identical, so that the most precise results will be obtained.  The GC/MS instrument in par-
   48

-------
                                                                                             Method 1666
     ticular will provide the most reproducible results if dedicated to the settings and conditions required
     for the analyses of volatiles by this method.
9.7  Depending on specific program requirements, field replicates may be collected to determine the pre-
     cision of the sampling technique, and spiked samples may be required to determine the accuracy of
     the analysis when the internal-standard method is used.

10.  Calibration

Calibration of the GC/MS system is performed by direct aqueous injection (Section 10.3) or purging the
compounds of interest and their labeled analogs from reagent water at the temperature to be used for anal-
ysis of samples (Section 10.2).
10.1  Assemble the GC/MS apparatus and establish the operating conditions to be used for sample analysis
     (Section 6.3.3.1 or Section 6.3.3.2). By injecting standards into the GC, demonstrate that the analyt-
     ical system meets the minimum levels in Tables  3 or 4 for the compounds for which calibration is to
     be performed, and the  mass-intensity criteria in Table 5 for 50 ng BFB.
     10.1.1   Mass-spectral libraries—Detection and identification of the compounds of interest are
             dependent upon the spectra stored in user-created libraries.
             10.1.1.1   For the compounds in Tables 1 and 2, and other compounds for which the GC/MS
                     is to be calibrated, obtain a mass  spectrum of each pollutant and labeled compound
                     and each internal standard by analyzing an authentic standard either singly or as
                     part  of a mixture in which there is no interference between closely eluted compo-
                     nents. Examine the spectrum to determine that only a single compound is present.
                     Fragments not attributable to the compound under study indicate the presence of an
                     interfering compound. Adjust the analytical conditionsand scan rate (for this test
                     only) to produce an undistorted spectrum at the GC peak maximum. An undis-
                     torted spectrum will usually be obtained if five complete spectra are collected
                     across the upper half of the GC peak. Software algorithms designed to "enhance"
                     the spectrum may eliminate distortion, but may also eliminate authentic m/z's or
                     introduce other distortion.
             10.1.1.2  The authentic reference spectrum is obtained under BFB tuning conditions (Sec-
                     tion  10.1 and Table 5) to normalize it to spectra from other instruments.
             10.1.1.3  The spectrum is edited by saving the five most intense mass-spectral peaks and all
                     other mass-spectral peaks greater than 10% of the base peak. The spectrum may
                     be further edited to remove common interfering masses. If five mass-spectral
                     peaks cannot be obtained under the scan conditions given in Section 6.4, the mass
                     spectrometer may be scanned to  an m/z lower than 20 to gain additional spectral
                     information. The spectrum obtained is stored for reverse search and for compound
                     confirmation.
10.2  Assemble the GC/MS apparatus and establish operating conditions given in Section 6.3.3.1. By
     injecting standards into the GC, demonstrate that the analytical system meets the minimum levels in
     Table 3 for the compounds for which  calibration is  to be performed, and the mass-intensity criteria in
     Table 5 for 50 ng BFB.
     10.2.1   Assemble the purge-and-trap device. Pack the trap as shown in Figure 3 and  condition over-
             night at 170-180 "€ by backflushing with an inert gas at a flow rate of 20-30 mL/min.  Con-
             dition traps daily for a minimum of 10 minutes prior to use.
                                                                                                 49

-------
Method 1666
                    10.2.1.1  Analyze the aqueous performance standard (Section 7.8.2) according to the purge-
                            and-trap procedure in Section 11. Compute the area at the primary m/z (Table 7)
                            for each compound.  Compare these areas to those obtained by injecting 1 |jL of
                            the methanolic standard (Section 7.5.1) to determine compound recovery.  The
                            recovery shall be greater than 50% for the PMI analytes. Maximum allowable
                            recovery for the PMI analytes found in Table 1 are shown in Table 8. This recov-
                            ery is demonstrated initially for each purge-and-trap GC/MS system. The test is
                            repeated only if the purge-and-trap or GC/MS systems are modified in any way
                            that might result in a change in recovery.

                    10.2.1.2  Demonstrate that 100 ng o-xylene (or o-xylene-d10) produces an area at m/z 106 (or
                            116) approximately one-tenth that required to exceed the linear range of the sys-
                            tem. The exact value must be determined by experience for each instrument. It is
                            used to match the calibration range of the instrument to the analytical range and
                            detection limits required.
       10.3 Assemble the GC/MS system for direct aqueous injection and establish the operating conditions to be
            used for sample analysis (Section 6.3.3.2). By injecting standards into the GC, demonstrate that the
            analytical system meets the minimum levels in Table 4 for the compounds for which calibration is to
            be performed, and the mass-intensity criteria in Table 5 for 50 ng BFB.
            Demonstrate that 100 ng o-xylene (or o-xylene-d10) produces an area at m/z 106 (or 116) approxi-
            mately one-tenth that required to exceed the linear range of the system.  The exact value must be
            determined by experience for each instrument. It is used to match the calibration range of the instru-
            ment to the analytical range and detection limits required.
       10.4 The following calibration steps are to be performed for both the purge-and-trap PMI analytes found
            in Table 1 and the direct aqueous injection PMI analytes found in Table 2, as appropriate.
            10.4.1   Calibration by isotope dilution—The isotope dilution approach is  used for the PMI analytes
                    when appropriate labeled compounds are available and when interferences do not preclude
                    the analysis.  If labeled compounds are not available, or interferences are present, the internal
                    standard method (Section 10.4.2) is used.  A calibration curve encompassing the concentra-
                    tion range of interest is prepared for each compound determined. The relative response (RR)
                    vs. concentration in micrograms per liter is plotted or computed using a linear regression.
                    An example of a calibration curve for o-xylene using o-xylene-dlo is given in Figure 6. Also
                    shown are the ±10% error limits (dotted lines). Relative response is determined according to
                    the procedures described below. A minimum of five data points are required for calibration.
                    10.4.1.1  The relative response (RR) of pollutant to labeled compound is determined from
                            isotope ratio values calculated from acquired data. Three isotope ratios are used in
                            this process:
                             Rx = The isotope ratio measured in the pure pollutant (Figure  7A)

                             Ry = The isotope ratio of pure labeled compound (Figure 7B)

                             Rm= The isotope ratio measured in the analytical mixture of the
                                   pollutant and labeled compounds (Figure 7C)
   50

-------
        	Method 1666


         The correct way to calculate RR is:
                                m
         If Rn, is not between 2Ry and 0.5R*, the method does not apply and the sample is
         analyzed by the internal standard method (Section 10.4.2).
10.4.1.2  In most cases, the retention times of the pollutant and labeled compound are simi-
         lar, and isotope ratios (R's) can be calculated from the EICP areas, where:

                               (area at rrij/z)
                               (area at m2/z)

         If either of the areas is zero, it is assigned a value of 1 in the calculations; that is, if
         area of nij/z = 50721, and area of m2/z = 0, then R = 50721/1 = 50720.
         The data from these analyses are reported to three significant figures (see Section
         14.6). In order to prevent rounding errors from affecting the values to be reported,
         all calculations performed prior to the final determination of concentrations should
         be carried out using at least four significant figures. Therefore, the calculation of R
         above is rounded to four significant figures.
         The m/z's are always selected such that R* > Ry.  When there is a difference in
         retention times (RT) between the pollutant and labeled compounds, special precau-
         tions are required to determine the isotope ratios.
         R» Ry, and Rm are defined as follows:
                              area m,
                         R =
                          x
                          R =
                           y
area mJ^ at /?T_ J
                              area m,
                         RnT
areamJzlatRTj]
                                                                                     51

-------
Method 1666
                    10.4.1.3   An example of the above calculations can be taken from the data plotted in Figure
                             7 for o-xylene and oxylene-d,0. For these data:

                                                      . ,68900
                                                  i


                                          V 6dm= 0.00001640

                                                96868   , ,„
                                          RnT  82508= L174

                    10.4.1 .4   The RR for the above data is then calculated using the equation given in Section
                             10.4.1.1. For the example, rounded to four significant figures, RR = 1.174. Not all
                             labeled compounds elute before their pollutant analogs.
                             To calibrate the analytical system by isotope dilution, analyze an aliquot of each of
                             the aqueous calibration standards (Section 7.8.1) spiked with an appropriate con-
                             stant amount of the labeled compound spiking solution (Section 7.7), using the
                             appropriate procedure in Section 10. Compute the RR at each concentration.
                    10.4.1 .5   Linearity — If the ratio of relative response to concentration for any compound is
                             constant (less than 20% coefficient of variation) over the five-point calibration
                             range, an averaged relative response/concentration ratio may be used for that com-
                             pound; otherwise, the complete calibration curve for that compound shall be used
                             over the five-point calibration range.
            10.4.2   Calibration by internal standard — Used when criteria for isotope dilution (Section 10.4. 1)
                    cannot be met. The method is applied to pollutants having no labeled analog and to the
                    labeled compounds.  The internal standards used for volatiles analyses are bromochlo-
                    romethane, 1,4-difluorobenzene, chlorobenzene-ds, and tetrahydrofuran-dg. Concentrations
                    of the labeled compounds and pollutants without labeled analogs are computed relative to
                    the nearest eluting internal standard, as shown in Tables 3 and 4.
                    10.4.2.1   Response factors — Calibration requires the determination of response factors (RF)
                             which are defined by the following equation:
                                              RF =
                             Where:
                             As = The EICP area at the characteristic m/zfor the compound
                                   in the daily standard
                             AJS = The EICP area at the characteristic m/zfor the internal
                                   standard
                             C{s= The concentration ( \ng/L) of the internal standard

                             Cs = The concentration of the pollutant in the daily standard
   52

-------
                                                                                            Method 1666
            10.4.2.2  The RF is determined at 10, 20, 50,100, and 200 ug/L for the pollutants (option-
                     ally at 5 times or more these concentrations for highly water-soluble pollutants; see
                     Section 7.8), in a way analogous to that for calibration by isotope dilution (Section
                     10.4.1). The RF is plotted against concentration for each compound in the stan-
                     dard (Cs) to produce a calibration curve.
            10.4.2.3  Linearity—If the RF for any compound is constant (less than 35% coefficient of
                     variation) over the five-point calibration range, an averaged RF may be used for
                     that compound; otherwise, the complete calibration curve for that compound shall
                     be used over the five-point range.
     10.4.3  Combined calibration—By adding the isotopically labeled compounds and internal standards
            (Section 7.7) to the aqueous  calibration standards (Section 7.8.1), a single set of analyses can
            be used to produce calibration curves for the isotope-dilution and internal-standard methods.
            These curves are verified each shift (Section 15.5) by analyzing the aqueous performance
            standard (Section 7.8.2). Recalibration is required only if calibration and ongoing perfor-
            mance (Section 15.5) criteria cannot be met.

11.  Purge, Trap, and GC/MS Analysis

Samples containing less than 1% solids are analyzed directly as aqueous samples (Section 11.4).  Samples
containing 1% solids or greater are analyzed as  solid samples utilizing one of two methods, depending on
the levels of pollutants in the sample. Samples containing 1% solids or greater and low to moderate levels
of pollutants are analyzed by purging a known weight of sample added to 5 mL of reagent water (Section
11.5). Samples containing 1% solids or greater and high levels of pollutants are extracted with methanol,
and an aliquot of the methanol extract is added to reagent water and purged (Section 11.6).
11.1  Determination of percent solids.
     11.1.1   Weigh 5 to 10 g of sample into a tared beaker.
     11.1.2  Dry overnight (12 hours minimum) at 110 °C (±5 °Q, and cool in a desiccator.
     11.1.3  Determine percent solids as  follows:

                              „   ...    weight of sample dry
                              % solids = —rr-^	—,—- x 100
                                        weight of sample wet

11.2  Remove standards and samples from cold storage and bring to 20-25 °C.
11.3  Adjust the purge gas flow rate to 40 mL/min (±4 mL/min).
11.4  Samples containing less than 1% solids.
     11.4.1   Mix the sample by shaking vigorously. Remove the plunger from a 5-mL syringe and attach
            a closed syringe valve.  Open the  sample bottle and carefully pour the sample into the
            syringe barrel until it overflows. Replace the plunger and compress the sample. Open the
            syringe valve and vent any residual air while adjusting the sample volume to 5 mL (±0.1
            mL).  Because this process of taking an aliquot destroys the validity of the sample for future
            analysis, fill a second syringe at this time to protect against possible loss of data.
     11.4.2  Add an appropriate amount  of the labeled compound spiking solution (Section 7.7.1)
            through the valve bore, then close the valve.
                                                                                                 53

-------
Method 1666
            11.4.3   Attach the syringe valve assembly to the syringe valve on the purging device. Open both
                    syringe valves and inject the sample into the purging chamber. Purge the sample per Sec-
                    tion 11.7.

       11.5 Samples containing 1% solids or greater and low to moderate levels of pollutants.
            11.5.1   Mix the sample thoroughly using a clean spatula and remove rocks, twigs, sticks, and other
                    foreign matter.

            11.5.2   Weigh 5 g (±1 g) of sample into a purging vessel (Figure 2). Record the weight to three
                    significant figures.
            11.5.3   Add 5 mL (±0.1 mL) of reagent water to the vessel.
            11.5.4   Using a metal spatula, break up any lumps of sample to disperse the sample in the water.
            11.5.5   Add an appropriate amount of the labeled compound spiking solution (Section 7.7.1) to the
                    sample  in the purge vessel.  Place a cap on the purging vessel and shake vigorously to fur-
                    ther disperse the sample. Attach the purge vessel to the purging device, and purge the sam-
                    ple per  Section 11.7.

       11.6 Samples containing 1 % solids or greater and high levels of pollutants, or samples requiring dilution
            by a factor of more than 100 (see Section 16).
            11.6.1   Mix the sample thoroughly using a clean spatula and remove rocks, sticks, twigs, and other
                    foreign matter.
            11.6.2   Weigh 5 g (±1 g) of sample into a calibrated IS- to 25-mL centrifuge tube. Record the
                    weight of the sample to three significant figures.
            11.6.3   Add 10 mL of methanol to the centrifuge tube. Cap the tube and shake it vigorously for 15
                    to 20 seconds to disperse the sample in the methanol.  Allow the sample to settle  in the
                    tube. If necessary, centrifuge the sample to settle suspended particles.
            11.6.4   Remove approximately 0.1% of the volume of the supernatant methanol using a 15- to 25-
                    (jL syringe. This volume will be in the range of 10-15 uL.
            11.6.5   Add this volume of the methanol extract to 5 mL reagent water in a 5-mL syringe, and ana-
                    lyze per Section 10.4.1.
            11.6.6   For further dilutions, dilute  1 mL of the supernatant methanol (Section 10.6.4) to 10 mL,
                    100 mL, 1000 mL, etc., in reagent  water. Remove a volume of this methanol extract/
                    reagent water mixture equivalent to the volume in Section 10.6.4, add it to 5  mL reagent
                    water in a 5- mL syringe, and analyze per Section 11.4.

       11.7  Purge the sample for 11 minutes (±0.1 minute) at 45 °C (±2 °C).
       11.8  After the 11-minute purge time, attach the trap to the chromatograph and set the purge-and-trap
            apparatus to the desorb mode (Figure 5). Desorb the trapped compounds  into the GC column by
            heating the trap to 170-180 °C while backflushing with carrier gas at 20-60 mL/min for 4 minutes.
            Start MS data acquisition upon start of the desorb cycle, and start the GC  column temperature pro-
            gram 3 minutes  later. Section 6.3.3.1 provides the recommended operating conditions for the gas
            chromatograph.  Table 3 provides the retention times and minimum levels that can be achieved
            under these conditions.  An example of the separations achieved by the column listed is shown in
            Figure 8. Other columns may be used provided the requirements in Section 9 are met.
   54

-------
                                                                                           Method 1666
11.9  After desorbing the sample for 4 minutes, recondition the trap by purging with purge gas while main-
     taining the trap temperature at 170-180 °C. After approximately 7 minutes, turn off the trap heater
     and stop the gas flow through the trap. When cool, the trap is ready for the next sample.
11.10 While analysis of the desorbed compounds proceeds, remove and clean the purge device.  Rinse with
     tap water, clean with detergent and water, rinse with tap and distilled water, and dry for a minimum of
     1 hour in an oven at a temperature greater than 15Q°C.

12.  Direct Aqueous Injection and GC/MS Analysis

Samples containing less than 1% solids are analyzed directly as aqueous samples (Section 12.3).  Samples
containing 1% solids or greater are analyzed after equilibration with reagent water containing labeled PMI
analytes and internal standards (Section 12.4).
12.1  Determine percent solids as in Section 11.1.
12.2  Remove standards and samples from cold storage and bring to 20-25 °C.
12.3  Samples containing less than 1% solids.
     12.3.1  Allow solids to settle and remove 5 mL of sample.
     12.3.2  Add an appropriate amount of the labeled compound spiking solution (Section 7.7.2).
     12.3.3  Inject 1 |iL or more directly into the GC injection port.  The temperature of the injection
            block should be great enough to immediately vaporize the entire sample. An example of the
            separations achieved by the column listed is shown in Figure 9.
12.4  Samples containing 1% solids or greater.
     12.4.1  Mix the sample thoroughly using a clean spatula and remove rocks, twigs, sticks and other
            foreign matter.
     12.4.2  Add 5 g (±1 g) of sample to a 10-mL centrifuge tube. Using a clean metal spatula, break up
            any lumps of sample. Record the sample weight to three significant figures.
     12.4.3  Add an appropriate amount of the labeled compound spiking solution (Section 7.7.2) to the
            sample in the centrifuge tube.

     12.4.4  Add a measured quantity (Y* 0.1 mL) of reagent water  to the tube so as to minimize head-
            space.
     12.4.5  Place a cap on the centrifuge tube leaving little or no headspace. Place the tube in the sonica-
            tor for a minimum of 5 minutes, turning occasionally. For most samples this should be suffi-
            cient time to distribute labeled and native analytes between the solid and aqueous phases and
            to establish equilibrium. Some sample matrices may require more sonication.
     12.4.6  On completion of sonication, centrifuge the sample and inject 1 pJL or more of supemate
            directly into the GC injection port The temperature of the injection block should be great
            enough to immediately vaporize the entire sample.
12.5  Liquid samples containing high solids concentrations, such as sludges or mud, may be weighed into a
     10-mL centrifuge tube, have labeled compound spiking solution (Section 7.7.2) added, and be soni-
     cated as in Section 12.4.5. Centrifugation and injection are to be performed as in Section 12.4.6.
                                                                                                55

-------
Method 1666
       13.   Qualitative Determination

       Identification is accomplished by comparison of data from analysis of a sample or blank with data stored in
       the mass-spectral libraries. For compounds for which the relative retention times and mass spectra are
       known, identification is confirmed per Sections 13.1 and 13.2.
       13.1   A labeled compound or pollutant having no labeled analog (Tables 1 and 2).
             13.1.1   The signals for all characteristic m/z's stored in the spectral library (Section 10.1.1) shall be
                    present and shall maximize within the same two consecutive scans.
             1 3.1 .2   Either ( 1 ) the background corrected EICP areas or (2) the corrected relative intensities of the
                    mass-spectral peaks at the GC peak maximum shall agree within a factor of 2 (0.5-2 times)
                    for all masses stored in the library.
             13.1 .3   The relative retention time shall be within the window established in the initial precision and
                    accuracy study (Section 9.2).
       13.2   Pollutants having a labeled analog (Tables 1 and 2).

             13.2.1   The signals for all characteristic m/z's stored in the spectral library (Section 10.1.1) shall be
                    present and shall maximize within the same two consecutive scans.
             13.2.2   Either (1) the background corrected EICP areas or (2) the corrected relative intensities of the
                    mass-spectral peaks at the GC peak maximum shall agree within a factor of 2 for all masses
                    stored in the spectral library.
             13.2.3   The relative retention time between the pollutant and its labeled analog shall be within the
                    window established in the initial precision and accuracy study (Section 9.2).
       13.3   The m/z's present in the sample mass spectrum that are not present in the reference mass spectrum
             shall be accounted for by contaminant or background ions. If the sample mass spectrum is contami-
             nated, or if identification is ambiguous, an experienced spectrometrist (Section 1.4) is to determine
             the presence or absence of the compound.
       14.  Quantitative Determination

       14.1  Isotope dilution — Because the pollutant and its labeled analog exhibit the same effects upon purging
            and desorption, or equilibration combined with gas chromatography, correction for recovery of the
            pollutant can be made by adding a known amount of a labeled compound to every sample prior to
            purging or equilibration. Relative response (RR) values for sample mixtures are used in conjunction
            with the calibration curves described in Section 10.4.1 to determine concentrations directly, so long
            as labeled compound spiking levels are constant. For the o-xylene example given in Figure 7 (Sec-
            tion 10.4.1.3), RR would be equal to 1.174. For this RR value, the o-xylene calibration curve given in
            Figure 6 indicates a concentration of 31.8 ug/L.
       14.2  Internal standard — For the compounds for which the system was calibrated (Table 1 and Table 2)
            according to Section 10.4.2, use the response factor determined during the calibration to calculate the
            concentration from the equation below, where the terms are as defined in Section 10.4.2.1.


                                                           A  xC.
                                                             s    is
                                         Concentration = -? — ^Tor-
                                                             is
   56

-------
                                                                                             Method 1666
14.3  The concentration of the pollutant in the solid phase of the sample is computed using the concentra-
     tion of the pollutant detected in the aqueous solution, as follows:


                  /-    .  ^       i-j/   n  >       YLx. aqueous cone (lig/L)
                  Concentration in solid (lig/kg)= 	-,	^—,	   ,. .  —=-=
                                      ^6  6/   sample wt. (kg) x percent solids x DF
                      where:
                      "percent solids" is from Section 11.1.3 or Section 12.1
                      Y = volume of water in liters (L)from Section 12.4.4
                      DF = dilution factor (as a decimal number), where necessary
     Where the aqueous concentration is in mg/L, the result will be in mg/kg.
14.4  Sample dilution—If the EICP area at the quantitation m/z exceeds the calibration range of the sys-
     tem, the sample is diluted by successive factors of 10 until the area is within the calibration range. If
     dilution of high-solids  samples by greater than a factor of 100 is required for purge-and-trap analysis,
     then extract the sample with methanol, as described in Section 11.6.
14.5  Dilution of samples containing high concentrations of compounds not in Table 1 or Table 2—When
     any peak in the mass spectrum is saturated, dilute the sample per Section 14.4.
14.6  Report results for all pollutants and labeled compounds found in all standards, blanks, and samples to
     three significant figures. For samples containing less than 1% solids, the units are micrograms or
     milligrams  per liter (ug/L or mg/L); and for undiluted samples containing 1% solids or greater, units
     are micrograms or milligrams per kilogram (ug/kg or mg/kg).
     14.6.1  Results for samples that have been diluted are reported at the least dilute level at which the
            area at the quantitation m/z is within the calibration range (Section 14.4), or at which no m/z
            in the spectrum is saturated  (Section 14.5).  For compounds having a labeled analog, results
            are reported at the least dilute level at which the area at the quantitation m/z is within the cal-
            ibration range  (Section 14.4) and the labeled compound recovery is within the normal range
            for the method (Section 16.2).

15.  System Performance

15.1  At the beginning of each 8-hour shift during which analyses are performed, system calibration and
     performance shall be verified for the pollutants and labeled compounds (Table 1 or Table 2).  For
     these tests,  analysis of the aqueous performance standard (Section 7.8.2) shall be used to verify all
     performance criteria.  Adjustment and/or recalibration (per Section 10) shall be performed until all
     performance criteria are met. Only after all performance criteria are met may blanks and samples be
     analyzed.
15.2  BFB spectrum validity—The criteria in Table 5 shall be met.
15.3  Retention times.
     15.3.1  Purge-and-trap analysis—The absolute retention times of the internal standards shall fall
            within ±30 seconds of the following—bromochloromethane, 954 seconds; 1,4-difluoroben-
            zene, 1052 seconds; chlorobenzene-ds, 1359 seconds. The relative retention times of all pol-
            lutants and labeled compounds shall fall within 5% of the value given in Table 3.
                                                                                                 57

-------
Method 1666
            15.3.2   Direct aqueous injection analysis—The absolute retention time of tetrahydrofuran-dg shall be
                    263 seconds (±30 seconds). The relative retention times of all pollutants and labeled com-
                    pounds shall fall within 10% of the value given in Table 4.
       15.4 GC resolution—The valley height between o-xylene and o-xylene-d10  (at m/z 106 and 116 plotted on
            the same graph) shall be less than 10% of the taller of the two peaks.
       15.5 Calibration verification and ongoing precision and accuracy—Compute the concentration of each
            pollutant (Table 1 or Table 2) by isotope dilution (Section 10.4.1) for those compounds that have
            labeled analogs. Compute the concentration of each pollutant that has no labeled analog by the inter-
            nal standard method (Section 10.4.2). Compute the concentrations of the labeled compounds by the
            internal standard method. These concentrations are computed based on the calibration data deter-
            mined in Section 10.
            15.5.1   For each pollutant and labeled compound, compare the concentration with the corresponding
                    limit for ongoing accuracy in Table 6. If all compounds meet the acceptance criteria, system
                    performance is acceptable and analysis of blanks and samples may continue. If any individ-
                    ual value falls outside the range given, system performance is unacceptable for that com-
                    pound.


       Note:   The large number of compounds in Table 6 present a substantial probability that one or more will
               fail the acceptance criteria when all compounds are analyzed. To determine if the analytical sys-
       	tern is out of control, or if the failure may be attributed to probability, proceed as follows.	


                    15.5.1.1  Analyze a second aliquot of the aqueous performance standard (Section 7.8.2).
                    15.5.1.2  Compute the concentration for only those compounds that failed the first test (Sec-
                            tion 15.5.1). If these compounds now pass, system performance is acceptable for
                            all compounds, and analyses of blanks and samples may proceed. If, however, any
                            of the compounds fail again, the measurement system is not performing properly
                            for these compounds. In this event, locate and correct the problem or recalibrate
                            the system (Section 10), and repeat the entire test (Section 15.1) for all compounds.
            15.5.2   Add results that pass the specification in Section 15.5.1.2 to initial (Section 9.2) and previous
                    ongoing data.  Update QC charts to form a graphic representation of laboratory performance
                    (Figure 8). Develop a statement of accuracy for each pollutant and labeled compound by cal-
                    culating the average percent recovery (R)  and the standard deviation of percent recovery (sr).
                    Express the accuracy as a recovery intervial from R - 2sr to R + 2sr. For example, if
                    R = 95% and sr = 5%, the accuracy is 85-105%.

       16.  Analysis of Complex Samples

       16.1  Some samples may contain high levels (>1000 ug/kg) of the compounds of interest and of interfering
            compounds. Some samples will foam excessively when purged. Others will overload the trap or the
            GC column.
       16.2  When the recovery of any labeled compound is outside the range given in Table 6, dilute samples by
            a factor of 10 with reagent water and analyze this diluted sample. If the recovery remains outside of
            the range for this diluted sample, the aqueous performance standard shall be analyzed (Section 15.1)
            and calibration verified (Section 15.5). If the recovery for the labeled compound in the aqueous per-
            formance standard is outside the range given in Table 6,  the analytical system  is out of control. In
   58

-------
                                                                                            Method 1666
     this case, the instrument shall be repaired, the performance specifications in Section 15 shall be met,
     and the analysis of the undiluted sample shall be repeated. If the recovery for the aqueous perfor-
     mance standard is within the range given in Table 6, then the method does not apply to the sample
     being analyzed, and the result may not be reported for regulatory compliance purposes.
16.3  When a high level of the pollutant is present, reverse-search computer programs may misinterpret the
     spectrum of chromatographically unresolved pollutant and labeled compound pairs with overlapping
     spectra. Examine each chromatogram for peaks greater than the height of the internal standard
     peaks. These peaks can obscure the compounds of interest.

17.  Method Performance

17.1  This method was developed and validated in a single laboratory.
17.2  Chromatograms of the aqueous performance standards (Sections 7.8.2 and 15.1) are shown in Fig-
     ures 8 and 9.

18.  Waste Management

18.1  It is the laboratory's responsibility to comply with all federal, State, and local regulations governing
     waste management, particularly the hazardous waste identification rules and land-disposal restric-
     tions. In addition it is the laboratory's responsibility to protect air, water, and land resources by min-
     imizing and controlling all releases from fume hoods and bench operations.  Also, compliance is
     required with any sewage discharge permits and regulations.
18.2  Samples containing acids at a pH of less than 2 are hazardous and must be neutralized before being
     poured down a drain or must be handled as hazardous waste.
18.3  For further information on waste management, consult "The Waste Management Manual for Labora-
     tory Personnel," available from the American Chemical Society's Department of Government Rela-
     tions and Science Policy, 1115 16th Street, NW, Washington, DC 20036.

-------
Method 1666
                                              References

       1.    "Performance Tests for the Evaluation of Computerized Gas Chromatography/Mass Spectrometry
            Equipment and Laboratories," U.S. EPA, EMSL Cincinnati, OH 45268, EPA-600/4-80-025 (April
            1980).

       2.    Bellar, T. A. and Lichtenberg, J. J., "Journal American Water Works Association," 66, 739 (1974).

       3.    Bellar, T. A. and Lichtenberg, J. J., "Semi-Automated Headspace Analysis of Drinking Waters and
            Industrial Waters for Purgeable Volatile Organic Compounds," in Measurement of Organic Pollutants
            in Water and Wastewater, C. E. VanHall, ed., American Society for Testing Materials, Philadelphia,
            PA, Special Technical Publication 686 (1978).

       4.    National Standard Reference Data System, "Mass Spectral Tape Format," U.S. National Bureau of
            Standards (1979 and later attachments).

       5.    "Working with Carcinogens," DREW, PHS, NIOSH, Publication 77-206 (1977).

       6.    "OSHA Safety and Health Standards, General Industry," 29 CFR 1910, OSHA 2206 (1976).

       7.    "Safety in Academic Chemistry Laboratories," American Chemical Society Publication, Committee
            on Chemical Safety (1979).

       8.    "Methods 330.4 and 330.5 for Total Residual Chlorine," USEPA, EMSL Cincinnati, OH 45268,
            EPA-4-79-020 (March 1979)"Handbook of Analytical Quality Control in Water and Wastewater Lab-
            oratories," U.S. EPA, EMSL Cincinnati, OH 45268, EPA-4-79-019 (March 1979).

       9.    "Handbook for Analytical Quality Control in Water and Wastewater Laboratories," U.S. EPA, EMSL
            Cincinnati, OH 45268, EPA-600/4-79-019 (March 1979).
   60

-------
                                                                                                          Method 1666
Table 1. Volatile PMI Analytes Amenable to Purge-and-Trap and Determined by GC/MS Using Isotope-
Dilution and Internal-Standard Techniques
CASRN1
628-63-7
7141-0
123-86-4
71-36-3
75-65-0
110-82-7
141-78-6
98-01-1
142-82-5
110-54-3
78-84-2
67-63-0
108-21-4
108-20-3
107-31-3
108-10-1
109-66-0
109-99-9
75-69-4
108-38-3
106-42-3
95-47-6
EPA-EGD
977
978
979
1036
1343
1333
1736
981
1334
1335
982
1044
983
960
991
1341
984
1345
552
1332

1331
                                                                                 Labeled Compound
Analog
| CASRN1
EPA-EGD
                                                                              ,0
                                                                             13C
                                                                              ,6
                                                                             d2
                                                                             J10
53001-22-2
 1735-17-7
84508-45-2

33838-52-7
21666-38-6
 1693-74-9
                                                                                      41051-88-1
                                                                                      56004-61-6
1243
1233
1636

1234
1235
                            	Pollutant
PMI Analyte
n-Amyl acetate
n-Amyl alcohol
n-Butyl acetate
n-Butyl alcohol
tert-Butyl alcohol
Cyclohexane
Ethyl acetate
Furfural2
n-Heptane
n-Hexane
Isobutyraldehyde2
Isopropanol
Isopropyl acetate
Isopropylether
Methyl formate
Methylisobutyl ketone
n-Pentane3
Tetrahydrofuran
Trichlorofluoromethane
m,p-Xylene

o-xylene
1 Chemical Abstracts Service Registry Number
2 These aldehydes from the PMI may be analyzed by this method or Method 1667
3 n-Pentane is not explicitly a PMI analyte. However, the sum of the concentrations of n-pentane, n-hexane, and n-heptane are
  to be used to estimate the concentration of petroleum naphtha in PMI wastewaters.
1245
                1232
                1231
                                                                                                              61

-------
Method 1666
Table 2. Volatile PMI Analytes
Internal-Standard Techniques
Determined by Direct Aqueous Injection GCMS
Using Isotope-Dilution and
Pollutant Labeled Compound
PMI Analyte
Acetonitrile
Diethylamine
Dimethylamine
Dimethyl sulfoxide
Ethanol
Ethylene glycol
Formamide
Methanol
Methylamine
Methyl cellosolve
n-Propanol
Triethylamine
CASRN1
75-05-8
109-89-7
124-40-3
67-68-5
64-17-5
107-21-1
75-12-7
67-56-1
74-89-5
109-864
71-23-8
121-44-8
EPA-EGD Analog
972 dg
986
987
1037 de
1734 de
1038
988
1735 d3
989
1040
755 1-d,
990
| CASRN1 EPA-EGD
2206-26-0 1272


2206-27-1 1237
1516-08-1 1634


1849-29-2 1635


not avail. 1255

         1 Chemical Abstracts Service Registry Number
    62

-------
                                                                                                             Method 1666
Table 3. Gas Chromatographic Retention Times and Minimum Levels for Volatile PMI Analytes Determined
by Purge-and-Trap GC/MS
    EGD
    No.1      PMI Analyte
      991     Methyl formate
      552     Trichlorofluoromethane
      984     n-Pentane
     1344     Isopropanol
     1243     tert-Butyl alcohol-d10
     1343     tert-Butyl alcohol
     1235     n-Hexane-d14
      982     Isobutyraldehyde
     1335     n-Hexane
      960     Isopropylether
     1636     Ethyl acetate-13c
     1736     Ethyl acetate
      181     Bromochloromethane (int std)
     1245     Tetrhydrofuran-d8
     1345     Tetrahydrofuran
     1233     Cyclohexane-d12
     1333     Cyclohexane
     1234     n-Heptane-d,6
     1336     n-Butanol
     1334     n-Heptane
      985     1,4-Difluorobenzene (int std)
      983     Isopropyl acetate
     1341     Methylisobutyl ketone
      978     n-Amyl alcohol
      979     n-Butyl acetate
      981     Furfural
      207     Chlorobenzene-ds (int std)
     1232     p-Xylene-d10
     1332     m,p-Xylene
     1231     o-Xylene-d10
      977     n-Amyl acetate
     1331     o-Xylene
 1 Three-digit EGD numbers beginning with 0,1,5, or 9 indicate a pollutant quantified by the internal standard method; beginning
  with 2 or 6 indicate  a labeled compound quantified by the internal standard method; beginning with 3 or 7 indicate a pollutant
  quantified by isotope dilution.The initial '1* in four-digit EGD numbers is to be ignored in applying these rules.
 2 This is the minimum level at which the entire analytical system shall give recognizable mass spectra (background-corrected)
  and acceptable calibration points, taking into account method-specific sample and injection volumes.  The concentration in the
  aqueous or solid phase is determined using the equations in Section 14.

Mean (sec) |
526
613
622
687
730
741
820
823
839
865
925
925
954
956
964
981
996
1013
1015
1033
1052
1128
1157
1202
1268
1354
1359
1368
1379
1413
1417
1424
Retention Time
EGD Ref
181
181
181
181
181
1243
181
181
1235
181
181
1636
181
181
181
181
1233
985
985
1234
985
985
985
985
207
207
207
207
1232
207
207
1231

Relative
0.551
0.642
0.652
0.720
0.765
1.016
0.860
0.863
1.023
0.907
0.970
1.000
1.000
1.002
1.012
1.028
1.015
0.963
0.964
1.020
1.000
1.072
1.100
1.143
0.933
0.996
1.000
1.007
1.008
1.040
1.043
1.008
ML2
(M9/L)
100
10
10
200

100

10
10
5

10


20

5

500
10

10
10
500
5
500


10

5
5

-------
Method 1666
        Table 4. Gas Chromatographic Retention Times and Minimum Levels for Volatile PMI Analytes by Direct
        Aqueous Injection GC/MS
                                                                      Retention Time
             No.1     PMIAnalyte                         Mean (sec)       EGDRef        Relative         (mg/L)
              989     Methylamine                               81            1245        0.308           200
             1635     Methlyateohol-dg                           85           1245        0.323
             1735     Methyl alcohol                              85.5          1635        1.006             50
              987     Dimethylamine                             93           1245        0.354           200
             1634     Ethyl alcohol-^                           103           1245        0.394
             1734     Ethyl alcohol                              104           1634        1.010             20
             1272     Acetonitrile-dg                             119           1245        0.452
              972     Acetonitrite                               121            1272        1.017              5
             1255     n-PropanoM-d,                           170           1245        0.464
              755     n-Propanol                               170.5          1255        1.003             20
              986     Diethylamine                              188           1245        0.717           200
             1245     Tetrahydrofuran-dg(intstd)                  263           1245        1.000
             1040     Methyl ce)losofve(2-MethoxyethanoO         290           1245        1.103             50
              990     Triethylamine                             372           1245        1.414           200
             1038     Elhyleneglycol                            398           1245        1.513           200
              988     Formamide                               400           1245        1.521           1000
             1237     Dimethyl sulfoxide-de                       639           1245        2.431
             1037     Dimethyl sulfoxkJe                         643           1237        1.006           100
         1 Three-digit EQD numbers beginning with 0, 1, 5, or 9 indicate a pollutant quantified by the internal standard method; beginning
           with 2 or 6 indicated a labteted compound quantified by the internal standard method; beginning with 3 or 7 indicate a poUutant
           quantified by isotope dilution.  The initial '1 ' in four-digit EGD numbers is to be ignored in applying these rules.
         2 This is the minimum level at which the entire analytical system shall give recognizable mass spectra (background-corrected)
           and acceptable calforation points, taking into account method-specific sample and injection volumes. The concentration in the
           aqueous or solid phase is determined using the equations in Section 14.

        Table 5. BFB Mass-Intensity Specifications
                                     m/z             Intensity Required
                                     50              15-40% of m/z 95
                                     75              30-60% of m/z 95
                                     95              base peak, 100%
                                     96              5-9% of m/Z 95
                                     173              less than 2% of m/z 174
                                     1 74              greater than 50% of m/z 95
                                     175              5-9% of m/z 174
                                     176              95-101% of m/z 174
                                     177              5-9% of m/z 176
    64

-------
Method 1666
Table 6. Acceptance Criteria for PMI Analytes
Acceptance Criteria for Performance Tests (% of Spike Level)

EGO
No. PMI Analyte
972 Acetonrtrile
977 Amyl acetate
978 Amyl alcohol
979 n-Butyl acetate
1036 n-6utyl alcohol
1343 tert-Butyl alcohol
1333 Cydohexane
986 DiethylamJne
987 Dimethyiamine
1037 Dimethyl sulfoxide
1734 Ethanol
1736 Ethyl acetate
1038 Ethylene glycol
988 Formamide
981 Furfural
1334 n-Heptane
1335 n-Hexane
982 IsobutyraJdehyde
1044 Isopropanol
983 Isopropyl acetate
960 Isopropyl ether
1735 Methanol
989 Methylamine
1040 Methyl cellosotve
991 Methyl formate
1 341 Methylisobutyl ketone
984 n-Pentane
755 n-Propand
975 Tetrahydrofuran
552 Trichlorofluoromethane
990 Triethylamine
1332 m.p-Xytene
1331 o-Xytene

Spike Level
50mg/L
10ng/L
200jig/L
10ng/L
200|xg/L
50ng/L
10jig/L
250mg/L
250mg/L
250mg/L
50mg/L
10jig/L
250mg/L
500mg/L
100»ig/L
Labeled and Native
PMI Analyte
Initial Precision and Labek
Accuracy Am
s X Recov

Labeled and
KlPMI Native PMI
lyte Analyte Ongoing
ery (P) Accuracy (R)
5 94-103 99-110 93-103
14 80-107
75 16-166
20 88-129
108 d-190
121 d-202
78-109
10-172
86-130
d-199
d-212
26 82-134 8-156 80-136
31 71-132
38 61-136
68-134
58-139
16 84-116 59-122 83-118
15 66-96 103-121 65-97
48 60-157 58-159 57-160
195 d-310
113 60-286
186 d-282
d-326
51-296
d-297
10fig/L 37 87-161 14-128 84-164
10U0/L
10ng/L
100ng/L
10ng/L
10jig/L
50mg/L
250mg/L
250mg/L
50jig/L
10»ig/L
10ufl/L
50mg/L
34 86-154 5-157 84-157
54 67-176
284 d-418
32 82-147
21 86-127
63-180
d-441
80-150
84-129
26 57-109 100-120 55-111
36 61-133
13 99-125
73 20-165
42 78-162
52 51-155
59-136
98-126
14-171
75-165
47-159
25 42-93 54-149 40-95
10ufl/L 89 35-214 42-178 28-221
20jig/L
250mg/L
20WJ/L
10|ig/L
9 90-108
31 71-133
89-109
69-135
15 90-121 76-109 89-122
16 84-116 84-100 82-118

-------
Method 1666
Table 7. Characteristic m/z's for Volatile PMI Analytes
Spike Level Primary m/z

PMI Analyte ftig/L)
Acetonitrile
n-Amyl acetate 10
n-Amyl alcohol 100
n-Butyl acetate 10
n-Butyl alcohol 200
tert-Butyl alcohol 50
Cyclohexane 10
Diethylamine
Dimethylamine
Dimethyl sulfoxide
Ethanol
Ethyl acetate 10
Ethylene glycol
Formamide
Furfural 100
n-Heptane 10
n-Hexane 10
Isobutyraldehyde 10
Isopropanol 100
Isopropyl acetate 10
Isopropyl ether 10
Methanol
Methylamine
Methyl cellosolve (2-methoxyethanol)
Methyl formate 50
Methylisobutyl ketone 10
n-Pentane 10
n-Propanol
Tetrahydrofuran 10
Trichlorofluoromethane 20
Tn'ethylamine
m,p-Xylene 20
o-Xylene 10
Labeled (Native/
(mg/L) Analog Labeled)
50 dg 41/44
43
70
43
56
d,0 59/66
d12 56/96
250 58
250 44
250 de 47/50
50 d5 31/33
13Q 43/44
250 31
500 45
96
d16 71/82
d,4 57/66
72
45
43
45
50 dg 31/33
250 30
250 45
60
43
43
50 1-d, 31/32
d8 72/80
101
250 86
d,0 106/116
d10 106/116
Reference
Compound
1272
207
985
207
985
1234
1233
1245
1245
1237
1634
1636
1245
1245
207
1234
1235
181
181
985
181
1635
1245
1245
181

181
1255
1245
181
1245
1232
1231
    66

-------
                                                                                                        Method 1666
Table 8. Maximum Recoveries for PMI Analytes by Purge-and-Trap GC/MS


                                    PMI Analyte         Maximum Recovery (%)
                             n-Amyl acetate                       110
                             n-Amyl alcohol                       300
                             n-Butyl acetate                       120
                             n-Butyl alcohol                       440
                             tert-Butyl alcohol                     130
                             Cyclohexane                        120
                             Ethyl acetate                        130
                             Furfural                            170
                             n-Heptane                          140
                             n-Hexane                           140
                             Isobutyraldehyde                     150
                             Isopropanol                         250
                             Isopropl acetate                      130
                             Isopropyl ether                       120
                             Methyl formate                       120
                             Methylisobutyl ketone                 130
                             n-Pentante                          130
                             Tetrahydrofuran                      150
                             Trichlorofluoromethane                110
                             m,p-Xylene                         120
                             o-Xylene                           120
                                                                                                             67

-------
Method 1666
              Optional
              Foam Trap
                                            Exit 1/4 in. O.D.
                                        —14mmO.D.
                                           Inlet 1/4 in. O.D.
          Exit 1/4 In. O.D.
           10 mm Glass Frit,
           Medium Porosity
Sample Inlet

2-Way Syringe Valve

17-cm, 20-Gauge Syringe Needle

6 mm O.D. Rubber Septum

Inlet, 1/4 in. O.D.
                                                                                      Stainless Steel, 1/16 in. O.D.
                                                                                      13x
                                                                                      Molecular Sieve
                                                                                      Purge-Gas Filter
                                                                                         Purge-Gas
                                                                                         Flow Control
                                   Figure 1.   Purging Device for Waters
                                                                                                             52-025-6
   68

-------
                                                                                               Method 1666
        Purge Inlet Fitting
        Sample Outlet Fitting
3 in. Long x 6 mm O.D. Glass Tubing
                      40-mLVial
                                                                            Septum
                                                                               Cap
                                                                                          52-025-7
              Figure 2.     Purging Device for Soils or Waters

-------
Method 1666
                     Packing Detail
Construction Detail
                                5 mm Glass Wool
                           7.7 cm Silica Gel
                           15cmTenaxGC
                           •-  1cm3%OV-1
                           £-  5 mm Glass Wool
                             Trap Inlet
          Compression
          Fitting Nut
          and Ferrules
          14ft,7y/ft
          Resistance Wire
          Wrapped Solid
                                                                                        Thermocouple/
                                                                                        Controller
                                                                                        Sensor
             Electronic
             Temperature
             Control and
             Pyrometer
                                                                                         Tubing, 25 cm
                                                                                         0.105in.l.D.
                                                                                         0.125in.O.D.
                                                                                         Stainless Steel
                                         Figure 3.   Trap Construction and Packings
                                                                                                             52-025-8
    70

-------
                                                                                                    Method 1666
Carrier-Gas
Flow Control

Pressure
Regulator
 Purge-Gas
 Flow Control

 13x Molecular
 Sieve Filter
                  Liquid Injection Ports
                      ,— Column Oven
                                       Optional 4-Port Column
                                   A   Selection Valve

                                                    Trap Inlet
6-Port
Valve
                                      Confirmatory Column

                                      To Detector

                                      Analytical Column
                               Vent
                                           Purging
                                           Device
                                    Note:
                                    All lines between trap
                                    and GC should be heated
                                    to 80°C.
                                                                                             52-025-9
       Figure 4.     Schematic of Purge-and-Trap Device—Purge Mode
                                                                                                          71

-------
Method 1666
            Carrier-Gas
            Flow Control
          Pressure
          Regulator
           Purge-Gas
           Flow Control
            13x Molecular
            Sieve Filter
                                                 Optional 4-Port Column
                                                 Selection Valve
 Liquid Injection Ports

     i— Column Oven
"JULJU"*-
                                                                          Confirmatory Column
                                                                         To Detector
                                                                          Analytical Column
                                                                 Inlet
                                                            Trap,
                                                            200°C
                                                     Purging
                                                     Device
                    Note:
                    All lines between trap
                    and GC should be heated
                    to 80°C.
                          Figure 5.   Schematic of Purge-and-Trap Device—Desorb Mode
                                                                                                52-025-10
    72

-------
                                                                   I        I
                                                                                                  Method 1666
 10 —
1.0 —
0.1  —
                I
                2
I
5
 I         I          I
10       20        50
 Concentration (pg/L)
 I        I
100     200
                    The Dotted Lines Enclose a ±10% Error Window.
        Figure 6.     Relative Response Calibration Curve for o-Xylene
                                                                                           52-025-11
                                                                                                       73

-------
Method 1666
               (A)
                                                                   Area=168920
                                                                                     M/Z116

                                                                                     M/Z106
               (B)
                           Area=60960
                                                                                     M/Z116

                                                                                     M/Z106
               (C)
M/Z106
M/Z116
96868
82508
                                                                                     M/Z116

                                                                                     M/Z106
               Figure 7.     Extracted Ion Current Profiles for (A) o-Xylene, (B) o-Xylene-d10,and
                            (C) a Mixture of o-Xylene and o-Xylene-d10
                                                                                                    52-025-12
    74

-------
                                                                                        Method 1666
170000 -


160000 -

150000 -

140000 -

130000 -

120000 -

110000 -

100000 -

 90000 -

 80000 -

 70000 -

 60000 -

 50000 -

 40000 -

 30000 -

 20000 -

 10000 -


     0
File
Operator
Acquired
Instrument
Sample Name
Misc Info
Vial Number
D:\DATA\HOTPURGBAUG1994\082694\HPSTD1.D
rln
26 Aug 94 2:27 pm using AcqMethod HOTPURGE
HP-1
Hotpurge Method Std. 08/26/94
hotpurge analytes (50 ppb IS/SS)
2
                    "UUJ
                                                 LMj
                    10
                          15
                                20
                     Figure 8.  Chromatogram of Aqueous Performance
                               Standard of Analytes from Table 1
                                                                                    52-025-13
                                                                                            75

-------
Method 1666
           100 -
        %FS
                                                          Sample
                                                          Lab File ID
                                                          Analyzed
                                                          instrument ID
STD. 400/2000
4K15A02
11/15/94
VG04
                                                                                     10.0
                           12.0
                                  Figure 9.  Chromatogram of Aqueous Performance
                                            Standard of Analytes from Table 2
                                                                                                       52425-14
    76

-------
                                                                                                Method 1666

&
            120,000
            100,000
                           I	I      I      I
                    <•
             80,000-
                    1      2
                                               o-Xylene-d10
                                                                                       -- +3s
1      1
34
   ^
    6
Analysis
 Number
1
7
1
8
 1
10
                                                          -3s


,.<
1.00'









                                                                                           -3s
                    6/1     6/1    6/1     6/1     6/2     6/2    6/3     6/3    6/4    6/5
                                                   Date
                                                 Analyzed
                                                                                            52-025-15
       Figure 10.    Quality Control Charts Showing Area (top graph) and Relative Response of
                    o-Xylene to o-Xylene-d10 (lower graph) Plotted as Function of Time or Analysis
                    Number
                                                                                                     77

-------
                      Method 1667

Formaldehyde, Isobutyraldehyde, and Furfural
                     by Derivatization and
       High-Pressure Liquid Chromatography
                                 January 1995

-------
                                  Method  1667


   Formaldehyde,  Isobutyraldehyde, and Furfural by Derivatization and

                     High-Pressure Liquid  Chromatography



1.   Scope and Application

1.1   This method is designed to meet the survey and monitoring requirements of the EPA's Engineering
     and Analysis Division (EAD). The method is used to determine certain organic pollutants specific to
     the pharmaceutical manufacturing industry (PMI) that can be derivatized and analyzed by high-pres-
     sure liquid chromatography (HPLC).
1.2  The chemical compounds listed in Table 1 may be determined in waters, soils, and municipal sludges
     by the method.
1.3  The detection limits of the method are usually dependent on the level of interferences rather than
     instrumental limitations. The limits in Table 2 are the minimum levels that can be reliably quantified
     by this method with no interferences present.
     Furfural (2-furaldehyde) forms two relatively stable geometric isomers upon derivatization with 2,4-
     dinitrophenylhydrazine (DNPH). The first isomer (probably anti-) elutes after the formaldehyde
     derivative and before the isobutryaldehyde derivative. The second isomer (probably syn-) elutes
     after the isobutyraldehyde derivative. Experience with this system has shown that the best quantita-
     tive results (lowest method detection limits) are obtained using the area from the first eluted peak
     rather than that from the second peak or the sum of the two areas. This method is for use only by ana-
     lysts experienced with HPLC or under the close supervision of such qualified persons.
1.4  This method is performance-based. The analyst is permitted to modify the method to overcome
     interferences or to lower the cost of measurements, provided that all performance criteria in this
     method are met. The requirements for establishing method equivalency are given in Section 9.1.2.

2.   Summary of the Method

2.1   For solid wastes or for aqueous wastes containing significant amounts of solid material, the aqueous
     phase, if any, is separated from the solid phase and stored for later analysis. If necessary, the particle
     size of the solids in the waste is reduced.  The solid phase is extracted with an amount of extraction
     fluid equal to 20 times the weight of the solid phase. The extraction fluid employed is a function of
     the alkalinity of the solid phase of the waste. Following extraction, the aqueous extract is separated
     from the solid phase by filtration employing 0.6- to 0.9-um glass fiber-filter.
2.2  If compatible (i.e., multiple phases will not form on combination), the initial aqueous phase of the
     waste is added to the aqueous extract, and these liquids are analyzed together. If incompatible, the
     liquids are analyzed separately and the results are mathematically combined to yield a volume-
     weighted average concentration.
2.3  A measured volume of aqueous sample or an appropriate amount of solids leachate is buffered to
     pH = 5 and derivatized with 2,4-dinitrophenylhydrazine (DNPH), using either the solid-sorbent or
     methylene chloride derivatization/extraction option. If the solid-sorbent option is used, the deriva-
     tive is extracted using solid-sorbent cartridges, followed by elution with ethanol. If the methylene
                                                                                             81

-------
Method 1667
             chloride option is used, the derivative is extracted with methylene chloride.  The methylene chloride
             extracts are concentrated using the Kudema-Danish (K-D) procedure and solvent exchanged into
             methanol prior to HPLC analysis. Liquid chroraatographic conditions are described that permit the
             separation and measurement of formaldehyde, isobutyraldehyde, and furfural derivatives in the
             extract by absorbance detection at 365 nm.
       2.4   The quality of the analysis is assured through reproducible calibration and testing of the derivatiza-
             tion/extraction procedure and the HPLC system.

       3.    Definitions

       There are no method-specific definitions to be noted for this document.

       4.    Interferences

       4.1   Method interferences may be caused by contaminants in solvents, reagents, glassware, and other
             sample processing hardware that lead to discrete artifacts and/or elevated baselines in chromato-
             grams. All of these materials must be routinely demonstrated to be free from interferences under the
             conditions of the analysis by analyzing laboratory reagent blanks as described in Section 9.3.
             4.1.1   Glassware must be scrupulously cleaned. Clean all glassware as soon as possible after use
                    by rinsing with the last solvent used. This should be followed by detergent washing with hot
                    water, and rinses with tap water and reagent water. It should then be drained, dried, and
                    heated in a laboratory oven at 130 ^C) for several hours before use. Solvent rinses with meth-
                    anol  may be substituted for the oven heating.  After drying and cooling, glassware should be
                    stored in a clean environment to prevent any accumulation of dust or other contaminants.
             4.1.2   The use of high-purity reagents and solvents helps to minimize interference problems. Purifi-
                    cation of solvents by distillation in all-glass systems may be required.
       4.2   Analysis of formaldehyde is complicated by its ubiquitous occurrence in the environment. Acetic
             acid, even high-purity acetic acid, is  often contaminated with formaldehyde. For this reason, a phtha-
             late buffer is used in this method instead of an acetate buffer. Wherever acetic acid is used, it must be
             demonstrated to be formaldehyde free.
       4.3   Matrix interferences may be caused by contaminants that are coextracted from the sample.  The
             extent of matrix interferences will vary considerably from source to source, depending upon the
             nature and diversity of the matrix being sampled. If matrix interferences occur, some additional
             cleanup may  be necessary.
       4.4   The extent of interferences that may be encountered using liquid chromatographic techniques has not
             been fully assessed.  Although the HPLC conditions described allow for resolution of the  specific
             compounds covered by this method,  other matrix components may interfere.

       5.    Safety

       5.1   The toxicity or carcinogenicity of each compound or reagent used in this method has not been pre-
            cisely determined; however, each chemical compound should be treated as a potential health hazard.
            Exposure to these compounds should be reduced to the lowest possible level. The laboratory is
            responsible for maintaining a current awareness file of OSHA regulations regarding the safe han-
            dling of the chemicals specified in this method. A reference file of material safety data sheets should
            also be made  available to all personnel involved in these analyses. Additional information on labora-
            tory safety can be found in References 1 through 3.
   82

-------
                                                                                            Method 1667
5.2  Formaldehyde has been classified as a potential carcinogen. Primary standards of formaldehyde
     should be prepared in a hood, and a NIOSH/MESA approved toxic gas respirator should be worn
     when high concentrations are handled.

6.   Equipment and Supplies
6.1   Reaction vessel—250-mL Florence flask.
6.2  Separatory funnel—250-mL, with polytetrafluoroethylene (PTFE) stopcock.
6.3  Kuderna-Danish (K-D) apparatus.
     6.3.1   Concentrator tube—10-mL, graduated (Kontes K-570050 or equivalent). A ground-glass
            stopper is used to prevent evaporation of extracts.
     6.3.2   Evaporation flask—500-mL (Kontes K-570001-500 or equivalent). Attach to concentrator
            tube with springs, clamps, or equivalent.
     6.3.3   Snyder column—Three-ball macro (Kontes K-503000-0121 or equivalent).
     6.3.4   Snyder column—-Two-ball micro (Kontes K569001-0219 or equivalent).
     6.3.5   Springs—'/a" (Kontes K-662750 or equivalent).
6.4  Vials—10-mL, 25-mL, glass with PTFE-lined screw-caps or crimp-tops.
6.5  Boiling chips—Solvent-extracted with methylene chloride, approximately 10/40 mesh (silicon car-
     bide or equivalent).
6.6  Balance—Analytical, capable of accurately weighing to the nearest 0.0001 g.
6.7  pH meter—Capable of measuring to the nearest 0.01.
6.8  High-performance liquid chromatograph (modular).
     6.8.1   Pumping system—Isocratic, with constant flow control capable of 1.00 mL/min.
     6.8.2   High-pressure injection valve with 20-uL loop.
     6.8.3   Column—250 mm long x 4.6 mm inside diameter (i.d.),  5-um particle size, C18 (or equiva-
            lent).
     6.8.4   Absorbance detector—365 nm.
     6.8.5   Strip-chart recorder compatible with the detector. Use of a data system is recommended.
6.9  Glass-fiber filter paper, 0.6- to 0.9-um.
6.10  Solid-sorbent cartridges—Packed with 500 mg C18 (Baker or equivalent).
6.11  Vacuum manifold—Capable of simultaneous extraction of up to  12 samples (Supelco or equivalent).
6.12  Sample reservoirs—50-mL capacity (Supelco or equivalent).
6.13  Pipet—Capable of accurately delivering 0.10 mL of solution (Pipetman or equivalent).
6.14  Water bath—Heated, with concentric ring cover, capable of temperature control of ± 2 °C at 80-90 °C
     The bath should be used under a hood.

-------
Method 1667
       7.   Reagents and Standards

       7.1   Reagent grade chemicals shall be used in all tests. Unless otherwise indicated, it is intended that all
            reagents shall conform to the specifications of the Committee on Analytical Reagents of the Ameri-
            can Chemical Society, where such specifications are available. Other grades may be used, provided
            it is first ascertained that the reagent is of sufficiently high purity to permit its use without lessening
            the accuracy of the determinations.
       7.2  Reagent water—Water in which the compounds of interest and interfering compounds are not
            detected by this method. It may be generated by any of the methods in this subsection.
            7.2.1    Activated carbon—Pass tap water through a carbon bed (Calgon Filtrasorb-300, or equiva-
                    lent).
            7.2.2    Water purifier—Pass tap water through a purifier (Millipore Super Q, or equivalent).
            7.2.3    Boil and purge—Heat tap water to 90-100 °C and bubble contaminant-free inert gas through
                    it for approximately 1 hour. While still hot, transfer the water to screw-cap bottles  and seal
                    with a PTFE-lined cap.
       7.3  Methylene chloride—HPLC grade or equivalent.
       7.4  Methanol—HPLC grade or equivalent.
       7.5  Ethanol (absolute)—HPLC grade or equivalent.
       7.6  2,4-Dinitrophenylhydrazine (DNPH, 70% w/w) in reagent water.
       7.7  Formalin (37.6% w/w) in reagent water.
       7.8  Acetic acid (glacial), demonstrated to be formaldehyde-free.
       7.9   Potassium acid phthalate.
       7.10 Sodium hydroxide solutions, 1 N, and 5 N.
       7.11  Sodium chloride.
       7.12 Sodium sulfate solution, 0.1 M.
       7.13 Hydrochloric acid, 0.1 N.
       7.14 Extraction fluid—Dilute 64.3 mL of 1.0 N sodium hydroxide and 5.1 mL of glacial acetic acid to 900
            mL with reagent water. Further dilute to 1 L with reagent water. The pH should be 4.93 ± 0.02.  If
            not, adjust with acid or base.
       7.15 Stock standard solutions.
            7.15.1  Stock formaldehyde (approximately 1.00 mg/mL)—Prepare by diluting 265 uL formalin to
                   100 mL with reagent water.
                   Standardization of formaldehyde stock solution—Transfer a 25-mL aliquot of a 0.1 M
                   sodium sulfate solution to a beaker and record the pH. Add a 25-mL aliquot of the formalde-
                   hyde stock solution (Section 7.15.1) and record the pH. Titrate this mixture back to the orig-
                   inal pH using 0.1 N hydrochloric acid. The formaldehyde concentration is calculated using
                   the following equation:
   84

-------
                                                                                            Method 1667
                  Concentration (mg/mL)  = 30.03 x (NHCl) x (mLHCl) x25

                      where:
                      NHCL = Normality of the HCl solution
                      mL HCl= mL of standardized HCl solution
                      30.03   = Molecular weight of formaldehyde
     7.15.2  Stock formaldehyde, isobutyraldehyde, and furfural—Prepare by adding 265 uL of formalin,
            0.100 g of isobutyraldehyde, and 0.100 g of furfural to 90 mL of reagent water and dilute to
            100 mL.  The concentrations of isobutyraldehyde and furfural in this solution are 1.00 mg/
            mL.  Calculate the concentration of formaldehyde in this solution using the results of the
            assay performed in Section 7.15.1.
     7.15.3  Stock standard solutions must be replaced after six months, or sooner if comparison with
            check standards indicates a problem.
     7.15.4  Aqueous performance standard—An aqueous performance standard containing formalde-
            hyde (nominally 100 ug/L), isobutyraldehyde at 100 ug/L, and furfural at 100 ug/L shall be
            prepared daily and analyzed each shift to demonstrate performance (Section 9).
     7.15.5  Preparation of calibration standards.
            7.15.5.1   Prepare calibration standard solutions of formaldehyde, isobutyraldehyde, and fur-
                     fural in reagent water from stock standard solution (Section 7.15.2). Prepare these
                     solutions at the following concentrations (in fig/mL) by serial dilution of the stock
                     standard solution: 50,20,10. Prepare additional calibration standard solutions at
                     the following concentrations, by dilution of the appropriate 50,20, or 10 ug/mL
                     standard:  5,0.5, 2,0.2,1,0.1. Make further dilutions if appropriate.
7.16  Reaction solutions.
     7.16.1  DNPH (1.00 mg/mL)—Dissolve 142.9 mg 70% (w/w) reagent in 100 mL of absolute etha-
            nol.  Slight heating or sonication may be necessary to effect dissolution.
     7.16.2  Phthalate buffer (0.1 N)—Prepare by dissolving 20.42 g of potassium acid phthalate in 1 L of
            reagent water. Adjust pH to 5 by addition of sodium hydroxide or hydrochloric acid, as nec-
            essary.
     7.16.3  Sodium chloride solution (saturated)—Prepare by mixing an excess of the reagent-grade
            solid with reagent water.

8.   Sample Collection, Preservation,  and Storage

8.1   Grab samples are collected in glass containers having a total volume greater than 20 mL. For aque-
     ous samples that pour freely, fill sample bottles so that no air bubbles pass through the sample as the
     bottle is filled and seal each bottle so that no air bubbles are entrapped. Maintain the hermetic seal on
     the sample bottle until time of analysis.
8.2   Samples are maintained at 0-4 tl from the time of collection until analysis. Samples must be deriva-
     tized within five days of collection and analyzed within three days of derivatization.

-------
Method 1667
       9.    Quality Control

       9.1   Each laboratory that uses this method is required to operate a formal quality assurance program (Ref-
             erence 4).  The minimum requirements of this program consist of an initial demonstration of labora-
             tory capability and analysis of standards and blanks as tests of continued performance. Laboratory
             performance is compared to established performance criteria to determine if the results of analyses
             meet the performance characteristics of the method.
             9.1.1    The analyst shall make an initial demonstration of the ability to generate acceptable accuracy
                    and precision with this method. This ability is established as described in Section 9.2.
             9.1.2    In recognition of advances that are occurring in analytical technology, and to allow the ana-
                    lyst to overcome sample matrix interferences, the analyst is permitted certain options to
                    improve separations or lower the costs of measurements. These options include alternative
                    extraction, concentration, cleanup procedures, and changes in columns and detectors.  Alter-
                    native techniques, such as substitution of immunoassay, and changes that degrade method
                    performance are not allowed. If an analytical technique other than the techniques specified
                    in this method is used, that technique must have a specificity equal to or better than the spec-
                    ificity of the techniques in this method for the analytes of interest.
                    9.1.2.1    Each time a modification is made to this method, the analyst is required to repeat
                             the procedure in Section 9.2.  If the detection limit of the method will be affected
                             by the change, the laboratory is required to demonstrate that the method detection
                             limit (MDL) (40 CFR Part 136, Appendix B) is lower than one-third the regulatory
                             compliance level. If calibration will be affected by the change, the analyst must
                             recalibrate the instrument per Section 10.
                    9.122   The laboratory is required to maintain records of modifications made to this
                             method. These records include the information below, at a minimum.
                             9.1.2.2.1    The names, titles, addresses, and telephone numbers of the analyst(s)
                                       who performed the analyses and modification, and of the quality con-
                                       trol officer who witnessed and will verify the analyses and modifica-
                                       tion.
                             9.1.2.2.2    A list of pollutant(s) measured, including name and CAS Registry
                                       Number.
                             9.1.2.2.3    A narrative stating the reason(s) for the modification.
                             9.1.2.2.4    Results from all quality control (QC) tests comparing the modified
                                       method to  this method, including:
                                       (a) Calibration (Section 10.1.2)
                                       (b) Calibration verification (Section 10.1.2.2)

                                       (c) Initial precision and accuracy (Section 9.2)

                                       (d) Analysis of blanks (Section 9.3)
                                       (e) Accuracy assessment (Section 9.5)
                            9.1.2.2.5    Data that will allow an independent reviewer to validate each determi-
                                       nation by tracing the instrument output (peak height, area, or other sig-
                                       nal) to the final result.  These data are to include:
                                       (a) Sample numbers and other identifiers
   86

-------
                                                                                             Method 1667
                                (b) Extraction dates
                                (c) Analysis dates and times
                                (d) Analysis sequence/run chronology
                                (e) Sample weight or volume (Section 11)
                                (f)  Extract volume prior to each cleaning step (Section 11.1.2)
                                (g) Final extract volume prior to injection (Section 11.3.4.5 or Section
                                   11.3.5.5)
                                (h) Injection volume (Section 12)
                                (i)  Dilution data, differentiating between dilution of a sample or an
                                   extract
                                (j)  Instrument and operating conditions
                                (k) Column and operating conditions (nature of column, dimensions,
                                   flow rates, solvents, etc.)
                                (I)  Detector operating conditions (wavelength, etc.)
                                (m) Chromatograms, printer tapes, and other recording of raw data
                                (n) Quantitation reports, data system outputs, and other data necessary
                                   to link raw data to the results reported
     9.1.3   Analyses of blanks are required to demonstrate freedom from contamination and that the
            compounds of interest and interfering compounds have not been carried over from a previ-
            ous analysis (Section 4). The procedures and criteria for analysis of a blank are described in
            Section 9.3.
     9.1.4   The laboratory shall, on an ongoing basis, demonstrate through the analysis of the aqueous
            performance standard (Section 7.15.4) that the analysis system is in control. This procedure
            is described in Section 10.
     9.1.5   The laboratory shall maintain records to define the quality of data that is generated.
9.2   Initial precision and accuracy—To establish the ability to generate acceptable precision and accuracy,
     the analyst shall perform the following operations for compounds to be calibrated.
     9.2.1   Analyze four aliquots of the aqueous performance standard (Section 7.15.4) according to the
            method beginning in Section 11. Use the solid-sorbent option or the methylene chloride
            option, whichever will be used routinely.
     9.2.2   Using results from Section 9.2.1, compute the average percent recovery (X) and the standard
            deviation of the recovery (s) for each compound.
     9.2.3   For each compound, compare s and X with the corresponding limits for initial precision and
            accuracy found in Table 3. If s and X for all compounds meet the acceptance criteria, system
            performance is acceptable and analysis of blanks and samples may begin. If, however, any
            individual s exceeds the precision limit or any individual X falls outside the range for accu-
            racy, system performance is unacceptable for that compound. This is an indication that the
            analytical system is not performing properly for the compound(s) in question. In this event,
            correct the problem and repeat the entire test (Section 9.2.1).
                                                                                                 87

-------
Method 1667
       9.3   Blanks—Reagent water blanks are analyzed to demonstrate freedom from contamination.
             "With each sample lot (samples analyzed on the same 8-hour shift), a blank shall be analyzed immedi-
             ately after analysis of the aqueous performance standard (Section 9.1.4) to demonstrate freedom from
             contamination. If any of the compounds of interest or any potentially interfering compound is found
             in a blank at greater than 10 ug/L, analysis of samples is halted until the source of contamination is
             eliminated and a blank shows no evidence of contamination at this level.
       9.4   The specifications contained in this method can be met if the apparatus used is calibrated properly,
             then maintained in a calibrated state. The standards used for calibration (Section 7.15.5), calibration
             verification (Section 10.1.2.2) and for initial (Section 9.2) and ongoing (Section 9.1.4) precision and
             accuracy should be identical, so that the most precise results will be obtained.
       9.5   Depending on specific program  requirements, field replicates may be collected to determine the pre-
             cision of the sampling technique, and spiked samples may be required to determine the accuracy of
             the analysis.

       10.   Calibration

       10.1   Establish liquid chromatographic operating parameters to produce a retention time equivalent to that
             indicated in Table 2 for formaldehyde derivative.  Suggested chromatographic conditions are pro-
             vided in Section 12.1. Prepare derivatized calibration standards according to the procedure in Section
             10.1.1. Calibrate the chromatographic system using the external standard technique (Section 10.1.2).
             10.1.1  Process each calibration standard solution through the derivatization option used for sample
                    processing (Section 11.3.4 or 11.3.5).
             10.1.2  External standard calibration procedure.
                    10.1.2.1   Analyze each derivatized calibration standard using the chromatographic condi-
                             tions specified in Section 12.1, and tabulate peak area against concentration
                             injected.  The results may be used to prepare calibration curves for formaldehyde,
                             isobutyraldehyde, and furfural.
                    10.1.2.2   The working calibration curve must be verified at the beginning of each 8-hour
                             shift by the measurement of one or more calibration standards. If the response for
                             any analyte varies from the previously established responses by more than 10%,
                             the test must be repeated using a fresh calibration standard after it is verified that
                             the analytical system is in control. Alternatively, a new calibration curve may be
                             prepared for that compound. If an autosampler is available, it is convenient to pre-
                             pare a calibration curve daily by analyzing standards along with test samples.

       11.  Sample Extraction, Cleanup, and Derivatization

       11.1  Extraction of solid samples.

            11.1.1   All solids must be homogeneous.  When the sample is not dry, determine the dry weight of
                   the sample using a representative aliquot.

                   11.1.1.1  Determination of dry weight—In certain cases, sample results are desired based on
                            a dry weight basis. When such data is desired, a portion of the sample is weighed
                            out at the same time as the portion used for the analytical determination.
   88

-------
                                                                                            Method 1667
Warning: The drying oven should be contained in a hood or vented. Significant laboratory contamination
_ may result from drying a heavily contaminated hazardous waste sample. _


            11 .1 .1 .2  Immediately after weighing the sample for extraction, weigh 5-1 0 g of the sample
                     into a tared crucible. Determine the percent dry weight of the sample by drying
                     overnight at 105 °C. Allow to cool in a desiccator before weighing.
                                             g of dry sample
                                                of sample
                                                           *100
     11 .1 .2  Measure 25 g of solid into a 500-mL bottle with a PTFE-lined screw-cap or crimp-top, and
            add 500 mL of extraction fluid (Section 6.13). Extract the solid by rotating the bottle at
            approximately 30 rpm for 18 hours. Filter the extract through glass-fiber filter paper and
            store in a sealed bottle at 4 °C.  Each mL of extract represents 0.050 g of solid.
11.2   Cleanup and separation.
     11.2.1  Cleanup procedures may not be necessary for a relatively clean sample matrix. The cleanup
            procedures recommended in this method have been used for the analysis of various sample
            types.  If particular circumstances demand the use of an alternative cleanup procedure, the
            analyst must meet the specifications in Section 9.1.2.
     11 .2.2  If the sample is not clean, or the complexity is unknown, the entire sample should be centri-
            fuged at 2500 rpm for 10 minutes.  Decant the supernatant liquid from the centrifuge bottle
            and filter through glass-fiber filter paper into a container that can be tightly sealed.
11.3  Derivatization.
     11.3.1  For aqueous samples, measure a 50- to 100-mL aliquot of sample. Quantitatively transfer
            the sample aliquot to the reaction vessel (Section 6.1).
     11.3.2  For solid samples, 1-10 mL of leachate (Section 11.1.2 or Section 11.2.2) will usually be
            required. The amount used for a particular sample must be determined through preliminary
            experiments.
     11 .3.3  Derivatization and extraction of the derivative can be accomplished using the solid-sorbent
            (Section  11.3.4) or methylene chloride option (Section 11.3.5).
     11 .3.4  Solid Sorbent Option
            11.3.4.1  Add 4 mL of phthalate buffer (Section 7.16.2) and adjust the pH to 5.0 ±0.1 with
                     sodium hydroxide or hydrochloric acid. Add 10 mL of DNPH reagent, adjust the
                     total volume to approximately 100 mL with reagent water, seal the container and
                     place on a wrist-action shaker at room temperature for 1  hour. Samples or stan-
                     dards containing high analyte concentrations may require more DNPH reagent for
                     complete reaction.
            11 .3.4.2  Assemble the vacuum manifold and connect to a water aspirator or vacuum pump.
                     Assemble solid sorbent cartridges containing a minimum of 1.5 g of C18 sorbent,
                     using connectors supplied by the manufacturer, and attach the sorbent train to the
                     vacuum manifold. Condition each cartridge by passing 10 mL dilute phthalate
                     buffer (10 mL 5 N phthalate buffer dissolved in 250 mL of reagent water) through
                     the sorbent cartridge train.
                                                                                                89

-------
Method 1667
                    11.3.4.3  Remove the reaction vessel from the shaker and add 10 mL of saturated sodium
                            chloride solution to the vessel.
                    11.3.4.4  Add the reaction solution to the sorbent train and apply a vacuum so that the solu-
                            tion is drawn through the cartridges at a rate of 3 to 5 mL/min. Release the vacuum
                            after the solution has passed through the sorbent.
                    11.3.4.5  Elute each cartridge train with approximately 9 mL of absolute ethanol, directly
                            into a 10-mL volumetric flask. Dilute the solution to volume with absolute etha-
                            nol, mix thoroughly, and place in a tightly sealed vial until analyzed.
            11.3.5   Methylene chloride option.
                    11.3.5.1  Add 5 mL of phthalate buffer (Section  7.16.2) and adjust the pH to 5.0 ±0.1 with
                            sodium hydroxide or hydrochloric acid. Add 10 mL of DNPH reagent, adjust the
                            volume to approximately 100 mL with reagent water, seal the container, and place
                            on a wrist-action shaker at room temperature for 1 hour. Samples or standards with
                            high analyte concentrations may require more DNPH reagent for complete reac-
                            tion.
                    11.3.5.2  Extract the solution with three 20-mL portions of methylene chloride, using a 250-
                            mL separatory funnel, and combine the methylene chloride layers. If an emulsion
                            forms upon extraction, remove the entire emulsion and centrifuge at 2000 rpm for
                            10 minutes. Separate the layers and proceed with the next extraction.
                    11.3.5.3  Assemble a K-D concentrator by attaching a 10-mL concentrator tube to a 500-mL
                            evaporator flask. Wash the K-D apparatus with 25 mL of extraction solvent to
                            complete the quantitative transfer.
                    11.3.5.4  Add one or two clean boiling chips to the evaporation flask and attach a three-ball
                            Snyder column. Prewet the Snyder column by adding about 1 mL of methylene
                            chloride to the top. Place the K-D apparatus on a hot water bath (80-90 °C)  so that
                            the concentrator tube is partially immersed in the hot water and the entire lower
                            rounded surface of the flask is bathed with hot vapor.  Adjust the vertical position
                            of the apparatus and the water temperature, as required, to complete the concentra-
                            tion in 10-15 minutes.  At the proper rate of distillation, the balls of the column
                            will actively chatter, but the chambers will not flood with condensed solvent.
                            When the apparent volume of the liquid reaches  10 mL, remove the K-D apparatus
                            and allow it to drain and cool for at least 10 minutes.
                    11.3.5.5  Prior to liquid chromatographic analysis, the solvent must be exchanged to metha-
                            nol. The analyst must ensure quantitative transfer of the extract concentrate. The
                            exchange is performed as described below.
       11.4  After cooling and draining as described in Section 11.3.5.4, momentarily remove the Snyder column
            and add 5 mL of methanol and a new boiling chip. Attach the micro Snyder column. Concentrate
            the extract using 1 mL of methanol to prewet the Snyder column.  Place the K-D apparatus on the
            water bath so that the concentrator tube is partially immersed in the hot water.  Adjust the vertical
            position of the apparatus and the water temperature, as required, to complete the concentration.
            When the apparent volume of the liquid reaches less than 5 mL, remove the K-D apparatus and allow
            it to drain and cool for at least 10 minutes.
            Remove the Snyder column and rinse the flask and its lower joint with 1-2 mL of methanol and add
            to the concentrator tube. A 5-mL syringe is recommended for this operation.  Adjust the extract vol-
            ume to  10-mL with methanol. Stopper the concentrator tube and store refrigerated at 4
-------
                                                                                           Method1667
     processing will not be performed immediately. If the extract will be stored longer than two days, it
     should be transferred to a vial with a PTFE-lined screw-cap or crimp-top. Proceed with the liquid
     chromatographic analysis if further cleanup is not required.

12.  High-Pressure  Liquid Chromatography

12.1  Chromatographic conditions.
     Column: C18,250 mm long x 4.6 mm i.d., 5-um particle size (or equivalent).

     Mobile Phase: Methanol/water, 75:25 (v/v), isocratic at 30 °C.
     How Rate:  l.OmL/min.
     UV Detector: 365 nm.
     Injection Vol.: 20 uL.
12.2  Analysis.
     12.2.1   Analyze samples by HPLC using conditions described in Section 12.1. Table 2 lists the
            retention times and MDLs that were obtained under these conditions. Other HPLC columns,
            chromatographic conditions, or detectors may be used if the requirements of Section 9 are
            met.
     12.2.2  The width of the retention-time window used to make identifications should be based upon
            measurements of actual retention-time variations of standards over the course of a day.
            Three times the standard deviation of the retention time  for a compound can be used to cal-
            culate a suggested window size; however, the experience of the analyst should weigh heavily
            in the interpretation of chromatograms.
     12.2.3  If the peak area exceeds the linear range of the calibration curve, a smaller sample volume
            should be used. Alternatively, the final solution may be  diluted with ethanol or methanol, as
            appropriate, and reanalyzed.

     12.2.4  If the peak area measurement is prevented by the presence of observed interferences, further
            cleanup may be required.
12.3  Calculations.

     12.3.1   Calculate each response factor (RF) as follows (mean value based on 5 points):

                                      concentration of standard
                               RF =
                                          area of the signal
                                                       5
                                                          RF.
                                     mean RF = RF =
                                                                                               91

-------
Method 1667
            12.3.2   Calculate the concentration of formaldehyde, isobutyraldehyde, and furfural as follows:

                               \ig\mL = RFx. area of signal x concentration factor
                             where:
                                                      final volume of extract
                           concentration factor = /m-rifl/ sampie (or leachate) volume

                    12.3.2.1   For solid samples, a factor must be included in the equation to account for the
                             weight of the sample used and, if desired, to correct for dry weight.

       13.  Method Performance

       13.1  The MDLs listed in Table 2 were obtained using reagent water and methylene chloride extraction.
            Similar results can be obtained using the solid-sorbent method.
       13.2  This method has been tested for linearity of recovery from spiked reagent water and has been demon-
            strated to be applicable over the range from the ML to 50 times the ML.
       13.3  A representative chromatogram is presented as Figure 1.

       14.  Waste Management

       14.1  It is the laboratory's responsibility to comply with all federal, State, and local regulations governing
            waste management, particularly the hazardous waste identification rules and land-disposal restric-
            tions. In addition, it is the laboratory's responsibility to protect air, water, and land resources by min-
            imizing and controlling all releases from fume hoods and bench operations. Also, compliance is
            required with any sewage discharge permits and regulations.
       14.2  Samples containing acids at a pH of less than 2 are hazardous and must be neutralized before being
            poured down a drain or must be handled as hazardous waste.
       14.3  For further information on waste management, consult "The Waste Management Manual for Labora-
            tory Personnel," available from the American Chemical Society's Department of Government Rela-
            tions and Science Policy, 1115 Sixteenth Street, NW, Washington, DC 20036.
   92

-------
                                                                                       Method 1667
                                       References

1.    "Working with Carcinogens," DHEW, PHS, NIOSH, Publication 77-206 (1977).

2.    "OSHA Safety and Health Standards, General Industry," 29 CFR 1910, OSHA 2206, (1976).

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

4.    "Handbook of Analytical Quality Control in Water and Wastewater Laboratories," U.S. EPA, EMSL
     Cincinnati, OH 45268, EPA-600/4-79-019 (March 1979).
                                                                                           93

-------
Method 1667	

        Table 1. PMI Analytes to Which This Method Applies
                                   PMI Analyte                               CASRN1
                                   Formaldehyde                              50-00-0
                                   Furfural                                   98-01-1
                                   Isobutyraldehyde                           78-84-2
                                   1 Chemical Abstracts Service Registry Number.
        Table 2. Retention Times and Minimum Levels (MLs) for PMI Analytes


                                                                Retention Time1                 ML2
                  PMI Analyte                                      (seconds)                  (ug/L)
                  Formaldehyde                                         326                     50
                  Furfural                                              495                     50
                  Isobutyraldehyde                                      714                     50
                  1 Retention times are for the DNPH derivative.
                  o
                   This is the minimum level at which the entire analytical system shall give a recognizable signal and an acceptable cal-
                    ibration point, taking into account method-specific sample and injection volumes.
        Table 3. Requirements for Initial Precision and Accuracy
                                                    Spike               Average Percent         Standard Deviation
         PMI Analyte                             Level (ug/L)              Recovery (X)                   (s)
         Formaldehyde                                  20                   25-187                     81
         Furfural                                      100                   70-102                     16
         Isobutyraldehyde                               10                   45-121                     38
   94

-------
                                                                                               Method 1667
250 -
200 -
150 -
100 -
 50 -
                                   10 PPM STANDARD.        Amount: 1.000.
                                   Acquired on 17-08-94 at 00:43:45   Reported on 08-17-94 at 10:40:44
                                          10
I
15
20
                 Figure 1. Chromatogram of the 2,4-DNPH Derivative of Formaldehyde,
                          Furfural, and Isobutyraldehyde
                                                                                             52-025
                                                                                                   95

-------
                   Method 1671

Volatile Organic Compounds Specific to the
           Pharmaceutical Manufacturing
                    Industry by GC/FID
                              January 1995

-------
                                 Method  1671
                  Volatile Organic Compounds Specific to the
             Pharmaceutical Manufacturing Industry by GC/FID
1.   Scope and Application

1.1   This method is for survey and monitoring under the Clean Water Act. The method is used to deter-
     mine certain non-purgeable volatile organic pollutants specific to the pharmaceutical manufacturing
     industry (PMI) that are amenable to direct aqueous injection gas chromatography (GC) and detection
     by a flame ionization detector (FID).
1.2   The PMI analytes listed in Table 1 may be determined in waters, soils, and municipal sludges by this
     method.
1.3   The detection limits of Method 1671 are usually dependent on the level of interferences rather than
     instrumental limitations. The minimum levels (MLs) in Table 2 are the level that can be attained
     with no interferences present.
1.4   This method is recommended for use by, or under the supervision of, analysts experienced in the
     operation of gas chromatographs and in the interpretation of chromatograms.
1.5   This method is performance-based. The analyst is permitted to modify the method to overcome
     interferences or to lower the cost of measurements, provided that all performance criteria in this
     method are met.  The requirements for establishing method equivalency are given in Section 9.2.

2.   Summary of the Method

2.1   The percent solids content of the sample is determined. If the solids content is less than 1%, internal
     standard(s) is added to a 5-mL sample. If the solids content of the sample is greater than 1 %, 5 mL of
     reagent water and internal standard(s) are added to a 5-g aliquot of sample.
     The mixture is sonicated in a centrifuge tube with little or no headspace for 5 minutes. During this
     period the analytes and the internal standard will equilibrate between the solid and aqueous phases.
     In some cases, additional sonication will be necessary to establish equilibrium. The resulting suspen-
     sion is centrifuged and the supernatant liquid analyzed.
22   An appropriate amount of the aqueous solution (or supernate) is injected into the GC. The com-
     pounds are separated by the GC and detected by the FID.

3.   Definitions

     There are no definitions specific to this method.

4.   Interferences

4.1   Impurities in the carrier gas, organic compounds outgassing from the GC plumbing, and solvent
     vapors in the laboratory account for the majority of contamination problems encountered with this
                                                                                           99

-------
Method 1671
             method. The analytical system is demonstrated to be free from interferences under conditions of the
             analysis by analyzing reagent water blanks initially and with each sample batch (samples analyzed
             on the same 8-hour shift), as described in Section 9.4.
       4.2   Samples can be contaminated by diffusion of volatile organic compounds through the bottle seal dur-
             ing shipment and storage. A field blank prepared from reagent water and carried through the sam-
             pling and handling protocol may serve as a check on such contamination.
       4.3   Contamination by carryover can occur when high-level and low-level samples are analyzed sequen-
             tially. To reduce carryover, the syringe is cleaned or replaced with a new syringe after each sample is
             analyzed. When an unusually concentrated sample is encountered, it is followed by analysis of a
             reagent water blank to check for carryover.  Syringes are cleaned by washing with soap solution,
             rinsing with tap and distilled water, and drying in an oven at 100-125°C. Other parts of the system
             are also subject to contamination; therefore, frequent bakeout and purging of the entire system may
             be required.
       4.4   Interferences resulting from samples will vary considerably from source to source, depending on the
             diversity of the site being sampled.

       5.    Safety

             The toxicity or carcinogenicity of each analyte, compound, or reagent used in this method has not
             been precisely determined; however, each chemical compound should be treated as a potential health
             hazard. Exposure to these compounds  should be reduced to the lowest possible level. The laboratory
             is responsible for maintaining a current awareness file of OSHA regulations regarding the safe han-
             dling of the chemicals specified in this method. A reference file of material safety data sheets should
             also be made available to all personnel involved in these analyses. Additional information on labora-
             tory safety can be found in References 2-4.

       6.    Equipment and Supplies

       6.1    Sample bottles and septa
             6.1.1    Bottles—25- to 40-mL with polytetrafluoroethylene (PTFE)-lined screw-cap (Pierce 13075,
                   or equivalent). Detergent wash, rinse with tap and distilled water, and dry at >105°C for a
                   minimum of 1 hour before use.
            6.1.2   Septa—PTFE-faced silicone (Pierce 12722, or equivalent), cleaned as above and baked at
                   100-200°C for a minimum of 1 hour.
       6.2   Gas chromatograph—Shall be Linearly temperature programmable with initial and final holds, and
            shall produce results which meet the calibration (Section 10), quality assurance (Section 9), and per-
            formance tests (Section 13) of this method.

            6.2.1    Column—30 m long x 0.32 mm i.d. fused-silica microbore column coated with 4-um of
                   bonded poly(dimethylpolysiloxane) (Supelco SPB-1 Sulfur, or equivalent).

            6.2.2    GC operating conditions.
                   Temperatures:
                   Column—2 minutes at 40°C, 10°C per minute to 180°C.
                   Injection port—200°C
   100

-------
                                                                                            Method 1671
            FED—300°C
            Carrier gas—Hydrogen at a head pressure of 10 psig.
     An injector split may be used in order to optimize peak shape and repeatability.
6.3   Syringes—5-mL, gas-tight glass hypodermic, with Luer-lok tips.
6.4   Micro syringes—10-, 25-, and 100-|jL.
6.5   Syringe valves—2-way with Luer ends, PTFE.
6.6   Bottles—15-mL, screw-cap with PTFE liner.
6.7   Balances.
     6.7.1   Analytical, capable of weighing 0.1 mg.
     6.7.2   Top-loading, capable of weighing 10 mg.
6.8   Equipment for determining percent moisture.
     6.8.1   Oven, capable of being temperature-controlled at 110°C (±5°C).
     6,8.2   Desiccator.
     6.8.3   Beakers—50-, 100-mL.
6.9   Centrifuge apparatus.
     6.9.1   Centrifuge capable of rotating 10-mL centrifuge tubes at 5000 rpm.
     6.9.2   Centrifuge tubes, 10-mL, with screw-caps (PTFE-lined) to fit centrifuge.
6.10  Sonication apparatus capable of sonicating 10-mL centrifuge tubes and thoroughly agitating con-
     tents.

7.   Reagents and Standards
7.1   Reagent water: Water in which the compounds of interest and interfering compounds are not
     detected by this method.  It may be generated by any of the following methods:
     7.1.1   Activated  carbon—pass tap water through a carbon bed (Calgon Filtrasorb-300, or equiva-
            lent).
     7.1.2   Water purifier—Pass tap water through a purifier (Millipore Super Q, or equivalent).
     7.1.3   Boil and purge—Heat tap water to between 90 and 100°C and bubble contaminant-free inert
            gas through it for approximately 1 hour. While still hot, transfer the water to screw-cap bot-
            tles and seal with a PTFE-lined cap.
7.2   Sodium thiosulfate—ACS granular.
7.3   Standard  solutions—Purchased as solutions or mixtures with certification to their purity, concentra-
     tion, and  authenticity, or prepared from materials of known purity and composition. If compound
     purity is 96% or greater, the weight may be used without correction to calculate the concentration of
     the standard.
     7.3.1   Place approximately 8 mL of reagent water in a 10-mL ground-glass-stoppered  volumetric
            flask. Allow the flask to stand unstoppered for approximately 10 minutes or until all wetted
                                                                                                101

-------
Method 1671
                    surfaces have dried. For each analyte, weigh the stoppered flask, add the compound, restop-
                    per, then immediately reweigh to prevent evaporation losses from affecting the measure-
                    ment.
             7.3.2   Liquids—Using a microsyringe, add sufficient liquid (about 100 mg) so that the final solu-
                    tion will have a concentration of about 10 mg/mL.
             7.3.3   Gases—Fill a valved 5-mL gas-tight syringe with the compound. Lower the needle to
                    approximately 5 mm above the meniscus. Slowly introduce the compound above the surface
                    of the meniscus.  The gas will dissolve in the solvent.  Repeat if necessary to reach desired
                    concentration.
             7.3.4   Fill the flask to volume, stopper, then mix by inverting several times. Calculate the concen-
                    tration in milligrams per milliliter (mg/mL, equivalent to micrograms per microliter [|ig/uL])
                    from the weight gain.
             7.3.5   Transfer the stock solution to a PTFE-sealed screw-cap bottle. Store, with minimal head-
                    space, in the dark at approximately 4°C.  Do not freeze.
             7.3.6   Replace standards after one month, or sooner if comparison with check standards indicate a
                    change in concentration.  Quality control check standards that can be used to determine the
                    accuracy of calibration standards may be available from the National Institute of Standards
                    and Technology, Gaithersburg, MD.
       7.4   Secondary standards—Using standard solutions (Section 7.3), prepare a secondary standard to con-
             tain each pollutant at a concentration of 100 mg/L or 500 mg/L for compounds with higher MLs.
             Where necessary, a concentration of 1000 mg/L may be used.
             7.4.1   Aqueous calibration standards—Using a syringe or a microsyringe, add sufficient secondary
                    standard (Section 7.4) to five reagent water aliquots to produce concentrations in the range of
                    interest

             7.4.2   Aqueous performance standard—An aqueous standard containing all pollutants and internal
                    standard(s) is prepared daily, and analyzed each shift to demonstrate performance (Section
                    13). This standard shall contain concentrations of pollutants and internal standard(s), as
                    appropriate, within a factor of 1 to 5 times the MLs of the pollutants listed in Table 1. It may
                    be one of the aqueous calibration standards described in Section 7.4.1.

       8.    Sample Collection, Preservation, and Handling

       8.1   Grab samples are collected in glass containers having a total volume greater than 20 mL. For aque-
            ous samples that pour freely, fill sample bottles so that no air bubbles pass through the sample as the
            bottle is filled and seal each bottle so that no air bubbles are entrapped. Maintain the hermetic seal on
            the sample bottle until time of analysis.
       8.2   Maintain samples at 4°C from the time of collection until analysis. Do not freeze.  If an aqueous
            sample contains residual chlorine, add sodium thiosulfate preservative (10 rag/40 mL) to the empty
            sample bottles just prior to shipment to the sample site. EPA Methods 330.4 and 330.5 may be used
            for measurement of residual chlorine (Reference 5). If preservative has been added, shake the bottle
            vigorously for 1 minute immediately after filling.
       8.3   For aqueous samples, experimental evidence indicates that some PMI analytes are susceptible to
            rapid biological degradation under certain environmental conditions. Refrigeration alone may not be
   102

-------
                                                                                             Method 1671
     adequate to preserve these compounds in wastewaters for more than seven days. For this reason, a
     separate sample should be collected, acidified, and analyzed when compounds susceptible to rapid
     biological degradation are to be determined.  Collect about 500 mL of sample in a clean container.
     Adjust the pH of the sampleto about 2 by adding hydrochloric acid (1+1) while stirring. Check pH
     with narrow range (1.4 to 2.8) pH paper. Fill a sample bottle as described in Section 8.1.  If residual
     chlorine is present, add sodium thiosulfate to a separate sample bottle and fill as in Section 8.1.

8.4  All samples shall be analyzed within 14 days of collection.

9.   Quality Assurance/Quality Control

9.1   Each laboratory that uses this method is required to operate a formal quality assurance program (Ref-
     erence 6). The minimum requirements of this program consist of an initial demonstration of labora-
     tory capability (Section 9.5) and analysis of standards (Sections 9.6 and 13) and blanks (Section 9.4)
     as tests of continued performance.  Each time a batch of samples is analyzed or there is a change in
     reagents or procedures, a method blank must be  analyzed as a safeguard against contamination.
9.2  In recognition of advances that are occurring in analytical technology, and to allow the analyst to
     overcome sample matrix interferences, the analyst is permitted certain options to improve separa-
     tions or lower the costs of measurements. These options include alternative concentration and
     cleanup procedures, and changes in columns and detectors. Alternative techniques, such as the sub-
     stitution of spectroscopy or immunoassay, and changes that degrade method performance are not
     allowed. If an analytical technique other than the techniques  specified in this method is used, that
     technique must have a specificity equal to or better than  the specificity of the techniques in this
     method for the analytes of interest.
     9.2.1   If the detection limit of the method will be affected by the change, the laboratory is required
            to demonstrate that the MDL (40 CFR Part 136, Appendix B) is lower than one-third the reg-
            ulatory compliance level.  If calibration will be affected by the change, the analyst must reca-
            librate the instrument per Section 10.
     9.2.2   The laboratory is required to maintain records of modifications  made to this method. These
            records include the information in this subsection, at a minimum.
            9.2.2.1   The names, titles, addresses, and telephone numbers of the analyst(s) who per-
                     formed the analyses and modification, and of the quality control officer who wit-
                     nessed and will verify the analyses and modification.
            9.2.2.2   A listing of pollutant(s) measured, by name and CAS  Registry Number.
            9.2.2.3   A narrative stating the reason(s) for the modification.
            9.2.2.4   Results from all quality control (QC) tests comparing the modified method to this
                     method including:
                     (a)  calibration (Section 10);
                     (b)  calibration verification (Section 13);

                     (c)  initial precision and accuracy (Section 9.5);

                     (d)  analysis of blanks (Section 9.4); and
                     (e)  accuracy assessment (Section 9.6 and 13).
                                                                                                703

-------
Method 1671
                    9.2.2.5   Data that will allow an independent reviewer to validate each determination by
                             tracing the instrumentoutput (peak height, area, or other signal) to the final result.
                             These data are to include:
                             (a) sample numbers and other identifiers;
                             (b) analysis dates and times;
                             (c) injection logs;
                             (d) analysis sequence/run chronology;
                             (e) sample weight or volume;
                             (f)  sample volume prior to each cleanup step, if applicable;
                             (g) sample volume after each cleanup step, if applicable;
                             (h) final sample volume prior to injection;
                             (i)  injection volume;
                             (j)  dilution data, differentiating between dilution of a sample or an extract;
                             (k) instrument and operating conditions;
                             (I)  column (dimensions, liquid phase, solid support, film thickness, etc,);
                             (m) operating conditions (temperature, temperature program, flow rates, etc.);
                             (n) detector (type, operating condition, etc,);
                             (o) chromatograms, printer tapes, and other recording of raw data; and
                             (p) quantitation reports, data system outputs, and other data necessary to link raw
                                 data to the results reported.
       9.3   With each sample batch, a matrix spike (MS) and matrix spike duplicate (MSD) are analyzed to
             assess precision and accuracy of the analysis. The relative percent difference (RPD) between the MS
             and MSD shall be less than 30% and compound recoveries shall fall within the limits specified in
             Table 3. Where RPDs are found to be 30% or greater, or where compound recoveries fall outside of
             the limits specified in Table 3, the analytical process is considered to be out of control and sample
             analysis shall cease until the analytical process is returned to control as evidenced by RPDs of less
             than 30% and compound recoveries within the limits specified.
       94   Analyses of blanks are required to demonstrate freedom from contamination and that the compounds
             of interest and interfering compounds have not been carried over from a previous analysis (Section
            4.3).
            94.1    With each sample batch (samples analyzed on the  same 8-hour shift), a blank shall be ana-
                    lyzed immediately after analysis of the aqueous performance standard (Sections 9.6 and 13)
                    to demonstrate freedom from contamination. If any of the compounds of interest or any
                    potentially interfering compound is found in a blank at greater man the ML (assuming a
                    response factor of 1 relative to the nearest-eluted internal standard for compounds not listed
                    in Table 1), analysis of samples is halted until the source of contamination is eliminated and
                    a blank shows no evidence of contamination at this level.
       9J5   Initial precision and accuracy—To establish the ability to generate acceptable precision and accuracy,
            the analyst shall perform the following operations for compounds to be calibrated.
   704

-------
                                                                                              Method 1671
     9.5.1    Analyze two sets of four 5-mL aliquot* (eight aliquot* total) of the aqueous performance
             standard (Section 7,4.2) containing the PMI analytes listed in Table 1.
     9.5.2    Using the first set of four analyses, compute the average recovery (X) in percent of spike
             level and standard deviation of the recovery (s) in percent of spike level, for each compound.
     9.5.3    For each compound, compare s and X with the corresponding limits for initial precision and
             accuracy found in Table 3.  If s and X for all compounds meet the acceptance criteria, system
             performance is acceptable and analysis of blanks and samples may begin.  If, however, any
             individual s exceeds the precision limit or any individual X falls outside the range for accu-
             racy, system performance is unacceptable for that compound.
     9J5A    Using the results of the second set of analyses, compute s and X for only those compounds
             that failed the test of the first set of four analyses (Section 9.5.3).  If these compounds now
             pass, the system performance is acceptable for all compounds, and analysis of blanks and
             samples may begin. If, however, any of the same compounds fail again, the analysis system
             is not performing properly for the compound(s) in question.  In this event, correct the prob-
             lem and repeat the entire test (Section 9.5).
9.6  The laboratory shall, on an ongoing basis, demonstrate through the analysis of the aqueous perfor-
     mance standard (Section 7.4.2) that the analysis system is in control. This procedure is described in
     Section 13.
9.7  Where available, field replicates may be used to validate the precision of the sampling technique.
9.8  The laboratory shall maintain records to define the quality of data that is generated.

10.  Calibration

10.1  Inject standards into the GC and adjust the sensitivity to detect an amount of each compound less
     than or equal to one-third of the ML listed in Table 2 for the analyte.
10.2  Internal standard calibration procedure. The analyst must select one or more internal standards that
     are similar in analytical behavior to the compounds of interest.  The analyst must further demonstrate
     that the measurement of the internal standards) is not affected by method or matrix interferences.
     Because of these limitations, no internal standard that would be applicable to all samples can be
     required in the method. The method was developed using tetrahydrofuran (THF) as an internal stan-
     dard.  Where  THF is not present in the sample matrix and no interference precludes its use, THF is to
     be used as an internal standard for application of this method.
     10.2.1   Prepare aqueous calibration standards at a minimum of five concentration levels for each
             analyte by carefully adding an appropriate amount of secondary standard to reagent water or
             to the matrix under study. One of me concentrations should be at or below the ML.  The con-
             centration range should bracket the concentrations expected in the samples and should not
             exceed the dynamic range of the GC/FID instrument.  These aqueous standards  must be pre-
             pared daily.

     10.2.2   Prepare a spiking solution containing the internal standard(s) using the procedures described
             in Sections 7.3 and 1A and add an appropriate amount of internal standard to each aqueous
             calibration standard.

     10.2.3   Using injections appropriate to optimize system sensitivity and separation of the analytes,
             analyze each calibration standard and tabulate peak height or area responses against concen-
                                                                                                  105

-------
Method 1671
                    tration for each compound and internal standard. Calculate response factors (RF) for each
                    compound as follows:
                              where:
                             As = Response for the analyte to be measured
                             A £5 = Response for the nearest eluting internal standard
                             Qy = Concentration of the nearest eluting internal standard
                             Cs = Concentration of the analyte to be measured
       If the RF value over the working range is a constant (less than 10% relative standard deviation), the RF can
       be assumed to be invariant and the average RF can be used for calculations.  Alternatively, the results can
       be used to plot a calibration curve of response ratios, Ag/Ajg, against concentration.
       11.  Sample Preparation
       Samples containing less than 1% solids are analyzed directly as aqueous samples. Samples containing 1%
       solids or greater are analyzed after equilibration with reagent water containing internal standard(s).
       11 .1  Determination of percent solids.
            11 .1 .1   Weigh 5-10 g of sample into a tared beaker.
            11.1.2   Dry overnight (12 hours minimum) at 110°C (±5°C), and cool in a desiccator.
            1 1 .1 .3   Determine the percent solids as follows :
                                               weight of sample dry
                                  %sollds  =  weigh, of sample ve,*100

       11 .2  Remove standards and samples from cold storage and bring to 20-25°C.
       11.3  Samples containing less than 1% solids.
            1 1 .3.1   Allow sou' ds to settle and remove 5 mL of sample.
            11 .3.2   Add an appropriate amount of internal standard spiking solution.
            11.3.3   Inject a sample directly into the GC. The temperature of the injection block should be great
                    enough to immediately vaporize the entire sample. An example of the separations achieved
                    by the column listed is shown in Figure 1 .

       Note:    Use of a 0.2-yL injection has been found to improve method sensitivity over a larger injection
               combined with a split sample.  Where possible, splitless injection should be used. All require-
               ments of this Method must be met regardless of type of injection used. _
   106

-------
                                                                                             Method 1671
11.4 Samples containing 1% solids or greater.
     11.4.1   Mix the sample thoroughly using a clean spatula and remove rocks, twigs, sticks, and other
             foreign matter.
     11.4.2   Add 5 g (±1 g) of sample to a tared 10-mL centrifuge tube. Using a clean metal spatula,
             break up any lumps of sample. Record the sample weight to three significant figures.
     11.4.3   Add an appropriate amount of internal standard spiking solution to the sample in the centri-
             fuge tube.
     11.4.4   Add a measured quantity (Y ± 0.1 mL) of reagent water to the tube so as to minimize head
             space.
     11.4.5   Place a cap on the centrifuge tube leaving little or no headspace. Place the tube in the sonica-
             tor for a minimum of 5 minutes, turning occasionally. For most samples this should be suffi-
             cient time to distribute the analytes and standard(s) between the solid and aqueous phases
             and to establish equilibrium.  Some sample matrices may require more sonication.
     11.4.6   On completion of sonication, centrifuge the sample and inject the same amount of supemate
             into the GC that was injected for the calibration standards.
11.5 For liquid samples containing high-solids concentrations, such as sludges or muds, weigh approxi-
     mately 5 g (to three significant figures) into a  10-mL centrifuge tube,  add an appropriate amount of
     internal standard solution, sonicate, centrifuge, and inject as in Section 11.4.6.

12.  Quantitative Determination

12.1  The calibration curve or averaged response factor determined during calibration is used to calculate
     the concentration.  For calculation using the averaged RF, the equation below is used, and the terms
     are as defined in Section 10.2.3.
                                                   As X Cis
                                   Concentration^ -j	=-=,
                                                   Ais x RF

12.2 The concentration of the pollutant in the solid phase of the sample is  computed using the concentra-
     tion of the pollutant detected in the aqueous solution, as follows:
                                        YL x aqueous cone (mg/L)
            Concentration in solid ([ig/kg) = 	sample wt (kg)	x percent solids x DF


                      where:

                      percent solids is from Section 11.1
                        7 = Volume of water in liters (L)from 11.4.4

                      DF = Dilution factor (as a  decimal number), if necessary
12.3 Sample dilution—If the calibration range of the system is exceeded, the sample is diluted by succes-
     sive factors of 10 until the sample concentration is within the calibration range.

12.4 Report results for all pollutants found in standards, blanks, and samples to three significant figures.
     For samples containing less than 1% solids, the units are milligrams per liter (mg/L); and for samples
     containing 1% solids or greater, units are milligrams per kilogram (mg/kg).
                                                                                                107

-------
Method 1671
       13.  System Performance

       13.1  At the beginning of each 8-hour shift during which analyses are performed, system calibration and
            performance shall be verified. Acceptance criteria for each compound (R) are found in Table 3.
            Adjustment and/or recalibration shall be performed until all performance criteria are met. Only after
            all performance criteria are met may blanks and samples be analyzed.
       13.2  Where THF is used as the internal standard, the absolute retention time of THF shall be 416 seconds
            (± 30 seconds). The relative retention times of all pollutants shall fall within 10% of the value given
            in Table 2.

       14.  Method Performance

       14.1  This method was developed and validated in a single laboratory.
       14.2  A chromatogram of the aqueous performance standard is shown in Figure 1.

       15.  Waste  Management

       15.1  It is the laboratory's responsibility to comply with all federal, state, and local regulations governing
            waste management, particularly die hazardous waste identification rules and land-disposal restric-
            tions. In addition, it is the laboratory's responsibility to protect air, water, and land resources by min-
            imizing and controlling all releases from fume hoods and bench operations.  Also, compliance is
            required with any sewage discharge permits and regulations.
       15.2  Samples containing acids at a pH of less than 2 are hazardous and must be neutralized before being
            poured down the drain or must be handled as hazardous waste.
       15.3  For further information on waste management consult "The Waste Management Manual for Labora-
            tory Personnel" and "Less is Better—Laboratory Chemical Management for Waste Reduction,"
            available from the American Chemical Society's Department of Government Relations and Science
            Policy, 1155 16th Street, N.W., Washington, DC 20036.
   108

-------
                                                                                       Method 1671
                                       References

1.    "Standard Test Method for Volatile Alcohols in Water by Direct Aqueous-Injection Gas Chromatog-
     raphy." 1994 Annual Book of ASTM Standards, Volume 11.02 (Water (II)). ASTM, 1916 Race
     Street, Philadelphia, PA 19103-1187.

2.    "Working with Carcinogens," DREW, PHS, NIOSH, Publication 77-206 (1977).

3.    "OSHA Safety and Health Standards, General Industry," 29 CFR 1910, OSHA 2206 (1976).

4.    "Safety in Academic Chemistry Laboratories," American Chemical Society Publication, Committee
     on Chemical Safety (1979).

5.    "Methods 330.4 and 330.5 for Total Residual Chlorine," USEPA, EMSL Cincinnati, OH 45268.

6.    "Method of Analytical Quality Control in Water and Wastewater Laboratories," USEPA, EMSL Cin-
     cinnati, OH 45268, EPA-4-79-019 (March, 1979).
                                                                                          109

-------
Method 1671
        Table 1. Non-Purgeable Water Soluble PMI Analytes to be Analyzed by Direct Aqueous Injection GC/FID
        and Internal Standard Techniques
        1
                             PMI Analyte
            Acetonitrile
            Diethylamine
            Dimethylamine
            Dimethyl sulfoxide
            Ethanol
            Ethylene glycol
            Formamide
            Methanol
            Methylamine
            Methyl cellosolve (2-methoxyethanol)
            n-Propanol
            Tn'ethylamine
Chemical Abstracts Service Registry Number
                                                                        CASRN
                                                                        75-05-8
                                                                       109-89-7
                                                                       124-40-3
                                                                        67-68-5
                                                                        64-17-5
                                                                       107-21-1
                                                                        75-12-7
                                                                        67-56-1
                                                                        74-89-5
                                                                       109-86-4
                                                                        71-23-8
                                                                       121-44-8
                                                                              i1
EPA-EGD
     972
     986
     987
    1037
     134
    1038
     988
     135
     989
    1040
     955
     990
   110

-------
                                                                                                         Method 1671
Table 2. Gas Chromatographic Retention Times and Minimum Levels for Non-purgeable Water Soluble PMI
Analytes by Direct Aqueous Injection GC/FID
 EGD
 No.                    PMI Analyte
 989        Methylamine
 135        Methanol
 987        Dimethylamine
 134        Ethanol
 972        Acetonitrile
 955        n-Propanol
 986        Diethylamine
 975        Tetrahydrofuran(intstd)
 1040       Methyl cellosolve (2-Methoxyethanol)
 988        Fomnamide
 1038       Ethylene glycol
 990        Triethylamine
 1037       Dimethyl sulfoxide
                                                                  Retention Time
ML1
Mean (sec)
128
139
165
188
203
307
341
416
429
473
495
518
676
EGD Ref
975
975
975
975
975
975
975
975
975
975
975
975
975
Relative
0507
0334
0396
0.452
0.488
0.737
OB19
1.000
1.030
1.136
1.189
1244
1.624
(mg/L)
50
50
50
50
50
50
50

20
100
100
50
20
1
 This is a minimum level at which the entire analytical system shall give an acceptable calibration point, taking into account
 method-specific sample and injection volumes. The concentration in the aqueous or solid phase is determined using the equa-
 tions in Section 12.
                                                                                                             111

-------
Method 1671
        Table 3. Acceptance Criteria for Performance Tests
                                                                         Acceptance Criteria (% of Spike Level)
                                                                      Initial Precision and Accuracy
           EGDNo.            PMIAnalyte              Spike Level
                972 Acetonitrile                             50
                986 Diethylamine                            50
                987 Dimethylamine                          50
               1037 Dimethyl sulfoxide                        50
                134 Ethanol                                50
               1038 Ethylene glycol                         100
                988 Formamkte                            200
                135 Methanol                               50
                989 Methylamine                            50
               1040 Methyl celtosolve (2-Methoxyethanol)        50
                955 n-Propanol                             50
                990 Triethylamine                            50
  s
30.2
14.8
27.2
 6.1
18.6
21.6
12.3
20.4
18.0
13.6
24.9
46.1
   X
85-146
65-100
98-153
31-100
93-131
100-149
96-121
79-121
79-116
 64-92
87-137
72-165
Ongoing
Accuracy
   R
 83-148
 63-100
 96-155
 30-100
 92-132
100-150
 95-122
 78-122
 78-117
  64-93
 85-139
 68-168
    112

-------
                                                                                       Method 1671
  100% -i
%FS   -
                I/
                                         Analysis: 4 MDL3,8,1
                                         Created at 10:36 on 01/Dec/94
                                         Sample*: 8
                                         Injections: 1
                                         Sample Name :MDL #3
             2     345678     9    10    11     12
                   Rgure 1.  Chromatogram of Aqueous Performance
                             Standard of Analytes from Table 1
                                                                             52-025-17
                                                                                          113

-------
                      Method 1673

Polyethylene glycol)-600 by Derivatization and
       High-Pressure Liquid Chromatography
                                 January 1995

-------
                                  Method 1673
                Polyethylene glycol)-600 by Derivatization and
                      High-Pressure  Liquid Chromatography
1.   Scope and Application

1.1   This method is designed to meet the survey and monitoring requirements of the EPA's Engineering
     and Analysis Division (BAD).  The method is used to determine poly(ethylene glycol)-600 (PEG-
     600, Chemical Abstracts Service Registry Number 25322-68-3),  a substance used in the pharmaceu-
     tical manufacturing industry (PMI), that can be derivatized and analyzed by high pressure liquid
     chromatography (HPLC).
1.2   PEG-600 is a mixture of oligomers with a molecular weight centered around 600 Da. The exact
     composition may vary from manufacturer to manufacturer and even between batches from the same
     manufacturer. This method has been developed for aqueous samples and is not, in its present form
     applicable to solids or sludges.
1.3   The detection limits of the method are usually dependent on the level of interferences rather than
     instrumental limitations. The detection limit provided in Table 1  is the minimum level that can be
     reliably quantified by this method with no interferences present.
1.4   This method is for use only by analysts experienced with HPLC or under the close supervision of
     such qualified persons.

2.   Summary of the Method

2.1   One liter of aqueous sample is placed into a liquid-liquid extractor and a known quantity of surrogate
     is added. Extraction with dichloromethane is carried out over an 18-hour period.  The dichlo-
     romethane extracts are dried over anhydrous sodium sulfate, evaporated to a small volume and dried
     again. Remaining dichloromethane is removed and the  water-free extract is derivatized using 3,5-
     dinitrobenzoyl chloride and pyridine in tetrahydrofuran. The tetrahydrofuran solution is diluted with
     diethyl ether, extracted to remove side products, evaporated, and solvent exchanged with acetonitrile/
     water and chromatographed on a reverse-phase C18 column with a solvent gradient of 40 percent ace-
     tonitrile/water to 100 percent acetonitrile. Detection is performed at 254 nanometers. The PEG-600
     derivative is identified by its retention time relative to that of the derivatized surrogate and quantified
     by external standard techniques. Derivatized samples must be stored at 4°C in an amber container
     and analyzed within 96 hours of preparation.

2.2   The quality of the analysis is assured through reproducible calibration and testing of the derivatiza-
     tion/extraction procedure and the HPLC system.
1.    This Chemical Abstracts Service Registry Number (CASRN) is common to all poly(ethylene glycol) oligomers
     and mixtures and does not specifically identify PEG-600.
                                                                                            117

-------
Method 1673
       3.   Definitions

       There are no method specific definitions to be noted for this document.

       4.   Interferences

       4.1   Method interferences may be caused by contaminants in solvents, reagents, glassware, and other
            sample processing hardware that lead to discrete artifacts and/or elevated baselines in chromato-
            grams. All of these materials must be routinely demonstrated to be free from interferences under the
            conditions of the analysis by analyzing laboratory reagent blanks as described in Section 9.3.
            4.1.1    Glassware must be scrupulously cleaned. Clean all glassware as soon as possible after use
                    by rinsing with the last solvent used. This should be followed by detergent washing with hot
                    water,  and rinses with tap water and reagent water.  It should then be drained, dried, and
                    heated in a laboratory oven at 130 degrees celsius (°C) for several hours before use. Solvent
                    rinses  with acetone may be substituted for the oven heating. After drying and cooling, glass-
                    ware should be stored in a clean environment to prevent any accumulation of dust or other
                    contaminants.
            4.1.2    The use of high purity reagents and solvents helps to minimize interference problems. Purifi-
                    cation  of solvents by distillation in all glass systems may be required.
       4.2   Matrix interferences may be caused by contaminants that are coextracted from the sample.  The
            extent of matrix interferences will vary considerably from source to source, depending upon the
            nature and diversity of the matrix being sampled. If matrix interferences occur, some additional
            cleanup may be necessary.
       4.3   The extent of interferences that may be encountered using liquid chromatographic techniques has not
            been fully assessed.  Although the HPLC conditions described allow for resolution of PEG-600,
            other matrix components may interfere.

       5.   Safety

       5.1   The toxicity or carcinogenicity of each compound or reagent used in this method has not been pre-
            cisely determined; however, each chemical compound should be treated as a potential health hazard.
            Exposure to these compounds should be reduced to the lowest possible level. The laboratory is
            responsible for maintaining a current awareness file of OSHA regulations regarding the safe han-
            dling of the chemicals specified in this method. A reference file of Material Safety Data Sheets
            should also be made available to all personnel involved in these analyses. Additional information on
            laboratory safety can be found in References 1 through 3.

       6.   Equipment and Supplies

       6.1   Reaction vessel—5-mL screw-cap vial with V-shaped chamber (Aldrich Zl 8,302-4, or equivalent)
            with polytetrafluoroethylene (PTFE) lined cap (Aldrich Zll,511-8, or equivalent).
       6.2   Liquid-liquid extractor—Capable of extracting 1 L of sample (Aldrich Zl 0,156-7, or equivalent)

       6.3   Kudema-Danish (K-D) apparatus.
            6.3.1   Concentrator tube—10-mL graduated (Kontes K-570050 or equivalent). A ground-glass
                   stopper is used to prevent evaporation of extracts.
   118

-------
                                                                                             Method 1673
     6.3.2   Evaporation flask — 500-mL (Kontes K-57000 1-500 or equivalent).  Attach to concentrator
            tube with springs, clamps, or equivalent.
     6.3.3   Snyder column— Three-ball macro (Kontes K-503000-0121 or equivalent).
     6.3.4   Snyder column — Two-ball micro (Kontes K569001-0219 or equivalent).
     6.3.5   Springs — One-half inch (Kontes K-662750 or equivalent).
6.4  Vials and bottles — 10-mL, 25-mL, 100-mL, and 1-L amber glass with PTFE-lined screw-caps.
6.5  Boiling chips — Solvent-extracted with methylene chloride, approximately 10/40 mesh (silicon car-
     bide or equivalent).
6.6  Balance — Analytical, capable of accurately weighing to the nearest 0. 1 mg.
6.7  High-performance liquid chromatograph (modular).
     6.7.1   Pumping system — Solvent programmable, with flow control capable of 2.00 milliliter per
            minute (mL/min).
     6.7.2   High-pressure injection valve with 50-microliter (uL) loop or autosampler.
     6.7.3   Column — 250 millimeter (mm) long x 4.6 mm inside diameter (ID), 5 jjm particle size, Beta-
            sil C18 (or equivalent).
     6.7.4   Absorbance detector — 254 nm.
     6.7.5   Strip chart recorder compatible with the detector. Use of a data system is recommended.
6.8  Glass-fiber filter paper, 0.6-0.9 urn.
6.9  Pipette — Capable of accurately delivering 0.1-1.0 mL of solution.
6.1 0 Water bath — Heated, with concentric ring cover, capable of temperature control of ± 2°C. The bath
     should be used under a hood.
6.11  Sand bath — Heated, with temperature control of ± 5°C.
6.12 Volumetric flasks — 100-mL, 1-L.
6.13 Microsyringes — lOO-pL,
7.   Reagents and Standards
7.1   Reagent grade or pesticide grade chemicals shall be used in all tests. Other grades may be used, pro-
     vided it is first ascertained that the reagent is of sufficiently high purity to permit its use without less-
     ening the accuracy of the determinations.
7.2   Reagent water:  Water in which the compounds of interest and interfering compounds are not
     detected by this method. It may be generated by any of the methods in this subsection.
     7.2.1   Activated carbon: pass tap water through a carbon bed (Calgon Filtrasorb-300, or equiva-
            lent).
     7.2.2   Water purifier:  Pass tap water through a purifier (Millipore Super Q, or equivalent).
     7.2.3   Boil and purge: Heat tap water to between 90 and 100°C and bubble contaminant free inert
            gas through it for approximately 1 hour. While still hot, transfer the water to screw-cap bot-
            tles and seal with  a PTFE-lined cap.
7.3   Dichloromethane — Pesticide grade or equivalent.
                                                                                                119

-------
Method 1673
       7.4  Acetonitrile—Pesticide grade or equivalent.
       7.5  Diethyl ether—Pesticide grade or equivalent.
       7.6  Tetrahydrofuran, anhydrous—Can be prepared by distillation from potassium benzophenone ketyl,
            or other, similar methods.
       7.7  Surrogate—Di(ethylene glycol) monohexyl ether (CASRN 112-59-4)
       7.8  Sodium sulfate, anhydrous.
       7.9  Hydrochloric acid solution—Dilute 100 mL of concentrated hydrochloric acid to approximately one
            liter with reagent water.
       7.10 Sodium bicarbonate solution—Dissolve 10 g of sodium bicarbonate in approximately 1 L of reagent
            water.
       7.11  Saturated sodium chlroide solution—Prepare a saturated solution in reagent water by adding reagent
            sodium chloride until no more will dissolve at room temperature.
       7.12 Pyridine, anhydrous (CASRN 110-86-l)(Aldrich 27,097-0, or equivalent).
       7.13 3,5-Dinitrobenzoyl chloride (CASRN 99-33-2) in anhydrous tetrahydrofuran (10 mg/mL)—Weigh
            1 g of 3,5-dinitrobenzoyl chloride into a 100-mL volumetric flask and fill to the mark with anhydrous
            tetrahydrofuran.  Store the resulting solution in an amber bottle protected from moisture.
       7.14 Stock standard solutions.
            7.14.1  Stock PEG-600 (approximately 10.0 mg/mL)—Prepare by weighing 1.0 g of PEG-600 into a
                   100-mL volumetric flask and diluting to volume with tetrahydrofuran (if aliquots of this
                   solution are to be derivatized, then anhydrous tetrahydrofuran must be used).
            7.14.2  Secondary standard—1.0,2.5,5.0,7.5 and 10.0 mg/L.  Measure 0.1,0.25,0.5,0.75, and 1.0
                   mL of stock solution (Section 7.14.1) into separate 1-L volumetric flasks and dilute to vol-
                   ume with reagent water.  Store in 1-L amber bottles.
            7.14.3  Surrogate standard—Prepare by weighing 1.0 g of di(ethylene glycol) monohexyl ether into
                   a 100-mL volumetric flask and diluting to volume with tetrahydrofuran (if aliquots of this
                   solution are to be derivatized, then anhydrous tetrahydrofuran must be used). Store in a 100-
                   mL amber bottle protected from moisture.
            7.14.4  Stock standard solution and surrogate solution must be replaced after six months, or sooner,
                   if comparison with check standards indicates a problem.
            7.14.5  Aqueous performance standard—An aqueous performance standard containing PEG-600 at
                   2.5 mg/L and surrogate at Img/L, shall be prepared daily, and analyzed each shift to demon-
                   strate performance (Section 9).

       8.    Sample Collection, Preservation, and Storage

       8.1   Grab samples are collected in glass containers having a total volume greater than one liter.  Fill sam-
            ple bottles so that no air bubbles pass through the sample as the bottle is filled and seal each bottle so
            that no air bubbles are entrapped. Maintain the hermetic seal on the sample bottle until time of anal-
            ysis.
       8.2   Samples are maintained at 0-4°C from the time of collection until analysis. Samples must be
            extracted within five days of collection, derivatized within seven days of extraction, and analyzed
            within four days of derivatization.
   120

-------
                                                                                             Method 1673
9.   Quality Control

9.1   Each laboratory that uses this method is required to operate a formal quality assurance program (Ref-
     erence 4). The minimum requirements of this program consist of an initial demonstration of labora-
     tory capability and analysis of standards and blanks as tests of continued performance. Laboratory
     performance is compared to established performance criteria to determine if the results of analyses
     meet the performance characteristics of the method.
     9.1.1    The analyst shall make an initial demonstration of the ability to generate acceptable accuracy
             and precision with this method. This ability is established as described in Section 9.2.
     9.1.2    In recognition of advances that are occurring in analytical technology, and to allow the ana-
             lyst to overcome sample matrix interferences, the analyst is  permitted certain options to
             improve separations or lower the costs of measurements. These options include alternate
             extraction, concentration, cleanup procedures, and changes in columns and detectors. Alter-
             native techniques, such as substitution of immunoassay, and changes that degrade method
             performance are not allowed. If an analytical technique other than the techniques specified
             in this method is used, that technique must have a specificity equal to or better than the spec-
             ificity of the techniques in this method for the analytes of interest.
             9.1.2.1   Each time a modification is made to this method, the analyst is required to repeat
                     the procedure in Section 9.2. If the detection limit of the method will be affected
                     by the change, the laboratory is required to demonstrate that the method detection
                     limit (MDL) (40 CFR Part 136, Appendix B) is lower than one-third the regulatory
                     compliance level. If calibration will be affected by the change, the analyst must
                     recalibrate the instrument per Section 10.
             9.1.2.2   The laboratory is required to maintain records of modifications made to this
                     method.  These records include the information below, at a minimum.
                     9.1.2.2.1   The names, titles, addresses, and telephone numbers of the analyst(s)
                               who performed the analyses and modification, and of the quality con-
                               trol officer who witnessed and will verify the analyses and modifica-
                               tion.
                     9.1.2.2.2   A Listing of pollutant(s) measured, by name and CAS Registry Number.
                     9.1.2.2.3   A narrative stating the reason(s) for the modification.
                     9.1.2.2.4   Results from all quality control (QC) tests comparing the modified
                               method to this method, including:
                               (a)  calibration (Section 10);
                               (b) calibration verification  (Section 10.1.3.1);
                               (c)  initial precision and accuracy (Section 9.2);
                               (d) analysis of blanks (Section 9.3);
                               (e)  accuracy assessment (Section 9.5).
                     9.1.2.2.5   Data that will allow an independent reviewer to validate each determi-
                               nation by tracing the instrument output (peak height, area, or other sig-
                               nal) to the final result.  These data are to include:
                               (a)  sample numbers and other identifiers;
                               (b) extraction dates;
                                                                                                 121

-------
Method 1673
                                       (c)  analysis dates and times;
                                       (d)  analysis sequence/run chronology;
                                       (e)  sample weight or volume;
                                       (f)  extract volume prior to each cleaning step (Section 11.1.2);
                                       (g)  final extract volume prior to injection;
                                       (h)  injection volume;
                                       (I)  dilution data, differentiating between dilution of a sample or an
                                           extract;
                                       (j)  instrument and operating conditions;
                                       (k)  column and operating conditions (nature of column, dimensions,
                                           flow rates, solvents, etc.)
                                       (I)  detector operating conditions (wavelength, etc.)
                                       (m)  chromatograms, printer tapes, and other recording of raw data; and
                                       (n)  quantitation reports, data system outputs, and other data necessary
                                           to fink raw data to the results reported.
            9.1.3    Analyses of blanks are required tn demonstrate fregdnm from rnntaminarirm and that PFfi-
                    600 and interfering compounds have not been carried over from a previous analyses (Section
                    4). The procedures and criteria for analysis of a blank are described in Section 93.
            9.1 A    The laboratory shall, on an ongoing basis, demonstrate through the analysis of the aqueous
                    performance standard (Section 7.14.5) mat the analysis system is in control  This procedure
                    is described in Section 10.
            9.1.5    The laboratory shall maintain records to define the quality of data that is generated.
       9.2  Initial precision and accuracy—To establish the ability to generate acceptable precision and accuracy,
            the analyst shall perform the following operations for compounds to be calibrated:
            9.2.1    Analyze f our afiquots of the aqueous performance standard (Section 7.14 J) according to the
                    method beginning in Section 11.
            9^2    Using results from Section 9.2.1, compute the average recovery (X) in percent of spike level
                    and the standard deviation of the recovery (s) in percent of the spike level for PEG-600 and
                    the surrogate.
            9^3    For each compound, compares  and Xwith the o)rrespondmg limits for initial precision and
                    accuracy found in Table 2. If s and X for all compounds meet the acceptance criteria, system
                    performance is acceptable and analysis of blanks and samples may begin. If, however, any
                    individual s exceeds the precision limit or any individual X falls outside the range for accu-
                    racy, system performance is unacceptable for that compound.  This is an indication that the
                    analytical system is not performing properly for the compound(s) in question. In this event,
                    correct the problem and repeat the entire test (Section 92.1).
       93   Blanks—Reagent water blanks are analyzed to demonstrate freedom from contamination.
            9.3.1    With each sample batch (samples analyzed on the same 8-hour shift), a blank shall be carried
                    through the extraction and derivatization procedure and be analyzed immediately after anal-
                    ysis of the aqueous performance standard (Section 9.1.4) to demonstrate freedom from con-
                    tamination. If PEG-600 or any potentially interfering compound is found in a blank at
                    greater than 200 ug/L, analysis of samples is halted until the source of contamination is elim-
                    inated and a blank  shows no evidence of contamination at this level.
   122

-------
                                                                                             Method 1673
9.4  The specifications contained in this method can be met if the apparatus used is calibrated properly,
     men maintained in a calibrated state.  The standards used for calibration (Section 10.1), calibration
     verification (Section 10.13.1) and for initial (Section 9.2) and ongoing (Section 9.1.4) precision and
     accuracy should be identical, so that the most precise results will be obtained.
9.5  Depending on specific program requirements, field replicates may be collected to determine the pre-
     cision of the sampling technique, and spiked samples may be required to determine the accuracy of
     the analysis

10.  Calibration

10.1  Establish liquid chromatographic operating parameters to produce a retention time equivalent to that
     given in Section 12.2.1. Prepare derivatized calibration standards according to the procedure in Sec-
     tion 10.1.1. Calibrate the chromatographic system using the external standard technique (Section
     10.1.3).
     10.1.1  Preparation of calibration standards. Prepare calibration standards by adding one mL of sur-
            rogate standard (Section 7.143) to each of the secondary standards (Section 7.14.2).
     10.12  Process each calibration standard solution through the extraction, concentration, and deriva-
            tization procedures described in Section 11.
     10.1.3  External standard calibration procedure. Analyze each derivatized calibration standard using
            the chromatographic conditions specified in Section 12.1, and tabulate peak area against con-
            centration injected.  The results may be used to prepare calibration curves for PEG-600
            10.1.3.1   The working calibration curve must be verified at the beginning of each 8-hour
                     shift by the measurement of one or more calibration standards. If the response for
                     any anaryte varies from the previously established responses by more than 10 per-
                     cent, the test must be repeated using a fresh calibration standard after it is verified
                     that the analytical system is in control Alternatively, a new calibration curve may
                     be prepared. If an autosampler is available, it is convenient to prepare a calibration
                     curve daily by analyzing standards along with test samples.

11.  Sample Extraction and Defivatization

11.1  Extraction of samples and standards.
     11.1.1  Place one liter of sample and 1 mL of surrogate standard (Section 7.143) or 1 L of calibra-
            tion standard (Section 10.1.1) in the liquid-liquid extractor and extract with pesticide grade
            dichloromethane for 18 hours.
     11.1.2  Dry the dichloromethane solution over anhydrous sodium sulfate and evaporate off the sol-
            vent using the Kudema-Danish procedure. Dry again over anhydrous sodium sulfate when
            the volume reaches  10-25 mL and use  a gentle stream of dry nitrogen to removemost of the
            remaining solvent Quantitatively transfer the residue to a V-shaped reaction vial using
            anhydrous dichloromethane or anhydrous tetrahydrofuran and remove the last of the solvent
            with a stream of dry nitrogen.
11.2  Derivatization
     112.1  After ensuring that the extract is free of water, add5 mL of derivatization solution (Section
            7.13) to the vial and two drops of anhydrous pyridine (Section 7.12).
     11.22  Heat the vial and contents in a sand bath at 60°C (±5°Q for 1 hour.
                                                                                                123

-------
Method 7673
            11.2.3  Cool the vial and quantitatively transfer the contents to a 125-mL separatory funnel. Add 50
                   mL of diethyl ether (ether) and sequentially extract with two 25-mL portions of dilute hydro-
                   chloric acid (Section 7.9), then two 25-mL portions of reagent water, then two 25-mL por-
                   tions of sodium bicarbonate solution (Section 7.10) and finally with two 25-mL portions of
                   saturated sodium chloride solution (Section 7.11). Take care not to lose any ether solution
                   during the extraction procedure.

            11.2.4  Place a small plug of glass wool in a funnel and add approximately 10 g of anhydrous
                   sodium sulfate to the funnel. Drain the ether solution through the sodium sulfate. If neces-
                   sary use a clean spatula to break up any lumps of sodium sulfate in the funnel, then rinse the
                   separatory funnel with two 10-mL portions of ether and drain through the anhydrous sodium
                   sulfate in the funnel.

            11.2.5  Quantitatively transfer the ether solution to a clean Kudema-Danish apparatus and evaporate
                   most of the solvent (alternatively, dry nitrogen can be used to evaporate the ether while
                   warming on a steam bath).
            11.2.6  Perform a solvent exchange with 40% acetonitrile in water, adjust the volume to 2 mL and
                   filter, if necessary, for analysis.

       12.  High-Pressure Liquid Chromatography

       12.1  Chromatographic conditions.

            Column: Betasil C18,250 mm long x 4.6 mm ID, 5-|jm particle size (Keystone 255-701, or equiva-
            lent).

            Mobile Phase: 40% acetonitrile/water to 100% acetonitrile over a period of 20 minutes.
            Flow Rate: 2.0 mL/min
            UV Detector: 254 nm
            Injection Vol.: 50 \iL
       12.2  Analysis.

            12.2.1   Analyze samples by HPLC, using conditions described in Section 12.1. The retention time
                   of the PEG-600 derivative relative to the surrogate derivative is centered about 0.63. The ML
                   achievable in reagent water is 1 mg/L. Other HPLC columns, chromatographic conditions,
                   or detectors may be used if the requirements of Section 9 are met.
            12.2.2   Because PEG-600 is a mixture of poly(ethylene glycol) oligomers, the exact nature of PEG-
                   600 samples from various manufacturers and different batches from a single manufacturer,
                   may vary. For this reason, concentrations of PEG-600 in a specific waste stream are best
                   determined when standards are prepared using the same batch of PEG-600 in use by the
                   pharmaceutical manufacturer at the time of discharge of the waste stream under analysis.
                   Where it is not possible to obtain such a sample, adequate results can be obtained by the  use
                   of a PEG-600 product as a standard that is  unrelated to the one in use by the pharmaceutical
                   manufacturer, and careful definition of an "elution range" for derivatized PEG-600 in both
                   the external standards and the samples. (See Section 12.2.3)
            12.2.3   An "elution range" or retention time window is defined as a characteristic period of time dur-
                   ing which the derivatized PEG-600 elutes from the chromatographic column.  This range
                   should encompass at least 90 percent of the PEG-600 derivative in both the standard and the
                   sample. The width of the retention time window used for quantitation should be based upon
   124

-------
                                                                                            Method 1673
             measurements of actual retention time variations of standards over the course of a day.
             Three times the standard deviation of the retention time for a compound can be used to cal-
             culate a suggested window size; however, the experience of the analyst should weigh heavily
             in the interpretation of chromatograms.

      12.2.4   If the peak area exceeds the linear range of the calibration curve, a smaller sample volume
             may be injected. Alternatively, the final solution (Section 11.2.6) may be diluted with 40%
             acetonitrile/water, as appropriate, and reanalyzed.

12.3  Calculations.

      12.3.1   Calculate each response factor (RF) as follows (mean value based on 5 points):

                                       concentration of standard
                               Rr = -
                                          area of the signal
                                                        5
                                      mean RF = RF =
     12.3.2   Calculate the concentration of PEG-600 as follows:

                       mg/mL = RF x area of signal x concentration factor

                      where:


                                               final volume of extract
                            concentratwn factor = initial sample volume


13.  Method Performance

13.1  The ML for PEG-600 was obtained using reagent water.
13.2  This method has been tested for linearity of recovery from spiked reagent water and has been demon-
     strated to be applicable over the range from the ML to 30 times the ML.
13.3  A representative chromatogram is presented as Figure 1 .

14.  Waste Management

14.1  It is the laboratory's responsibility to comply with all federal, state, and local regulations governing
     waste management, particularly the hazardous waste identification rules and land disposal restric-
     tions.  In addition it is the laboratory's responsibility to protect air, water, and land resources by min-
     imizing and controlling all releases from fume hoods and bench operations. Also, compliance is
     required with any sewage discharge permits and regulations.
14.2  Samples containing acids at a pH of less than 2 are hazardous and must be neutralized before being
     poured down a drain or must be handled as hazardous waste.

14.3  For further information on waste management, consult "The Waste Management Manual for Labora-
     tory Personnel," available from the American Chemical Society's Department of Government Rela-
     tions and Science Policy, 115 16th Street, NW, Washington, DC 20036
                                                                                               125

-------
Method 1673
                                             References

       1.    "Working with Carcinogens," DHEW, PHS, NIOSH, Publication 77-206 (1977).

       2.    "OSHA Safety and Health Standards, General Industry," 29 CFR 1910, OSHA 2206, (1976).

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

       4.    "Handbook of Analytical Quality Control in Water and Wastewater Laboratories," USEPA, EMSL
            Cincinnati, OH 45268, EPA-4-79-019 (March 1979).
   126

-------
                                                                                                         Method 1673
Table LPMIAnalyte
        PMIAnalyte                                         CASRN1             Minimum Level2
        Polyethylene glycol)-600                             25322-68-3               1 mg/L
        (PEG-600)
1  Chemical Abstracts Service Registry Number.
2  This is the minimum level at which the entire analytical system shall give a recognizable signal and an acceptable calibration
  point.
Table 2. Requirements for Initial Precision and Accuracy
                                                                                              Percent
                                                         Spike        Average Percent    Standard Deviation
 PMI Analyte and Surrogate                             Level (ug/L)       Recovery (X)             (s)
 PEG-600                                                1058               22-79                28
 Di(ethylene glycol) monohexyl ether                          215               d-102                53
 d = Detected (analyte must be detected)
                                                                                                             127

-------
Method 1673
             2.00 -,
                                             Reverse Phase Column
                                             40% ACN/HOH to 100% ACN
                                             Calibration Std.
                                             1 ppm w/surrogate
             1.50 -
             1.00 -
            0.50 -
                           Rgure 1.  Chromatogram of the Aqueous Performance Standard
                                                                                               62425-18
   128

-------