EPA Document No. 815-B-08-001
METHOD 314.2   DETERMINATION OF PERCHLORATE IN DRINKING WATER USING
                 TWO-DIMENSIONAL ION CHROMATOGRAPHY WITH SUPPRESSED
                 CONDUCTIVITY DETECTION
                                    Version 1.0
                                    May 2008
Herbert P. Wagner (Lakeshore Engineering Services, Inc.)
Barry V. Pepich (Shaw Environmental, Inc.)
Douglas Later, Chris Pohl, Kannan Srinivasan, Brian De Borba, Rong Lin (Dionex Corp. )
David J. Munch (U.S. EPA, Office of Ground Water and Drinking Water)
                          TECHNICAL SUPPORT CENTER
               OFFICE OF GROUND WATER AND DRINKING WATER
                  U. S. ENVIRONMENTAL PROTECTION AGENCY
                             CINCINNATI, OHIO 45268
                                      314.2-1

-------
                                      METHOD 314.2

       DETERMINATION OF PERCHLORATE IN DRINKING WATER USING TWO-
    DIMENSIONAL ION CHROMATOGRAPHY WITH SUPPRESSED CONDUCTIVITY
                                        DETECTION

1.  SCOPE AND APPLICATION

   1.1   This is a large volume (2 to 4 mL), two-dimensional (2-D) ion chromatographic (1C) method
         using suppressed conductivity detection for the determination of perchlorate in raw and
         finished drinking waters.  Because this method utilizes two dissimilar 1C columns it does not
         require second column confirmation.  Detection and quantitation in the second dimension are
         accomplished by suppressed conductivity detection. Precision and accuracy data have been
         generated for perchlorate using this 2-D 1C method in reagent water, finished groundwater,
         finished surface water and a Laboratory Fortified Synthetic Sample Matrix (LFSSM).  The
         single laboratory Lowest Concentration Minimum Reporting Level (LCMRL) has also been
         determined in reagent water.l

                                                      Chemical Abstract Services
            Analvte                                   Registry Number (CASRN)
            Perchlorate                                        14797-73-0
   1.2   The Minimum Reporting Level (MRL) is the lowest analyte concentration that meets Data
         Quality Objectives (DQOs) that are developed based on the intended use of this method. The
         single laboratory LCMRL is the lowest true concentration for which the future recovery is
         predicted to fall between 50 and 150 percent recovery with 99% confidence.  The single
         laboratory LCMRL for perchlorate was 0.060 ug/L and 0.038 ug/L using 2.0-mL and 4.0-mL
         injection volumes, respectively. The procedure used to determine the LCMRL is described
         elsewhere.l

   1.3   Laboratories using this method will not be required to determine the LCMRL, but will need to
         demonstrate that their laboratory MRL for this method meets the requirements described in
         Section 9.2.4.

   1.4   Detection Limit (DL) is defined as the statistically calculated minimum concentration that can
         be measured with 99% confidence that the reported value is greater than zero.2 The DL for
         perchlorate is dependent on sample matrix, fortification concentration, and instrument
         performance. Determining the DL for perchlorate in this method is optional (Sect. 9.2.6).
         The reagent water DL for perchlorate using 2.0- and 4.0-mL injection volumes was calculated
         to be 0.018 and 0.012 ug/L, respectively, using 7 reagent water (RW) replicates fortified at
         0.075 and 0.025 ug/L.  These values are provided in Table 2.

   1.5   This method is intended for use by analysts skilled in the operation of 1C instrumentation, and
         the interpretation of the associated data.
                                            314.2-2

-------
    1.6   METHOD FLEXIBILITY - In recognition of technological advances in analytical systems
         and techniques, the laboratory is permitted to modify the separation technique, 1C columns,
         concentrator column and mobile phase composition. However, any modifications must
         maintain the basic chromatographic elements of this new technique.  This includes the initial
         separation of analytes on a primary 1C column that are heart-cut after suppression and directed
         to a concentrator column. After concentration, analytes must then be eluted onto a second,
         dissimilar 1C column for separation and quantitation. For high sensitivity applications, the
         system should have a second-dimension column with a lower cross sectional area, which
         operates at a relatively lower flow rate in order to achieve sensitivity enhancement
         proportional to the ratio of the flow rate reduction.  Method modifications should be
         considered only to improve method performance. Modifications that are introduced in the
         interest of reducing cost or sample processing time, but result in poorer method performance
         (see Sect. 9.4 for criteria), may not be used. In all cases where method modifications are
         proposed, the analyst must first redetermine the cut window and establish an acceptable
         calibration.  Then the analyst must demonstrate acceptable method performance by
         conducting the procedures outlined in the initial demonstration of capability (IDC, Sect. 9.2),
         verifying that all ongoing QC acceptance criteria can be routinely met (Tables 4 and 5), and
         properly documenting the method changes (Sect. 9.4). Changes may not be  made to sample
         collection and preservation (Sect. 8) or to the quality control requirements (Sect. 9).

2.  SUMMARY OF METHOD

    2.1   Water samples are collected in the field using a sterile filtration technique.  A 2.0 to  4.0-mL
         sample aliquot is injected onto a 4-mm 1C column. Separation of perchlorate is achieved in
         the first dimension (1-D) using 35 mM KOH at a flow rate of 1.0 mL per minute.
         Approximately 5-6 mL of the suppressed eluent containing the perchl orate is  diverted from
         the first dimension column to a concentrator column used in place of the  sample loop of the
         second dimension (2-D) injection valve.  The concentrator column has low  backpressure but
         sufficient capacity to trap the perchlorate ions quantitatively in the suppressed eluent.  In this
         manner, perchlorate is separated from other matrix ions and concentrated on a trapping
         column.  The heart-cut portion of the  1-D chromatogram is eluted off the concentrator column
         and onto a smaller diameter (2 mm diameter) guard and analytical column that have  different
         selectivity from the first dimension columns to facilitate the 2-D separation using 65 mM
         KOH at a flow rate of 0.25 mL per minute. Perchlorate is quantitated using the external
         standard method.

         There are several  advantages to this version of the method. Like Method 314.1, this version is
         compatible with a large sample injection volume (up to 4.0-mL). This is due to the high
         capacity of the 1-D analytical column and its higher selectivity for perchlorate relative to the
         other matrix ions.  The suppressed eluent from the first dimension column, which is
         essentially water containing the anions of interest and a subset of the matrix interferences that
         co-elute in the heart-cut window, are efficiently retained and refocused on the concentrator
         column.  Because the second dimension is operated at a lower flow rate relative to the first
         dimension and uses a smaller cross sectional area column, this version of the method offers
         enhanced sensitivity. The increase in sensitivity of the present setup is directly proportional  to
         the flow rate ratio of the first dimension to the second dimension.  This yields method
         sensitivity comparable to the EPA perchlorate methods that utilize mass spectrometric

                                            314.2-3

-------
         detection (331.0 and 332.0). Finally, the 2-D 1C method combines two columns with different
         selectivity thereby eliminating the need for second column confirmation.

3.  DEFINITIONS

   3.1    ANALYSIS BATCH - A sequence of field samples, which are analyzed within a 24-hour
          period and include no more than 20 field samples.  An Analysis Batch must also include all
          required QC samples, which do not contribute to the maximum field sample total of 20. For
          this method, the required QC samples include:

               Laboratory Reagent Blank (LRB)
               Continuing Calibration Check (CCC) Standards
               Laboratory Fortified Sample Matrix (LFSM)
               Laboratory Duplicate (LD) or a Laboratory Fortified Sample Matrix Duplicate
               (LFSMD)
               Laboratory Fortified Synthetic  Sample Matrix CCC Standard (LFSSM CCC)

   3.2    ANALYTE FORTIFICATION SOLUTIONS (AFS) - The Analyte Fortification Solutions
          are prepared by  dilution of the Analyte Secondary Dilution Solutions (SDS) and are used to
          fortify the LFSMs and the LFSMDs with perchlorate. It is recommended that multiple
          concentrations be prepared so that the fortification levels can be rotated or adjusted to the
          concentration of target analyte in the native samples.

   3.3    CALIBRATION STANDARD (CAL) - A solution of the target analyte prepared from the
          Perchlorate Primary Dilution Solution or Perchlorate Stock Standard Solution.  The CAL
          solutions are not sterile filtered and used  to calibrate the instrument response with respect to
          analyte concentration.

   3.4    CONTINUING  CALIBRATION CHECK STANDARD (CCC) - A calibration check
          standard containing the method analyte which is analyzed periodically throughout an
          Analysis Batch,  to verify the accuracy of the existing calibration for that analyte. The CCCs
          are not sterile filtered.

   3.5    DETECTION LIMIT (DL) - The minimum concentration of an analyte that can be
          identified, measured and reported with 99% confidence that the analyte concentration is
          greater than zero. This is a statistical  determination (Sect. 9.2.6), and accurate  quantitation is
          not expected at this level.2

   3.6    LABORATORY DUPLICATE (LD) - Two sample aliquots (LDi and LD2), from a single
          field sample bottle,  and analyzed separately with identical procedures. Analyses of LDi and
          LD2 indicate precision associated specifically with laboratory procedures by removing
          variation contributed from sample collection, preservation, and storage procedures.

   3.7    LABORATORY FORTIFIED BLANK (LFB) - An aliquot of reagent water or other blank
          matrix to which  a known quantity of the method analyte is added. The LFB is  analyzed
          exactly like a sample, including the preservation procedures in Section 8.1. Its purpose is to
          determine whether the methodology is in control, and whether the laboratory is capable of
          making accurate and precise measurements.
                                            314.2-4

-------
3.8    LABORATORY FORTIFIED SAMPLE MATRIX (LFSM) - An aliquot of a field sample to
      which a known quantity of the method analyte is added.  The LFSM is processed and
      analyzed exactly like a field sample, and its purpose is to determine whether the field sample
      matrix contributes bias to the analytical results.  The background concentration of the analyte
      in the field sample matrix must be determined in a separate aliquot and the measured value in
      the LFSM corrected for native concentrations.

3.9    LABORATORY FORTIFIED SAMPLE MATRIX DUPLICATE (LFSMD) - A second
      aliquot of the field sample used to prepare the LFSM, which is fortified and analyzed
      identically to the LFSM. The LFSMD is used instead of the Laboratory Duplicate to assess
      method precision and accuracy when the occurrence of the target analyte is infrequent.

3.10  LABORATORY FORTIFIED SYNTHETIC SAMPLE MATRIX (LFSSM) - An aliquot of
      the LSSM (Sect. 7.2.2) which is fortified with perchlorate. The LFSSM is used to set the
      start time for the cut window in the 1-D during the initial demonstration of capability (IDC)
      (Sect. 9.2).  The LFSSM is also used to determine the precision and accuracy of the method
      during the IDC (Sects. 9.2.2 and 9.2.3). The LFSSM samples are treated like the CCCs and
      are not sterile filtered.

3.11  LABORATORY FORTIFIED SYNTHETIC SAMPLE MATIX CONTINUING
      CALIBRATION CHECK STANDARD (LFSSM CCC) - An aliquot of the LSSM (Sect.
      7.2.2) which is fortified with perchlorate at a concentration equal to one of the CCCs. In this
      method, a LFSSM CCC at a concentration equal to the highest calibration level is analyzed at
      the end of each Analysis Batch (Sect. 9.3.3) to confirm that the first-dimension heart cutting
      procedure has acceptable recovery in high inorganic matrices. The LFSSM CCC is not sterile
      filtered.

3.12  LABORATORY REAGENT BLANK (LRB) -  An aliquot of reagent water or other blank
      matrix that is treated exactly as a sample including exposure to all filtration equipment,
      storage containers and internal standards. The LRB is used to determine if the method
      analyte or other interferences are present in the laboratory environment, the reagents, or the
      apparatus.

3.13  LABORATORY SYNTHETIC SAMPLE MATRIX (LSSM) - An aliquot of reagent water
      that is fortified with 1000 mg/L of the sodium salts of chloride, bicarbonate and sulfate. This
      solution is prepared from the LSSMSS and is representative of a water sample that contains
      3000 mg/L of these common matrix anions.

3.14  LABORATORY SYNTHETIC SAMPLE MATRIX STOCK SOLUTION (LSSMSS) - The
      LSSMSS contains the common anions chloride,  sulfate and bicarbonate at 25.0 g/L.  This
      solution is used in the preparation of QC samples.

3.15  LOWEST CONCENTRATION MINIMUM REPORTING LEVEL (LCMRL) - The single-
      laboratory LCMRL is the lowest true concentration for which the future recovery is predicted
      to fall between 50 and 150 percent recovery with 99% confidence. l
                                       314.2-5

-------
   3.16   MATERIAL SAFETY DATA SHEET (MSDS) - Written information provided by vendors
          concerning a chemical's toxicity, health hazards, physical properties, fire, and reactivity data
          including storage, spill,  and handling precautions.

   3.17   MINIMUM REPORTING LEVEL (MRL) - The minimum concentration that can be
          reported by a laboratory as a quantified value for the target analyte in a sample following
          analysis.  This defined concentration must meet the criteria defined in Section 9.2.4 and must
          be no lower than the concentration of the lowest calibration standard for the target analyte.

   3.18   PRIMARY DILUTION STANDARD SOLUTION (PDS) - A solution containing the
          method analyte prepared in the laboratory from stock standard solutions and diluted as
          needed to prepare calibration solutions and other analyte-containing solutions.

   3.19   QUALITY CONTROL  SAMPLE (QCS) - A solution containing the method analyte at a
          known concentration that is obtained from a source external to the laboratory and different
          from the source of calibration standards. The QCS is used to verify the accuracy of the
          calibration standards and the integrity  of the calibration curve.

   3.20   REAGENT WATER (RW) - Purified water which does not contain any measurable quantity
          of the target analyte or interfering compounds at or above 1/3  the MRL.

   3.21   SECONDARY DILUTION STANDARD SOLUTION (SDS) - A solution containing the
          method analyte prepared in the laboratory from the PDS and diluted as needed to prepare
          calibration solutions and other analyte solutions.

   3.22   STOCK STANDARD SOLUTION (SSS) - A concentrated solution containing the method
          analyte prepared in the laboratory using assayed reference materials or purchased from a
          reputable commercial source, so that the concentration and purity of analytes are traceable to
          certificates of analysis.

4.  INTERFERENCES

   4.1   Interferences can be divided into three different categories: (i) direct chromatographic co-
         elution, where an analyte response is observed at very nearly the  same retention time (RT)  as
         the target analyte; (ii) concentration dependant co-elution, which is observed when the
         response of higher than typical concentrations of the neighboring peak overlaps into the
         retention window of the target analyte;  and (iii) ionic character displacement, where retention
         times may significantly shift due to the influence of high ionic  strength matrices (high mineral
         content or total dissolved solids) overloading the exchange sites on the column and
         significantly shortening the target analyte's retention time.

       4.1.1  A direct chromatographic co-elution may be solved by changing column selectivity in
             one or both dimensions of the 2-D 1C method, adjusting eluent strength in one or both
             dimensions, modifying the eluent with organic solvents  (if compatible with 1C columns),
             changing the detection system, or selective removal of the interference with sample
             pretreatment.  Sample dilution will have little to no effect  on direct chromatographic co-
             elution. The analyst must verify that any change made to the chromatographic

                                            314.2-6

-------
              parameters does not induce any negative affects on method performance by repeating and
              passing all the QC criteria as described in Section 9.2.

       4.1.2   Sample dilution may resolve some of the difficulties if the interference is the result of
              either concentration dependant co-elution or ionic character displacement, but it must be
              clarified that sample dilution will alter the MRL by a proportional factor equivalent to
              that of the dilution. Therefore, careful consideration of Data Quality Objectives (DQOs)
              should be given prior to performing such a dilution.

   4.2   Method interferences may be caused by contaminants in solvents, reagents (including reagent
         water), sample bottles and caps, and other sample processing hardware that lead to discrete
         artifacts and/or elevated baselines in the chromatograms.  All items such as these must be
         routinely demonstrated to be free from interferences (less than  1/3 the perchlorate MRL) by
         analyses of an LRB as described in Section 9.2.1  Subtracting blank values from sample
         results is not permitted.

   4.3   Matrix interferences may be caused by contaminants that are present in the sample. The
         extent of matrix interferences will vary considerably from source to source, depending upon
         the nature of the water.  Water samples high in organic carbon or total dissolved solids may
         have elevated baselines or interfering peaks.

       4.3.1   Equipment and containers used for sample collection and storage have the potential to
              introduce interferences. The potential for interferences from these sources must be
              investigated during the IDC (Sect. 9.2) by preparing  and analyzing a LRB.  This
              procedure should be repeated each time that a new brand and/or lot of materials are used
              to ensure that contamination does not hinder analyte identification and quantitation.

   4.4   When first-dimension heart cut windows are properly set, this method demonstrates adequate
         performance in water matrices that contain up to 1000 mg/L of each common anion (sulfate,
         carbonate, and chloride). While this addresses the majority of source and finished drinking
         waters, there is a possibility that water matrices could exceed this level.  The analyst should
         monitor all first dimension chromatograms (Sect. 11.3.4) to confirm that sample matrix does
         not overload the primary column capacity and require dilution.

5.  SAFETY

   5.1   The toxicity or carcinogenicity of each reagent used in this method has not been precisely
         defined.  Each chemical should be treated as a potential health hazard, and  exposure to these
         chemicals should be minimized.  Each laboratory is responsible for maintaining an awareness
         of OSHA regulations regarding safe handling of chemicals used in this method.  A reference
         file of MSDSs should be made available to all personnel involved in the chemical analysis.
         Additional references to laboratory safety are available.3"5

6.  EQUIPMENT AND SUPPLIES (References to specific brands or catalog numbers are included for
         illustration only and do not imply endorsement of the product.  This does not preclude the  use
         of other vendors,  supplies or configurations.)
                                             314.2-7

-------
6.1    NON-STERILE SAMPLE CONTAINERS - 125-mL brown Nalgene bottles (Fisher Cat. No.
      03-313-3C or equivalent).

6.2    STERILE SAMPLE FILTERS (Corning 26-mm surfactant free cellulose acetate 0.2-jim filter,
      Fisher Cat. No. 09-754-13 or equivalent).  If alternate filters are used they should be certified
      as having passed a bacterial challenge test.6 Additionally, if alternate filters or different lots
      of the recommended filters are used, they must be tested using a LRB and a LFB fortified at
      the MRL as outlined in Section 9.2 to insure that they do not introduce interferences or retain
      perchlorate.

6.3    SYRINGES - Sterile, silicone free disposable syringes (Henke Sass Wolf 20-mL Luer lock,
      Fisher Cat. No. 14-817-33 or equivalent).

6.4    VOLUMETRIC FLASKS - Class A, suggested sizes include 10, 50, 100, 250, 500 and 1000
      mL for preparation of standards and eluents.

6.5    GRADUATED CYLINDERS - Suggested sizes include 25 and 1000 mL.

6.6    AUTO PIPETTES - Capable of delivering variable volumes from  1.0 to 2500 |iL.

6.7    ANALYTICAL BALANCE - Capable of weighing to the nearest 0.0001 g.

6.8    DUAL ION CHROMATOGRAPHY SYSTEM WITH SUPPRESSED CONDUCTIVITY
      DETECTION (1C) - This section describes the instrument configuration that was used to
      collect the data in Section 17. A Dionex Model ICS-3000 Dual system (or equivalent)
      consisting of a Dual Pump (DP) module, Eluent Generator (EG) module,
      Detector/Chromatography (DC) module (single or dual temperature zone configuration), and
      Autosampler (AS) was used to collect the  data presented in this method. The 1C system
      should also have a temperature controlled  column compartment and be capable of operating
      above  room temperature (30 °C) and include dual 1C pumps and all required accessories,
      including guard, analytical, and  concentrator columns, detector/chromatography module, dual
      eluent generators, continuously-regenerated anion trap columns, compressed gases,
      autosampler, suppressors, carbonate removal devices (CRD), dual conductivity detectors, and
      a computer-based data acquisition and control system.  Additionally, the system must be
      capable of performing automated,  two-dimensional 1C, including performing inline column
      concentration and matrix elimination steps as described in Section  2.1.  A schematic diagram
      of the instrumentation for this 2-D 1C method is shown in Figure 1. Table  1 provides full
      details of the instrumental conditions.

   6.8.1   DUAL PUMP MODULE - A DP Dual Gradient-Gradient Pumping Module with dual
          channel degas devices (Dionex DP, P/N 061712 or equivalent) was used to generate the
          data for this method.  The dual pump  system used for method  development was capable
          of supplying a flow rate of approximately 1.0 mL/min to the first dimension column and
          approximately 0.25 mL/min to the second dimension column.

   6.8.2   ELUENT GENERATOR MODULE - A dual channel EG Module (Dionex EG, P/N
          061714 or equivalent) with dual potassium hydroxide cartridges (EluGen® Cartridge,
          EGC, P/N 058900 or equivalent) was used to prepare the potassium hydroxide eluent for
                                       314.2-8

-------
      both the first and second dimensions of this method. An equivalent eluent generator may
      be used and/or manually prepared eluents may also be used provided that adequate
      resolution, peak shape, capacity, accuracy, and precision (Sect. 9.2) are obtained.  Care
      must be exercised with manually prepared hydroxide eluents to prevent formation of
      carbonate in the eluent from exposure to the atmosphere, which can dramatically alter the
      chromatography and affect sensitivity.

   6.8.2.1 CONTINUOUSLY REGENERATED ANION TRAP COLUMNS - 1C eluent
          purification columns (Dionex CR-ATC, P/N 060477 or equivalent). Any in-line,
          resin-based manual or electrolytic trapping column that provides adequate eluent
          purification for ultra trace analysis and performance (Sect. 9.2) may be used. A CR-
          ATC device, or equivalent, was used for eluent purification for both the first and
          second dimension eluents.

          NOTE:  For the configuration in Figure 1, the pump and eluent generator modules in
          combination must be capable of delivering different isocratic eluent concentrations to
          the columns in the first and second dimension.  The same requirement applies for
          manually prepared eluents with a dual pumping system; the pump system must be
          capable of delivering two different, independent isocratic concentrations of eluents to
          the first and second dimension columns. In addition, the system should also be
          capable of providing a step isocratic eluent concentration change or a controlled
          gradient change to the first dimension column.  This allows the first dimension
          column to be cycled to a higher eluent concentration in order to clean residual matrix
          components from the column prior to introduction of the next sample.  This is
          important to ensure maximum column life and to minimize potential carryover and/or
          interferences from one sample to the next.

6.8.3  DETECTOR/CHROMATOGRAPHY MODULE - A DC Module (Dionex DC, P/N
      061793 or equivalent) equipped with dual injection valves and capable of maintaining the
      columns  and the suppressors at 30 °C,  and conductivity cell at 35 °C is recommended.

      NOTE: For optimal performance of this system, the conductivity cell should be set at a
      higher temperature than the analytical columns. For example, if the columns are set at 30
      °C, the cell should be set at 35 °C.

6.8.4  FIRST DIMENSION GUARD COLUMN - An 1C column, 4 x 50 mm (Dionex
      IonPac®AG20, P/N 063154 or equivalent).  Any guard column that provides adequate
      protection for the  analytical column and does not have an adverse effect on the peak
      shape may be used.

6.8.5  FIRST DIMENSION ANALYTICAL  COLUMN - An 1C column, 4 x 250 mm (Dionex
      IonPac®AS20, P/N 063148 or equivalent).  Any analytical column that provides adequate
      resolution, peak shape, capacity, accuracy, and precision (Sect. 9.2) may be used.  The
      separation mechanism for the first dimension analytical column must differ from the
      second dimension column in selectivity.

6.8.6  FIRST DIMENSION ANION SUPPRESSOR DEVICE - An 1C  eluent suppression
      device, 4 mm (Dionex Anion Self-Regenerating Suppressor, ASRS Ultra II, P/N 061561
                                     314.2-9

-------
       or equivalent).  An equivalent in-line suppression device that effectively suppresses the
       conductance of the eluent and converts the analyte to the hydronium form prior to
       conductivity detection, as well as provides adequate efficiency, resolution, peak shape,
       capacity, accuracy, precision, and a comparable MRL and DL (Sect. 9.2) may be used.
       Adequate baseline stability should be attained as measured by a baseline noise of no more
       than 5 nS per minute over the background conductivity. The first dimension suppressor
       must be compatible with (e.g., 4 mm in this example) the first dimension guard and
       analytical column.

       NOTE:  The conductivity suppressor was set to perform electrolytic suppression at a
       current setting of 150 mA. It was important to operate the suppressor in the external
       water mode to reduce baseline noise and achieve optimal method performance.

6.8.7   FIRST DIMENSION CARBONATE REMOVAL DEVICE - An 1C carbonate removal
       device, 4 mm (Dionex CRD, P/N 062983 or equivalent). Any in-line carbonate removal
       device that effectively removes the carbonate peak from the suppressed eluent stream
       prior to conductivity  detection and provides adequate efficiency,  resolution, peak shape,
       capacity, accuracy, and precision for perchlorate (Sect. 9.2) may be used.  The first
       dimension CRD must be compatible with (e.g., 4 mm in this example) the first dimension
       guard and analytical column.

6.8.8   FIRST DIMENSION CONDUCTIVITY DETECTOR - A Conductivity detector and
       integrated cell (Dionex CD P/N 061716, or equivalent) capable of providing data as
       required in Section 9.2. A Standard Bore Temperature Stabilizer (0.010 inch ID, Dionex
       P/N 062561 or equivalent) was also used to equilibrate the temperature of the eluent to
       that of the first dimension guard and analytical column.

       NOTE:  The conductivity detector cell temperature should be controlled at a temperature
       above the analytical column. For method development, the conductivity detector was set
       at 35 °C to minimize  bubble formation  and condensation between analytical column,
       suppressor and CRD  and stable temperature control of the detector cell itself.

6.8.9   CONCENTRATOR COLUMN - An 1C trapping column, 5 x 23 mm (Dionex UTAC-
       ULP1, P/N 063475 or equivalent). Any concentrator column that provides effective
       retention/trapping and release of perchlorate while providing the resolution, peak shape,
       capacity, accuracy, and precision (Sect. 9.2) may be used. The concentrator column
       should not release sulfonated leachates that would affect the quantitation of perchlorate.

   6.8.9.1 Alternate concentrator columns are allowed, but prior to their use, they must be
          evaluated to determine the first dimension cut window (Sect.  10.2.2) and they must be
          determined to have sufficient capacity to quantitatively trap perchlorate in the
          LFSSM CCC (Sect. 10.3.3) and should have relatively low backpressure since the
          concentrator column is placed as a post-suppressor device.

6.8.10  SECOND DIMENSION GUARD COLUMN - An 1C column, 2 x 50 mm (Dionex
       IonPac®AG16, P/N 055379 or equivalent).  Any guard column that provides adequate
       protection for the analytical column and does not have  an adverse effect on the peak
       shape may be used.
                                     314.2-10

-------
6.8.11  SECOND DIMENSION ANALYTICAL COLUMN - An 1C column, 2 x 250 mm
       (Dionex lonPac AS16, P/N 055378 or equivalent). Any analytical column that provides
       adequate resolution, peak shape, capacity, accuracy, and precision (Sect. 9.2) may be
       used. The separation mechanism for the second dimension analytical column must differ
       from the first dimension column.

6.8.12  SECOND DIMENSION ANION SUPPRESSOR DEVICE - An 1C eluent suppression
       device,  2 mm (Dionex Anion Self-Regenerating Suppressor, ASRS Ultra II, P/N 061562
       or equivalent). An equivalent in-line suppression device that effectively suppresses the
       conductance of the eluent prior to conductivity detection, and that provides adequate
       efficiency, resolution, peak shape, capacity, accuracy, precision, and a comparable MRL
       and DL (Sect. 9.2) may be used. Adequate baseline stability should be attained as
       measured by a baseline noise of no more than 5 nS per minute over the background
       conductivity. The second-dimension suppressor must be compatible with (e.g., 2 mm in
       this example) the  second-dimension guard and analytical column.

       NOTE: The conductivity suppressor was set to perform electrolytic suppression at a
       current  setting of 41 mA. It was important to operate the suppressor in the external water
       mode to reduce baseline noise and achieve optimal method performance.

6.8.13  SECOND DIMENSION CARBONATE REMOVAL DEVICE - An 1C carbonate
       removal device, 2 mm (Dionex CRD, P/N 062986 or equivalent).  Any in-line carbonate
       removal device that effectively removes carbonate from the suppressed eluent stream
       prior to conductivity detection of the method analyte and provides adequate efficiency,
       resolution, peak shape, capacity, accuracy, and precision (Sect. 9.2) may be used. The
       second  dimension CRD must be compatible with (e.g., 2 mm in this example) the second
       dimension guard and analytical column.

6.8.14  SECOND DIMENSION CONDUCTIVITY DETECTOR - A Conductivity detector and
       integrated cell (Dionex CD P/N 061716, or equivalent)  capable of providing data as
       required in Section 9.2. A Microbore Temperature Stabilizer (0.005 inch ID, Dionex P/N
       062562 or equivalent) was also used to equilibrate the temperature of the eluent to that of
       the second dimension guard and analytical column.

       NOTE: The conductivity detector cell temperature should be  controlled at a temperature
       above the analytical column. For method development, the conductivity detector was set
       at 35°C  to minimize bubble-formation and condensation between analytical column,
       suppressor and CRD and stabilize temperature control of the detector cell itself.

6.8.15  AUTOSAMPLER MODULE - An AS Autosampler Module with simultaneous injection,
       a sample prep option, and a large volume sample needle assembly (Dionex AS, P/N
       063105  or equivalent) was used to generate data for this method.  Any autosampler
       capable of automatically injecting up to 4.0 mL of sample may be used.

6.8.16  DATA  SYSTEM  - An interfaced data system such as Dionex, Chromeleon Version 6.7
       (or equivalent) is required to acquire, store, and output conductivity data.  The computer
       software should have the capability of processing stored conductivity data by recognizing
                                    314.2-11

-------
             and integrating a peak within a given retention time window. The software should be
             capable of constructing linear regressions or quadratic calibration curves, and calculating
             analyte concentrations using the calibrations.

   6.9   STERILE SAMPLE CONTAINERS - 125-mL sterile high-density polyethylene (HOPE)
         bottles (I-Chem 125-mL sterile HOPE bottle, Fisher Cat. No. N411-0125 or equivalent).

   6.10  SAMPLE FILTERS - Sterile sample filters (Corning 26-mm surfactant free cellulose acetate
         0.2-|im filter, Fisher Cat. No. 09-754-13 or equivalent).  If alternate filters are used they
         should be certified as having passed a bacterial challenge test.6 In addition, if alternate filters
         or different lots of the recommended filters are used, they must be tested using a LRB and a
         LFB fortified at the MRL as outlined in Section 9.2 to ensure that they do not introduce
         interferences or retain perchlorate.

   6.11  SYRINGES - 20-mL sterile, disposable syringes (Henke Sass Wolf 20-mL Luer lock, Fisher
         Cat. No.  14-817-33 or equivalent).

   6.12  VOLUMETRIC FLASKS - Class A, suggested sizes include 10, 50, 100, 250, 500 and 1000
         mL for preparation of standards and eluent.

   6.13  GRADUATED CYLINDERS -  Suggested sizes include 25 and 1000 mL.

   6.14  AUTO PIPETTES - Capable of delivering variable volumes from 1.0 |iL to 2500 |iL.

7.    REAGENTS AND STANDARDS

    7.1    REAGENTS - Reagent grade or better chemicals should be used in all tests. Unless
          otherwise indicated, it is intended that all reagents will conform to the specifications of the
          Committee on Analytical Reagents of the American Chemical Society (ACS), where such
          specifications are available.  Other grades may be used, provided it is first determined that
          the reagent is of sufficiently high purity to permit its use without lessening the quality of the
          determination.

      7.1.1  REAGENT WATER (RW)  - Purified water that does not contain any measurable
             quantity of the target analyte or interfering compounds at or above 1/3 the perchlorate
             MRL. The purity of the water required for this method cannot be over emphasized. The
             reagent water used during method development was purified from tap water using a
             Millipore ELIX-3 followed  by a Millipore Gradient A10 system. The water should
             contain no particles larger than 0.20 microns.

      7.1.2  ELUENT SOLUTIONS - Three hydroxide eluent concentrations were used to collect the
             data in Section 17. A potassium hydroxide eluent concentration of 35 mM was used for
             the first dimension matrix elimination separation on the AS20 column, and 65 mM for
             the second dimension separation on the AS 16 column.  In addition, a hydroxide eluent
             concentration of 100 mM was delivered to the first dimension column after the
             perchlorate heart-cut step was completed to ensure the column was properly cleaned prior
             to the next analysis. These eluents were automatically prepared using electrolytic eluent
             generation with the ICS-3000 EG Eluent Generator and EluGen potassium hydroxide
                                            314.2-12

-------
          cartridges (Sect 6.8.2).  They may also be manually prepared (Sect. 6.8.2) if adequate
          precautions are used to prevent carbonate formation from exposure to air.

   7.1.3   SODIUM BICARBONATE - (NaHCO3, CASRN 497-19-8) - Fluka Cat. No. 71627 or
          equivalent.

   7.1.4   SODIUM CHLORIDE - (NaCl, CASRN 7647-14-5) - Fisher Scientific Cat. No. S-271
          or equivalent.

   7.1.5   SODIUM SULFATE - (Na2SO4, CASRN 7757-82-6) - Fluka Cat. No. 71959 or
          equivalent.

7.2   STANDARD SOLUTIONS - When a compound purity is assayed to be 96 percent or greater,
      the weight can be used without correction to calculate the concentration of the stock standard.
      Solution concentrations listed in this section were used to develop this method and are
      included only as an example. Although estimated stability times for standard solutions
      are suggested in the following subsections, laboratories should use  standard QC
      practices to determine when their standards need to be replaced.

   7.2.1   PERCHLORATE STANDARD SOLUTIONS - Obtain the analyte as a solid standard or
          as a commercially prepared standard from a reputable standard manufacturer. Prepare
          the perchlorate stock and dilution solutions as described below.

       7.2.1.1 PERCHLORATE STOCK STANDARD SOLUTION (SSS) (1000 mg/L C1O4") -
             Preparation of this solution is accomplished using a solid NaClO4 standard. Weigh
             out 123.1 mg of NaClO4 into a 100-mL volumetric flask and dilute to volume with
             reagent water. When stored in opaque, plastic storage bottles,  the resulting stock
             solution may be stable for up to one year.

       7.2.1.2 PERCHLORATE PRIMARY DILUTION SOLUTION (PDS) (10.0 mg/L C1O4") -
             Prepare the Perchlorate PDS by adding 1.0 mL of the Perchlorate SSS to a 100-mL
             volumetric flask and dilute to volume with reagent water. This solution is used to
             prepare the Perchlorate Secondary Dilution Standard Solution, the Perchlorate
             Fortification Solutions and the Calibration Solutions below. When stored in opaque,
             plastic storage bottles, the resulting solution is stable for at least one month.

       7.2.1.3 PERCHLORATE SECONDARY DILUTION SOLUTION (SDS) (1.0 mg/L C1O4") -
             Prepare the 1.0 mg/L Perchlorate SDS by adding 10.0 mL of the Perchlorate PDS to a
             100-mL volumetric  flask and diluting to volume with reagent water. This solution is
             used to prepare the Perchlorate Fortification Solutions, CAL and CCC Standards
             listed below. When  stored in opaque, plastic storage bottles, the resulting solution is
             stable for at least one month.

       7.2.1.4 PERCHLORATE FORTIFICATION SOLUTIONS (PFS) (50, 200 and 500 |ig/L
             ClO/f) - The Perchlorate Fortification Solutions are prepared by dilution of the
             Perchlorate SDS and are used to fortify the Laboratory Fortified Blank (LFB), the
             Laboratory Fortified Sample Matrix (LFSM), the Laboratory Fortified Sample Matrix
             Duplicate (LFSMD) and the Laboratory Fortified Synthetic Sample Matrix (LFSSM)
                                       314.2-13

-------
   7.2.2
   with perchlorate. It is recommended that multiple concentrations be prepared so that
   the fortification levels can be rotated or adjusted to the concentration of the target
   analyte in the native samples.  When stored in opaque, plastic storage bottles, the
   resulting solutions are stable for at least one month.

LABORATORY SYNTHETIC SAMPLE MATRIX (LSSM) - Prepare a LSSM that
contains the common anions chloride, sulfate and bicarbonate at  1000 mg/L each as
follows.
      7.2.2.1 Weigh out 3.44 g of NaHCO3, 3.72 g of Na2SO4, and 4.00 g of NaCl (Fluka 71627,
             Fluka 71959, Fisher S-271, respectively, or equivalent). Add these to a 100-mL
             volumetric flask using a funnel and dilute to volume using reagent water. Dilute 4.0
             mL of this solution to 100 mL with RW to obtain the 1000 mg/L LSSM solution.

      CALIBRATION STANDARDS (CAL) - Prepare a calibration curve from dilutions of the
      Perchlorate PDS and the Perchlorate SDS using a minimum of five Calibration Standards,
      which span the concentration range of interest.  The lowest CAL standard must be at or below
      the MRL.  An example of the dilutions used to prepare the CAL standards used to collect the
      data in Section 17 are shown in the Table below.

      NOTE: CAL standards are not sterile filtered. This step must be omitted for the CALs in
      order to identify any potential losses associated with the sample filtration protocol.

CAL and CCC
Levels

CAL 1
CAL 2
CAL 3
CAL 4
CAL 5
CAL 6
Vol. of
C1O4
PDS
(HL)



30
50
100
Vol. of
C1O4
SDS
(HL)
30
50
100



Final
Vol. of
Std.
(mL)
100
100
100
100
100
100
Final
Cone, of
C1O4
(Hg/L)
0.30
0.50
1.00
3.00
5.00
10.0
7.4   CONTINUING CALIBRATION CHECK STANDARDS (CCC) - Prepare the CCC
      standards from dilutions of the Perchlorate PDS and the Perchlorate SDS at three
      concentrations representing the low, mid and high points of the calibration curve. The low-
      level CCC should be at or below the MRL. CCC Standards are not sterile filtered.

7.5   LABORATORY FORTIFIED SYNTHETIC SAMPLE MATRIX CCC STANDARD
      (LFSSM CCC) - In order to ensure that the first dimension cut window is functioning
      properly during each analysis batch, a CCC is also prepared in the 1000 mg/L common anion
      synthetic sample matrix. This solution is termed the Laboratory Fortified Synthetic Sample
      Matrix (LFSSM) CCC and is analyzed following the last high-level CCC during each
      Analysis Batch. The LFSSM CCC is not sterile filtered.
                                       314.2-14

-------
8.  SAMPLE COLLECTION, PRESERVATION, AND STORAGE

   8.1   SAMPLE COLLECTION

       8.1.1  Grab samples must be collected in accordance with conventional sampling practices.7

       8.1.2  When sampling from a cold water tap, open the tap and allow the system to flush until the
             water temperature has stabilized (usually approximately 3 to 5 minutes). Collect a
             representative sample from the flowing system using a beaker of appropriate size. Use
             this bulk sample to generate individual samples as needed.  A volume of approximately
             20-mL is required for each individual sample.

       8.1.3  When sampling from an open body of water, fill a beaker with water sampled from a
             representative area.  Use this bulk sample to generate individual samples as needed. A
             volume of approximately 20-mL of filtered sample is required for each individual
             sample.

       8.1.4  Once representative samples are obtained, they must be sterile filtered (Sect. 8.1.4.1) to
             remove any native microorganisms. Perchlorate is known to be susceptible to
             microbiological degradation.8  Wilkin et al.9 reported rapid natural microbial attenuation
             of perchlorate in surface water in the absence of nitrate following fireworks events in
             Oklahoma over a two-year period. They measured up to a 1000-fold increase of
             perchlorate concentration the day following the events, which was followed by a
             temperature-dependent reduction to native levels in 20 to 80 days. Samples are therefore
             sterile filtered to remove microbes and stored with headspace to reduce the potential for
             degradation by any remaining organisms.

          8.1.4.1 Remove a sterile syringe (Sect. 6.3) from its package and draw up approximately 25
                mL of the bulk sample. Remove a sterile syringe filter (Sect 6.2) from its package
                without touching the exit Luer connection.  Connect the filter to the syringe making
                sure that no water from the  syringe drops on the exterior of the filter. Depress the
                syringe plunger gently and discard the first 3-5 mL.  Open a sterile sample container
                (Sect. 6.9) without touching the interior.  Using gentle pressure, pass the sample
                through the filter into the sample container. During this process do not let the syringe
                or filter make contact with the sample container. Following filtration, seal the sample
                container tightly, label and prepare the container for shipment.  Syringes and filters
                are single use items and must be discarded after each sample.

   8.2   SAMPLE SHIPMENT AND STORAGE - Field samples must be chilled during shipment and
         must not exceed 10 °C during the first 48 hours after collection. Field samples should be
         confirmed to be at or below 10 °C when they are received at the laboratory.  Field samples
         stored in the lab must be held at or below 6 °C until analysis.

   8.3   SAMPLE HOLDING TIMES - Field samples that are  collected and stored as described in
         Sections  8.1 and 8.2 may be held for 28 days.
                                            314.2-15

-------
9.  QUALITY CONTROL

   9.1.   Quality Control requirements include the Initial Demonstration of Capability (IDC) and
         ongoing QC requirements that must be met when preparing and analyzing field samples.  This
         section describes each QC parameter, its required frequency, and the performance criteria that
         must be met in order to meet EPA data quality objectives. The QC criteria discussed in the
         following sections are summarized in Section 17, Tables 4 and 5. These QC requirements are
         considered the minimum acceptable QC criteria.  Laboratories are encouraged to institute
         additional QC practices to meet their specific needs.

   9.2   INITIAL DEMONSTRATION OF CAPABILITY (IDC) - The IDC must be successfully
         performed prior to analyzing any field samples. Prior to conducting the IDC, the analyst must
         first meet the calibration requirements outlined in Section 10. Requirements for the IDC are
         described in the following sections and are summarized in Table 4.

      9.2.1  DEMONSTRATION OF LOW SYSTEM BACKGROUND - Analyze a Laboratory
             Reagent Blank (LRB) processed through all sample collection steps outlined in Section
             8.1. The LRB must be sterile filtered using the same equipment used to collect field
             samples (Sect. 8.1).  Confirm that the LRB is reasonably free of contamination (<  1/3 the
             MRL) and that the criteria in Section 9.3.1 are met.

             NOTE: It is a good laboratory practice to include a blank in the calibration of any
             instrument. The method should also be checked for carryover by analyzing a RW blank
             immediately following the highest CAL standard.  If this RW sample does not meet the
             criteria outlined in Section 9.3.1 then carryover is present and should be identified and
             eliminated.

      9.2.2  DEMONSTRATION OF PRECISION - Prepare and analyze 7 replicate LFBs and
             LFSSMs fortified near the midrange of the initial calibration curve. All samples must be
             fortified and processed using the sample collection protocols described in Section 8.1.
             The percent relative  standard deviation (%RSD) of the results of the replicate analyses
             must be < 20 percent.

                  „,,,„„  Standard Deviation of Measured Concentrations  ,
                  %RSD =	xlOO
                                      Average Concentration

      9.2.3  DEMONSTRATION OF ACCURACY - Using the same set of replicate data generated
             for Section 9.2.2, calculate average recovery.  The average recovery of the replicate
             values must be within ± 20 percent of the true value.

                       n/_         Average Measured Concentration   inn
                       % Recovery =	x 100
                                        Fortified Concentration

      9.2.4  MINIMUM REPORTING LEVEL (MRL) CONFIRMATION - Establish a target
             concentration for the MRL based on the intended use of the method. The lowest
             calibration standard used to establish the initial calibration (as well as the low-level CCC)

                                           314.2-16

-------
       must be at or below the concentration of the MRL. Establishing the MRL concentration
       too low may cause repeated failure of ongoing QC requirements. Confirm or validate the
       MRL following the procedure outlined below.

   9.2.4.1 Fortify and analyze seven replicate LFBs at or below the proposed MRL
          concentration.  All samples must be fortified and processed using the sample
          collection protocols described in Section 8.1.  Calculate the mean (Mean) and
          standard deviation (S) for these replicates. Determine the Half Range for the
          Prediction Interval of Results (HRPIR) using the equation below.

                                      HRPIR =  3.963S

            where S is the standard deviation,  and  3.963 is a constant value for seven
            replicates.1

   9.2.4.2 Confirm that the upper and lower limits  for the Prediction Interval of Results (PIR =
          Mean +_ HRPIR) meet the upper and lower recovery limits as shown below.

              The Upper PIR Limit must be < 150 percent recovery.

                                   Mean + HRP1R
                               FortifiedConcentration

             The Lower PIR Limit must be > 50 percent recovery.

                                  Mean - HRPIP
                              	—	—	x  100> 50%
                              r ortijiedL oncentratron

   9.2.4.3 The MRL is validated if both the Upper and Lower PIR Limits meet the criteria
          described above (Sect. 9.2.4.2). If these criteria  are not met, the MRL has been set
          too low and must be determined again at a higher concentration.

9.2.5   QUALITY CONTROL SAMPLE (QCS) - Analyze a mid-level Quality Control Sample
       (Sect. 9.3.7) to confirm the accuracy of the calibration curve fit.

9.2.6   DETECTION LIMIT DETERMINATION (DL) (optional) - While DL determination is
       not a specific requirement of this method,  it may be required by various regulatory
       bodies associated with compliance monitoring.  It is the responsibility of the laboratory
       to determine  ifDL determination is required based upon the intended use of the data.

       Analyses for this  procedure should be done over at least 3 days. Prepare at least 7
       replicate fortified LFBs using the sample collection protocols described in Section 8.1.
       Use the solutions described in Section 7.2.1.4 to fortify at a concentration estimated to be
       near the DL.  This fortification concentration may be estimated by selecting a
       concentration at 2-5 times the noise level.  Analyze the seven replicates through all steps
       of Section  11.

                                     314.2-17

-------
          NOTE: If an MRL confirmation data set meets these requirements, a DL may be
          calculated from the MRL confirmation data, and no additional analyses are necessary.

          Calculate the DL using the following equation:

                       DL = St(n- 1, 1 - alpha = 0.99)

                 where:
                       t(n-1,1 - alpha = 0.99) = Student's t value for the 99% confidence level with n-1
                       degrees of freedom
                       n = number of replicates
                       S = standard deviation of replicate analyses.

          NOTE: Do not subtract blank values when performing DL calculations.

9.3     ONGOING QC REQUIREMENTS - This section describes the ongoing QC criteria that
       must be followed when processing and analyzing field samples.  Table 5 summarizes these
       requirements.

   9.3.1   LABORATORY REAGENT BLANK (LRB) - A LRB is analyzed during the IDC and is
          required with each Analysis Batch (Sect. 3.1) to confirm that potential  background
          contaminants are not interfering with the identification or quantitation  of perchlorate.
          The LRB must be sterile filtered using the same equipment used to collect field samples
          (Sect. 8.1). If within the retention time window of any analyte, the LRB produces a peak
          that would prevent the determination of perchl orate, determine the source of
          contamination and eliminate the interference before processing samples. Background
          from target analytes or contaminants that interfere with the measurement of target
          analytes must be <  1/3 the MRL. If the target analytes are detected in the LRB at
          concentrations equal to or greater than this level, then all data for the problem analyte(s)
          must be considered invalid for all samples which yielded a positive result.

       9.3.1.1 LRBs must be processed in the same manner as field samples including exposure to
              all sample collection devices (i.e., sterile filtration into sample containers). If samples
              are collected using devices that have not been previously evaluated by the laboratory,
              unused sample collection devices must be sent with the samples so that a LRB (and a
              LFB) may be processed in the laboratory.

          NOTE: Although quantitative data below the MRL may not be accurate enough for data
          reporting,  such data are useful in determining the magnitude of a background
          interference. Therefore, blank contamination levels may be estimated by extrapolation,
          when the concentration is below the MRL.

   9.3.2   CONTINUING CALIBRATION CHECK STANDARDS (CCC) - CCC standards are
          analyzed at the beginning of each Analysis Batch,  after every ten field  samples, and at the
          end of the  Analysis Batch.  The CCCs are not sterile filtered.  See Section 10.3 and Table
          5 for concentration requirements and acceptance criteria.
                                         314.2-18

-------
9.3.3   LABORATORY FORTIFIED SYNTHETIC SAMPLE MATRIX CCC STANDARD
       (LFSSM CCC) - A CCC standard prepared in the LSSM at the same concentration as the
       high-level CCC Standard should be analyzed at the end of each Analysis Batch. The
       LFSSM CCCs are not sterile filtered.  The LFSSM CCC is used to ensure the integrity of
       the sample pre-concentration/matrix elimination step and the chromatographic separation
       of perchlorate from other interfering anionic species in very high ionic matrices.  See
       Section 10.3.3. and Table 5 for concentration requirements and acceptance criteria.

9.3.4   LABORATORY FORTIFIED BLANK (LFB) -A LFB is required with each Analysis
       Batch. In successive analysis batches, the LFB fortification level must be rotated
       between low, medium, and high. The low concentration LFB must be at or below the
       MRL. Results of LFBs fortified at < MRL must be within 50-150% of the true value.
       Results of LFB analyses from all other concentrations must be 80-120% of the true value.
       If the LFB results do not meet these criteria, then all data for perchl orate must be
       considered invalid  for all samples in the Analysis Batch.

       NOTE: LFBs must be processed in the same manner as field samples including exposure
       to all sample collection devices (i.e., sterile filtration into sample containers). If samples
       are collected using devices that have not been previously evaluated by the laboratory,
       unused sample collection devices must be sent with the samples so that a LFB (and a
       LRB) may be processed in the laboratory.

9.3.5   LABORATORY FORTIFIED SAMPLE  MATRIX (LFSM) - Analysis of a LFSM (Sect.
       3.8) is required in each Analysis Batch. The LFSM is processed and analyzed exactly
       like a sample, including sterile filtration, and its purpose is to determine whether the
       sample matrix contributes bias to the analytical results. The background concentration of
       the analyte in the sample matrix must be determined in a separate aliquot and the
       measured value in the LFSM corrected for background concentrations. If a variety of
       different sample matrices are analyzed regularly, for example drinking water from
       groundwater and surface water sources, performance data should be collected for each
       source.

   9.3.5.1 Within each Analysis Batch, a minimum of one field sample is fortified as an LFSM
          for every 20 field samples analyzed.  The LFSM is prepared by spiking a field sample
          with an appropriate amount of the Perchl orate Fortification Solutions (Sect. 7.2.1.4).
          The fortification should be delivered in the smallest volume possible to minimize
          dilution of the sample.  Select a fortification concentration that is equal to or greater
          than the native  concentration, if known.  Use historical data and rotate through the
          designated concentrations when selecting a fortifying concentration.

   9.3.5.2  Calculate the  percent recovery (%REC) using the equation
            A = measured concentration in the fortified sample
            B = measured concentration in the unfortified sample
            C = fortification concentration
                                    314.2-19

-------
   9.3.5.3 Recoveries for samples fortified at concentrations near or at the MRL (within a factor
          of two times the MRL concentration) should be 50-150%. Recoveries for samples
          fortified at all other concentrations should be 80-120%. If the accuracy for any
          analyte falls  outside the designated range, and the laboratory performance for that
          analyte is shown to be in control in the CCCs, the recovery is judged to be matrix
          biased. The  result for that analyte in the unfortified sample is labeled suspect/matrix
          to inform the data user that the results are suspect due to matrix effects.

       9.3.5.3.1  Field samples that have a native perchlorate concentrations above the DL but
                below the MRL and are fortified at concentrations at or near the lowest
                calibration standard should be corrected for the native levels in order to obtain
                meaningful percent recovery values.  This example and the LRB (Sect. 9.3.1)
                are the only permitted use of analyte results below the MRL.

9.3.6   LABORATORY DUPLICATE OR LABORATORY FORTIFIED SAMPLE MATRIX
       DUPLICATE (LD or LFSMD) - Within each Analysis Batch, a minimum of one
       Laboratory Duplicate (LD) or Laboratory Fortified Sample Matrix Duplicate (LFSMD)
       must be analyzed.  Laboratory Duplicates check the precision associated with laboratory
       procedures.  If target analytes are not routinely observed in field samples, a LFSMD
       should be analyzed rather than a LD.

   9.3.6.1 Calculate the relative percent difference (RPD) for duplicate measurements  (LDi and
          LD2) using the equation
                                   LD  -LD
                           RPD=
:100
   9.3.6.2 RPDs for Laboratory Duplicates should be < 20%. Greater variability may be
          observed when Laboratory Duplicates have analyte concentrations that are within a
          factor of 2 of the MRL. At these concentrations Laboratory Duplicates should have
          RPDs that are < 50 percent.  If the RPD of any analyte falls outside the designated
          range, and the laboratory performance for that analyte is shown to be in control in the
          CCCs and LFSSM CCC, the recovery is judged to be matrix influenced. The result
          for that analyte in the unfortified field sample is labeled suspect/matrix to inform the
          data user that the results are  suspect due to matrix effects.

   9.3.6.3 If a LFSMD is analyzed instead of a Laboratory Duplicate, calculate the relative
          percent difference (RPD) for duplicate LFSMs (LFSM and LFSMD) using the
          equation.

                               LFSM-LFSMD
                              (LFSM + LFSMD)/2

   9.3.6.4 RPDs for duplicate LFSMs should be < 20%. Greater variability may be observed
          when LFSMs are fortified at analyte concentrations that are within a factor of 2 of the
                                     314.2-20

-------
             MRL. LFSMs fortified at these concentrations should have RPDs that are < 50%. If
             the RPD of any analyte falls outside the designated range, and the laboratory
             performance for that analyte is shown to be in control in the CCCs and LFSSM CCC,
             the precision is judged to be matrix influenced.  The result for that analyte in the
             unfortified field sample is labeled suspect/matrix to inform the data user that the
             results are suspect due to matrix effects.

   9.3.7  QUALITY CONTROL SAMPLES (QCS) - A QCS is required if an alternate
          commercial source is available  for the target analyte.  A QCS should be evaluated as part
          of the IDC (Sect. 9.2.5) and each time new PDS solutions are prepared.  If standards are
          prepared infrequently, analyze a QCS at least quarterly. The QCS should be fortified
          near the midpoint of the calibration range and analyzed as a CCC. The acceptance
          criteria for the QCS is the same as for the mid- and high-level CCCs (Section 10.3). If
          measured analyte concentrations are not of acceptable accuracy, check the entire
          analytical procedure to locate and correct the problem.

9.4    METHOD MODIFICATION QC REQUIREMENTS - The analyst is permitted to modify
       the separation technique, 1C columns, concentrator column and mobile phase composition.
       However, any modifications must maintain the basic chromatographic elements of this new
       technique (Sect 1.6).

   9.4.1  Each time method modifications are made, the analyst must first redetermine the cut
          window following Section 10.2.2 using perchlorate solutions fortified into the LSSM
          (Sect. 7.2.2) and then reestablish an acceptable initial calibration (Sect. 10.2.3). Alternate
          configurations may require modification of this technique.  If alternate procedures are
          used, the analyst must still employ the LSSM and RW matrices to set the start and stop
          times of the cut windows.

   9.4.2  Repeat the procedures of the IDC (Sect. 9.2) and verify that all Ongoing QC criteria can
          be met for the proposed method modification (Sect. 9.3).

   9.4.3  The analyst is also required to evaluate method performance for the proposed method
          modifications in real matrices that span the range of waters that the laboratory analyzes.
          This additional step is required  because modifications that perform acceptably during the
          IDC, which is conducted in reagent water and the LSSM, can fail ongoing method QC
          requirements in real matrices due to common method interferences. If, for example, the
          laboratory analyzes finished waters from both surface and groundwater municipalities,
          this requirement can be accomplished by assessing precision and accuracy (Sects. 9.2.2
          and 9.2.3) in a surface water with moderate to high total organic carbon ( e.g., 2 mg/L or
          greater) and a hard groundwater (e.g., 250 mg/L or greater).

   9.4.4  The results of Sections 9.4.1 - 9.4.3 must be appropriately documented by the analyst and
          should be independently assessed by the laboratory's QA officer prior to its  application to
          field samples.

       9.4.4.1 When implementing method modifications, it is the responsibility of the laboratory to
             closely review the results of Ongoing QC, and in particular, the results associated

                                         314.2-21

-------
                 with the LFSM (Sect. 9.3.5), the LFSMD  (Sect. 9.3.6) and the LFSSM CCCs (Sect.
                 9.3.3).  If repeated failures are noted, the modification must be abandoned.

10.  CALIBRATION AND STANDARDIZATION

    10.1   Demonstration and documentation of acceptable initial calibration for perchlorate is required
          prior to conducting the IDC and before any field samples are analyzed. If the initial
          calibration is successful, continuing calibration check standards are required at the beginning
          and end of each Analysis Batch, as well as after every tenth field sample.

          NOTE: CAL solutions and CCC standards are not processed with the sample collection
          protocols.  This step must be omitted for the CALs and CCCs in order to identify any
          potential losses associated with the sample filtration.

    10.2   INITIAL CALIBRATION - The initial calibration must be established prior to conducting
          the IDC (Sect. 9.2) and may be reestablished prior to analyzing field samples. However, it is
          permissible to verify the calibration with daily CCCs. Calibration should be performed using
          peak areas and the external standard technique. Calibration using peak heights is not
          permitted.

       10.2.1 INSTRUMENT CONDITIONS - Establish proper operating conditions. Operating
             conditions used during method development are described in Section 17 Table 1.
             Conditions different from those described may be used if the IDC QC criteria in Section
             9.2 are met. The procedure used to establish the first dimension cut window is provided
             in Section 10.2.2 below.

       10.2.2 FIRST DIMENSION CUT WINDOW DETERMINATION - Since EPA Method 314.2
             injects a large volume (2.0  - 4.0 mL) onto the first dimension column, the  affect of high
             ionic strength matrices on the perchlorate retention time requires careful consideration.
             Determination of the cut window is the first step in separating perchlorate from other
             interfering anionic matrix species.  Setting of the cut window in the first dimension must
             include evaluation of the retention time for perchlorate in both RW and the 1000 mg/L
             LSSM.

             NOTE: The procedure described below is applicable to the equipment configuration
             listed in Section 6. However, since acceptable modifications to this method require
             heart-cutting of perchlorate, the trapping of perchlorate on a concentrator column, and the
             subsequent separation and quantitation on a dissimilar analytical column, all modified
             versions of this procedure require the proper determination of the cut window.  Similar
             techniques must therefore be developed and used by analysts using modified methods.

          10.2.2.1   DETERMINING THE START TIME  OF THE CUT WINDOW - Inject an
                 aliquot (2.0 - 4.0 mL) of a 25 |ig/L perchlorate fortification in the 1000 mg/L LSSM
                 (without injection valve #2 on system #2 being activated) to determine the start time
                 for the cut window. The start time for the  cut window in the first dimension should
                 be  set at one minute prior to the start of the perchlorate elution in the 1000 mg/L
                 LSSM.

                                           314.2-22

-------
          NOTE: Depending upon the injection volume, the perchlorate in the LSSM may not
          appear as a distinct peak, but rather as a broad, smeared peak on the first dimension
          column. However, a distinct rise in baseline is evident when perchlorate starts to
          elute from the column (see Fig. 2 and 3).  In this matrix, there is also a pressure spike
          as well as a small baseline deflection which occurs when injection valve #2 in the
          second dimension is switched to the load position. This chromatographic event
          occurs approximately 30 seconds after injection valve #2 is switched. Consequently,
          it is recommended that a LSSM blank also be analyzed in order to ensure that the start
          of the elution of perchlorate is not misrepresented by the disruption of the baseline
          caused by the switching of injection valve #2.

   10.2.2.2   DETERMINING THE STOP TIME OF THE CUT WINDOW - The stop time
          for the cut window is established using reagent water. Inject an aliquot (2.0 - 4.0 mL)
          of the highest CAL standard in RW and determine when perchlorate is completely
          eluted off the first dimension column. The stop time for the cut window in the first
          dimension should be set at one minute after the perchlorate peak returns to baseline.

          NOTE: In RW, injection volume has less effect on the perchlorate retention time
          (lower ionic strength does not move the perchlorate forward).  However, in normal
          drinking water and high ionic strength matrices, the retention time for perchlorate can
          move forward significantly compared to elution in reagent water.  As noted above,
          there is a baseline deflection which occurs in the first dimension when injection valve
          #2 is switched back to the inject position. This event occurs approximately 30
          seconds after the valve is switched to the inject position.  Usually  setting the stop time
          for the cut window at one minute after the elution of perchlorate in RW in the first
          dimension is acceptable.  Since the stop time for the cut window is also the time at
          which injection valve #2 on system #2 is switched to the inject position, the elution of
          perchlorate must be completed at least 60 seconds before injection valve #2 is
          switched into the inject position when this instrument configuration  is used (see Figs.
          4 and 5).

10.2.3  CALIBRATION - Prepare a set of at least five CAL standards as described in Section
       7.3. The lowest concentration CAL standard must be at or below the  MRL. TheMRL
       must be confirmed using the procedure outlined in Section 9.2.4, after establishing the
       initial calibration.  Calibrate the conductivity detector using the external standard tech-
       nique. Calibration curves may be  generated using the 1C data system  through the use of
       first (linear) or second (quadratic)  order calibration curves.  Concentration-based
       weighting may be used. A quadratic fit is recommended for this method.

   10.2.3.1   CALIBRATION ACCEPTANCE CRITERIA - The validation of the calibration
          is determined by calculating the concentration of the analyte from each of the
          analyses used to generate the calibration curve. At least one CAL standard must be at
          or below the MRL. This calibration point should calculate to be 50 to 150 percent of
          its true value. All other higher concentration calibration points should calculate to be
          80 to 120 percent of their true values. If these criteria cannot be met, the analyst will
          have difficulty meeting ongoing QC criteria. In this case, corrective  action should be
          taken to reanalyze the calibration standards, restrict the range of calibration, or select
          an alternate method of calibration.
                                     314.2-23

-------
10.3   CONTINUING CALIBRATION CHECK (CCC) STANDARDS - The CCCs verify the
      calibration at the beginning and end of each group of analyses, as well as after every 10th
      field sample during analyses. The LRBs, LFBs, LFSSMs, LFSMs, LFSMDs, and CCCs are
      not counted as field samples. The beginning CCCs for each Analysis Batch must be at or
      below the MRL in order to verify instrument sensitivity and the accuracy of the calibration
      curve prior to the analysis of any field samples. Subsequent CCCs should alternate between
      a medium and high concentration.

   10.3.1  Inject an aliquot of the CCC standards and analyze with the same conditions used during
          the initial calibration.

   10.3.2  Calculate the concentration of the analyte in the CCC standards. The calculated amount
          for the analyte for mid and high level CCCs must be ± 20 percent of the true value. The
          calculated analyte amount for the lowest CCC level must be within ± 50 percent of the
          true value. If these conditions do not exist, then all data for the analyte must be
          considered invalid, and remedial action (Sect. 10.3.4) must be taken which may require
          recalibration. Any results from field samples that have been analyzed since the last
          acceptable calibration verification are invalid.

      10.3.2.1   The analyst should carefully review all first-dimension chromatograms  for the
             CCCs to ensure that the entire chromatographic peak elutes within the cut window. If
             this is not the case, the analyst should redetermine the cut window as per Section
             10.2.2.

   10.3.3  LABORATORY FORTIFIED SYNTHETIC SAMPLE MATRIX CCC STANDARDS
          (LFSSM CCC) - As chromatographic columns age, column performance can deteriorate.
          This deterioration will typically result in a decreased retention time for perchlorate,
          which can affect the perchlorate retention time in high inorganic matrices. A LFSSM
          CCC must be analyzed daily in each Analysis Batch at a concentration equal to the high-
          level CCC to ensure that the first dimension cut window is functioning properly for high
          inorganic strength waters. The QC acceptance criteria for the LFSSM CCC is the same as
          for the high-level CCCs (± 20%). If these conditions do not exist, then all data for the
          analyte must be considered invalid, and remedial action should be taken which  may
          require reestablishing the cut window (Sect. 10.2.2) and recalibration.

      10.3.3.1   The analyst should carefully review all first-dimension chromatograms  for the
             LFSSM CCC to ensure that the entire chromatographic peak elutes within the cut
             window.  If this is not the case, the analyst should redetermine the cut window as per
             Section 10.2.2.

   10.3.4  REMEDIAL ACTION - Failure to meet CCC or LFSSM CCC QC performance criteria
          requires remedial action. Maintenance such as confirming the integrity of the trapping
          efficiency of the concentrator column and matrix elimination step and/or regenerating or
          replacing the 1C guard and analytical columns require returning to the initial calibration
          step (Sect.  10.2).
                                        314.2-24

-------
11. PROCEDURE

   11.1   Important aspects of this analytical procedure include proper field sample collection (Sect.
          8.1) and storage (Sect. 8.2), ensuring that the instrument is properly calibrated (Sect. 10.2)
          and that all required QC are met (Sect. 9) during each Analysis Batch.  This section describes
          the procedures for field sample preparation and analysis. If alternative instrumentation
          and/or columns to those listed in this method are used, the procedure outlined in Section 9.2
          must be followed prior to analyzing field samples.

   11.2   SAMPLE PREPARATION

       11.2.1 Collect and store field samples as described in Sections 8.1 and 8.2.

       11.2.2 Process all LRBs, LFBs, LFSMs and LFSMDs using the sample collection protocols in
             Section 8.1.

   11.3   SAMPLE ANALYSIS

       11.3.1 Establish the instrument operating conditions as described in Table 1 of Section 17.
             Confirm that the analyte retention times for the calibration standards are stable.

       11.3.2 Establish a valid initial calibration following the procedures outlined in Section 10.2 or
             confirm that the calibration is still valid by running a low-level CCC as described in
             Section 10.3.  If establishing an initial calibration for the first time, complete the IDC as
             described in Section 9.2.

       11.3.3 Analyze field and QC samples at their required frequencies using the same conditions
             used to collect the initial calibration.  Table 6 shows an acceptable analytical sequence
             that contains all method-required QC samples.

       11.3.4 COMPOUND IDENTIFICATION - Establish an appropriate retention time window for
             perchlorate elution from the second dimension column in order to identify it in QC and
             field sample chromatograms.

          11.3.4.1   Since the ionic strength of drinking water matrices may vary considerably, the
                    cut window (Sect. 10.2.2) for perchlorate in the first dimension must be set
                    wide enough to account for the variability in the ionic strength of the
                    drinking water matrices. A window of approximately 5-6 minutes (depending
                    upon the injection volume) was found to be acceptable for the system used during
                    method development.  If the cut window is not set properly, analyte recoveries
                    may be reduced without affecting the second-dimension retention times.

          11.3.4.2   Retention times in the second dimension should be very stable as long as the first-
                    dimension heart cut window has been set properly (Sect. 10.2.2).  This is because
                    a majority of the common anions, which alter the ionic strength of field samples
                    causing sample-to-sample variation in 1C retention time, are eliminated during the

                                            314.2-25

-------
                    heart cut procedure. Retention times measured for perchlorate in RW and for the
                    LFSSM were essentially identical during method development.

          11.3.4.3   The QC requirements for each analysis batch include a first-dimension printout of
                    the final CCC and the high level LFSSM CCC chromatograms to ensure the heart
                    cut window is functioning properly. The analyst should also review all first-
                    dimension chromatograms to ensure acceptable chromatographic performance
                    within the first-dimension cut window. For example, if a sample had a much
                    higher conductivity (significantly above the LFSSM) caused by either very high
                    levels of interfering anions or perchlorate, the first-dimension cut window would
                    be overwhelmed.  In such an instance, the sample should be diluted and
                    reanalyzed.

          11.3.4.4   High ionic strength matrices have the potential to cause an increase in background
                    conductivity and severe tailing as the other anions elute from the first dimension
                    column and cause the perchlorate retention time to decrease.

       11.3.5 EXCEEDING CALIBRATION RANGE - The analyst must not extrapolate beyond the
             established calibration range.  If an analyte result exceeds the range of the initial
             calibration curve, the field sample may be diluted with reagent water and the diluted field
             sample re-injected. Incorporate the dilution factor into final concentration calculations.
             The dilution will also affect the perchlorate MRL.

12. DATA ANALYSIS AND CALCULATIONS

   12.1  Identify the analyte present in the field and QC Samples as described in Section  11.3.4.

   12.2  Calculate the perchlorate concentrations using the multi-point calibration established in
         Section 10.2. Quantify only those values that fall between the MRL and the highest
         calibration standard. Field samples with target analyte responses that exceed the highest
         calibration standard require dilution and reanalysis (Sect. 11.3.5).

       12.2.1 As noted in Section 9.3.5.3.1, it may be necessary to extrapolate below the MRL to
             estimate contaminants in LRBs and to correct for native levels of perchlorate below the
             MRL when field samples are fortified at or near the MRL.  These are the only permitted
             use of analyte results below the MRL.

   12.3  Calculations must utilize all available digits of precision, but final reported concentrations
         should be rounded to an appropriate number of significant figures  (one digit of uncertainty),
         typically two, and not more than three significant figures.

   12.4  Prior to reporting data, the laboratory is responsible for assuring that QC requirements have
         been met or that any appropriate qualifier is documented.

13. METHOD PERFORMANCE

   13.1  PRECISION, ACCURACY AND DETECTION LIMITS - Tables for these data are
         presented in Section 17.  The LCMRL for perchlorate is presented in Table 2 and was
                                            314.2-26

-------
         calculated using a procedure described elsewhere.1  The DL data is also reported in Table 2.
         Single laboratory precision and accuracy data are presented in Table 3.

   13.2  Figure 6 is a representation of both the first and second dimension chromatograms for a 5.0
         ug/L perchlorate fortification in reagent water and Figure 7 shows similar chromatograms for
         a 0.50 ug/L perchlorate fortification to a municipal  surface water, disinfected with chlorine.

14. POLLUTION PREVENTION

   14.1  For information about pollution prevention that may be applicable to laboratory operations,
         consult "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, D.C., 20036, or on-line at:
         http: //www. up s. edu/community/storeroom/Chemi cal_Wastes/wastearti cl es. htm.

15. WASTE MANAGEMENT

   15.1  The analytical procedures described in this method  generate relatively small amounts of waste
         since  only small amounts of reagents are used. The matrices of concern are finished drinking
         water or source water. However, the Agency requires that laboratory waste management
         practices be conducted consistent with all applicable rules and regulations, and that
         laboratories protect the  air, water, and land by minimizing and controlling all releases from
         fume  hoods and bench operations.  Also, compliance is required with any sewage discharge
         permits and regulations, particularly the hazardous waste identification  rules and land disposal
         restrictions.  For further information on waste management, see the publications of the
         American Chemical Society's Laboratory Environment, Health  & Safety Task Force on the
         Internet at http://membership.acs.Org/c/ccs/publications.htm.  Or see "Laboratory Waste
         Minimization and Pollution Prevention," Copyright © 1996 Battelle Seattle Research Center,
         which can be found at http://www.p2pays.org/ref/01/text/00779/index2.htm.

16. REFERENCES

   1. Winslow, S. D.; Pepich, B. V.; Martin, J. 1; Hallberg,  G. R.; Munch, D. 1; Frebis, C. P.;
      Hedrick, E. J.; Krop, R.  A. Statistical Procedures for Determination and Verification of
      Minimum Reporting Levels for Drinking water Methods. Environ. Sci. &  Technol. 2006, 40,
      281-288.

   2. Glaser, J.A.;  Foerst, D.L.; McKee, G. D.; Quave, S. A.; Budde, W. L. Trace Analyses for
      Wastewaters, Environ. Sci. & Technol. 1981, 15, 1426-1435.

   3. "Carcinogens - Working With Carcinogens," Department of Health, Education, and Welfare,
      Public Health Service, Center for Disease Control, National Institute for Occupational  Safety and
      Health, Publication No.  77-206, Aug. 1977.

   4. "OSHA  Safety and Health Standards, General Industry," (29CFR1910), Occupational Safety and
      Health Administration, OSHA 2206, (Revised, January 1976).

   5. "Guidelines,  Recommendations, and Regulations for Handling Antineoplastic Agents", Center
                                            314.2-27

-------
   for Disease Control, National Institute for Occupational Safety and Health,
   http://www.cdc.gov/niosh/topics/antineoplastic/pubs.html#b.

6.  Blosse, P.T., Boulter, E.M., Sundaram, S., "Diminutive Bacteria Implications for Sterile
   Filtration", Pall Corporation, East Hills, NY.

7.  ASTM Annual Book of Standards, Part II, Volume 11.01, D3370-82, "Standard Practice for
   Sampling Water," American Society for Testing and Materials, Philadelphia, PA, 1986.

8.  Xu, J.; Song, Y.;  Min, B.; Steinberg, L.;  Logan, B.E. . Microbial degradation of perchlorate:
   principles and applications. Environ. Engin. Sci.  2003, 20(5), 405-422.

9.  Wilkin, R. T.; Fine, D. D.; Brunett, N. G. Perchlorate Behavior in a Municipal Lake Following
   Fireworks Displays. Environ. Sci.  & Technol.  2007, 41, 3966-3971.
                                         314.2-28

-------
 17.0 TABLES, DIAGRAMS, FLOWCHARTS, AND VALIDATION DATA

 TABLE 1. INSTRUMENTAL CONDITIONS
Standard Conditions and Equipment for First Dimension Analyses
Ion Chromatograph:
Sample loop:
Load Volume:
Eluent Generator:
Dionex ICS-3000 Dual System
2.0 or 4.0 mL
4.7 mL
Dionex ICS-3000 EG Eluent Generator Module (P/N 061714), with
dual channel EluGen Catridges (P/N 058900), isocratic 35 mM
potassium hydroxide step changed or ramped to 100 mM. potassium
hydroxide following the elution of perchlorate (when injection valve #2
is switched to the inject position)
1.0 mL/min
AS Autosampler Module (P/N 063105) with Sequential Injection,
Sample Prep options and a large volume (8.2 mL) sample needle
assembly (P/N 061267)
Dionex lonPac AG 20 Guard 4 x 50 mm (P/N 063154)
Dionex lonPac AS 20 Analytical 4 x 250 mm (P/N 063148)
-2500 psi
Dionex ASRS Ultra II, 4-mm (P/N 061561) with  4-mm Carbonate
Removal Device (CRD, P/N 062983)
DC-Module (P/N 061793) maintained @ 30° C
Conductivity Detector with integrated cell (P/N 061716) held @ 35° C
45 min
Eluent Flow:
Autosampler:


Columns :

Typical System Back-pressure:
Conductivity Suppressor:

Chromatography Module
Detector:
Total analysis time:

Standard Conditions and Equipment for Second Dimension Analyses ^^:
Ion Chromatograph:
Sample loop:
Load Volume:
Eluent Generator:
Eluent Flow:
Columns :

Typical System Back-pressure:
Conductivity Suppressor:

Chromatography Module
Detector:
Total analysis time:
Dionex ICS-3000 Dual System
Dionex UTAC ULP1, 5 x 23-mm, concentrator column (P/N 063475)
Cut-window time set per procedures in Section 11.4 (5-6 mL)
Dionex ICS-3000 EG Eluent Generator Module (P/N 061714), with
dual channel EluGen Catridges (P/N 058900), isocratic 65 mM
potassium hydroxide
0.25 mL/min
Dionex lonPac AG 16 Guard 2 x 50 mm (P/N 055379)
Dionex lonPac AS 16 Analytical 2 x 250 mm (P/N 055378)
-2500 psi
Dionex ASRS Ultra II, 2-mm (P/N 061562) with 2-mm Carbonate
Removal Device (CRD, P/N 062986)
DC-Module (P/N 061793) maintained @ 30° C
Conductivity Detector with integrated cell (P/N 061716) held @ 35° C
45 min
   (a) Mention of trade names or commercial products does not constitute endorsement or
      recommendation for use. (b) It is the responsibility of the analyst to assure that the instruments,
      supplies, and conditions used are capable of meeting all method QC requirements.
                                          314.2-29

-------
 TABLE 2. LOWEST CONCENTRATION MRL AND DLs FOR PERCHLORATE
           WITH 2-D 1C
Injection
Volume
2.0 mL
4.0 mL
Analyte
cicv
LCMRLa (ug/L)
0.060
0.038
*DL (ug/L)
0.018b
0.012C
       aLCMRLs were calculated according to the procedure in reference 1.
       *The DL was calculated from data acquired on a single day.
       bReplicate fortifications at 0.075 |ig/L.
       °Replicate fortifications at 0.025 |ig/L.
TABLE 3. 1C PRECISION AND RECOVERY DATA FOR PERCHLORATE IN VARIOUS
          MATRICES WITH 2-D 1C (n=7) USING 2.0- AND 4.0-mL INJECTION VOLUMES
2.0-mL Injection Volume 4.0-mL Injection Volume
Matrix
Reagent Water

*LFSSM

Ground water

Surface water 1

Surface water 2

Spiked
Cone.
(HS/L)
0.50
5.0
0.50
5.0
0.50
5.0
0.50
5.0
0.50
5.0
Unspiked
Cone.
(HS/L)
<0.060
<0.060
<0.060
<0.060
0.20
0.20
0.50
0.50
0.74
0.74
Mean
% Rec.
94.7
96.1
94.6
95.0
92.4
97.0
95.3
97.1
91.8
96.2
% RSD
4.5
1.8
6.9
1.5
5.8
1.8
4.2
1.8
4.3
1.2
Unspiked
Cone.
(HS/L)
<0.038
<0.038
<0.038
<0.038
0.20
0.20
0.49
0.49
0.74
0.74
Mean
% Rec.
97.2
98.6
97.2
97.0
95.5
97.2
110
97.4
92.8
97.5
% RSD
4.2
0.58
4.6
1.3
2.5
2.1
3.4
1.1
1.2
0.85
 *Described in Section 3.10 and 7.2.2
                                          314.2-30

-------
TABLE 4.  INITIAL DEMONSTRATION OF CAPABILITY QUALITY CONTROL
           REQUIREMENTS
   Method
  Reference
   Requirement
    Specification and
       Frequency
    Acceptance Criteria
  Section
  9.2.1
Demonstration of
Low System
Background
Analyze a LRB prior to
any other IDC steps.
Demonstrate that perchlorate is
below 1/3 of the MRL (Section
9.3.1) and that possible
interferences from sampling
protocols do not prevent the
identification and
quantification of perchlorate.
  Section
  9.2.2
Demonstration of
Precision
Analyze 7 replicate LFBs
and LFSSMs fortified near
the mid-point of the
calibration curve
%RSDmustbe  < 20%.
  Section
  9.2.3
Demonstration of
Accuracy
Calculate average recovery
for replicates used in
Section 9.2.2.
Mean recovery + 20% of true
value.
  Section
  9.2.5
Quality Control
Sample
During IDC, each time a
new analyte PDS is made,
every time the instrument
is calibrated and at least
quarterly.	
The result for perchlorate must
be 80-120% of the true value.
                                           314.2-31

-------
TABLE 5.   ONGOING QUALITY CONTROL REQUIREMENTS (SUMMARY)
   Method
  Reference
  Requirement
  Specification and Frequency
     Acceptance Criteria
  Section 8.3
Sample Holding
Time
28 days when processed and
stored according to sections 8.1
and 8.2 with appropriate
preservation and storage.	
Sample results are valid only if
samples are extracted within
sample hold time.
  Section
  9.3.1
Laboratory
Reagent Blank
(LRB)
Analyze a LRB as part of the IDC
(Sect. 9.2), as part of each
calibration, and with each
Analysis Batch
Demonstrate that perchlorate is
below 1/3 of the MRL and that
possible interferences from
sampling protocols do not
prevent the identification and
quantification of perchlorate.
  Section
  9.3.2
  and 10.3
Continuing
Calibration Check
(CCC) Standards
Verify initial calibration by
analyzing a low-level CCC at the
beginning of each Analysis
Batch. Subsequent CCCs are
required after every 10 field
samples, and after the last field
sample in a batch.
Low CCC - at or below the MRL
concentration
Mid CCC - near midpoint in
calibration curve
High CCC - near the highest
calibration standard.
For each CCC the result must
be:
CCC Level        Result
< MRL     + 50% True value

 > MRL    + 20% True value
to high CAL
                                                                Recalibration is recommended
                                                                if these criteria are not met.
  Section
  9.3.3
Laboratory
Fortified
Synthetic Sample
Matrix CCCs
(LFSSM CCC)
In order to monitor the cut
window during an Analysis
Batch, a high-level CCC
standard, prepared in the LFSSM
(Sect. 9.3.3) is also required at
the end of each Analysis Batch.
For the LFSSM CCC the result
must be

 + 20% True value
  Section
  9.3.4
Laboratory
Fortified Blank
(LFB)
Analyze a LFB with each
analysis batch, rotating between
low, medium and high
concentration from batch to
batch.
                                Low LFB - at or below the MRL
                                concentration
                                Mid LFB - near midpoint in
                                calibration curve
                                High LFB - near the highest
                                calibration standard.	
                                           314.2-32
For each LFB the result must
be:
LFB Level         Result
< MRL     + 50% True value

 > MRL    + 20% True value
to high CAL

The result for perchlorate must
be 80-120% of the true value.

Recalibration is recommended
if these criteria are not met.

-------
TABLE 5. ONGOING QUALITY CONTROL REQUIREMENTS (Continued)
   Method
  Reference
   Requirement
     Specification and
        Frequency
     Acceptance Criteria
  Section
  9.3.5
Laboratory
Fortified Sample
Matrix (LFSM)
Analyze one LFSM per
Analysis Batch (20 field
samples or less).  Fortify the
LFSM with perchlorate at a
concentration close to but
greater than the native
concentration (if known).
Calculate LFSM recoveries.
Recoveries for the LFSM must
be calculated (Sect. 9.3.5.2).
The result must be:
                                                             LFSM Level
                                                                  Result
                                                              MRL    + 20% True value
                                                             to high CAL
  Section
  9.3.6
Laboratory
Duplicate (LD) or
Laboratory
Fortified Sample
Matrix Duplicate
(LFSMD)
Analyze at least one LD or
LFSMD daily, or with each
Analysis Batch (20 samples
or less), whichever is more
frequent.
RPD must be calculated (Sect.
9.3.6.1 for LD and Sect. 9.3.6.3
for LFSMD). The result must
be:
 Level              Result
<2xMRL     <  50% RPD
                                                             2xMRL
                                                             to high CAL
                                                                < 20% RPD
  Section
  9.3.7
Quality Control
Sample (QCS)
During IDC, each time a new
analyte PDS is made, every
time the instrument is
calibrated and at least
quarterly.
Results must be + 20% of the
expected value.
  Section
  10.2
Initial Calibration
Use external standard
calibration technique to
generate a first or second
order calibration curve.  Use
at least 5 standard
concentrations.
Check the calibration curve
as described in Section 10.2.
                                 Analyze a QCS near the mid-
                                 point of the calibration
                                 curve.
When each calibration standard
is calculated as an unknown
using the calibration curve, the
result should be:
  Level           Result
< MRL      + 50% True value

> MRL      + 20% True value
to high CAL

The result for perchlorate must
be 80-120% of the true value.

Recalibration is recommended
if these criteria are not met.
                                           314.2-33

-------
TABLE 6.  SAMPLE ANALYSIS BATCH WITH QC REQUIREMENTS
Injection
#
1
2
O
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Sample
Description
Laboratory Reagent Blank (LRB)
CCC at the MRL
LFB
Sample 1
Sample 2
Sample 2 - Laboratory Fortified Sample Matrix (LFSM)
Sample 2 - Laboratory Fortified Sample Matrix Duplicate
(LFSMD)
Sample 3
Sample 4
Sample 5
Sample 6
Sample 7
Sample 8
Sample 9
Sample 10
CCC at mid level
Sample 1 1
Sample 12
Sample 13
Acceptance
Criteria
< 1/3 MRL
Recovery of 50- 150%
< MRL 50-150% of value
> MRL 80-120% of value
normal analysis
normal analysis
Recovery of 80 - 120%
%RPD = + 20%
normal analysis
normal analysis
normal analysis
normal analysis
normal analysis
normal analysis
normal analysis
normal analysis
Recovery of 80 - 120%
normal analysis
normal analysis
normal analysis
                         CONTINUED on NEXT PAGE
                                  314.2-34

-------
TABLE 6. (Continued)
Injection
#
20
21
22
23
24
25
26
27
28
Sample
Description
Sample 14
Sample 15
Sample 16
Sample 17
Sample 18
Sample 19
Sample 20
CCC at high level*
LFSSM CCC at high level*
Acceptance
Criteria
normal analysis
normal analysis
normal analysis
normal analysis
normal analysis
normal analysis
normal analysis
Recovery of 80- 120%
Recovery of 80- 120%
* also require first dimension chromatogram printout
                                          314.2-35

-------
             Figure 1. EPA Method 314.2 System Schematic
 1st Dimension
                                   2nd  Dimension
                   Autasampler
waste	"


  Large Loop

   	Load
   — Inject
                     Injection Valve 1
4mmAG/AS20
  Column 1
 External Water
 Suppressor
                 CRD1
                                          waste —
                                  External Water-
                              Injection Valve 2
                                               i	
 CD 2
i

 CRD2


 Suppressor 2
                                   2mmAG/AS16Colurhh2
                                               EG
                         Concentrator
                           Column
                          	Transfer to 2-D
                          — Load Concentrator
                                                   ---- 1
                                            waste
                            314.2-36

-------
Figure 2. Setting the start time for the cut window in the first dimension using a 2.0-mL injection
volume of a 50 ug/L C1O4 fortification in the 1000 mg/L LSSM.
    0.639;
    0.500-
    0.375
    0.250-
          MS
                                             Baseline deflection
    0.125-  Cut window start set
              18.5 min
   -o.ooo
Start of baseline increase due to
elution of perchlorate @ 19.5 min
                                          314.2-37

-------
Figure 3 . Setting the start time for the cut window in the first dimension using a 4.0-mL injection

volume of a 25 ug/L C1O4 fortification in the 1000 mg/L LSSM.
                       Pressure Spike
 1.22=   0
       MS
 1.00-
 0.80-
 0.60-
 0.40-
 0.20-
-0.00
_Q 29 "I1 I ' !  > I ' ' ' I > > ' I ' ' > I ! ' ' I  > > > I ' ' > I > > ' I ' >


          13.0                  18.0
                                             Baseline deflection
               Cut window start set

               @> 20.0 min
                                                Start of baseline increase due to

                                                elution of perchlorate @ 21.0 min
                               mm
23.0
| I I I | 1 1 I


    26.0
30.0
                                         314.2-38

-------
Figure 4. Setting the stop time for the cut window in the first dimension using a 2.0-mL injection
volume of a 25 ug/L C1O4 fortification in RW.
 0.671'
       MS
 0.500'
 0.375'
 0.250"
 0.125"
-0.000
-0.125
-0.267 T I
                                      CIO/
Baseline deflection
                                         \
                                         Return to baseline (5) 22.5 min
Cut window end set
                                                                            23.5 min
           13.0  14.0  15.0 16.0  17.0  18.0  19.0 20.0  21.0 22.0  23.0 24.0 25.0 26.0  27.0 28.0 29.0
                                             314.2-39

-------
Figure 5. Setting the stop time for the cut window in the first dimension using a 4.0-mL injection
volume of a 10 ug/L CICV fortification in RW.
  0.400-
         uS
  0.300-
  0.200-
  0.100-
  o.ooo-
  -0.10CT
  -0.2001 ' '
       13.0
                                                cip4-
                           Baseline deflection
                                    \
i i i i i  i i i i i i i i  i

15.0     17.0
I I I I ,  I I I I I , ,
19.0     21.0
                                              Return to baseline
                                              @25.0 min
                                                          Cut window end
                                                          Set @ 26.0 min
25.0
27.0
29.0 30.0
                                          314.2-40

-------
Figure 6. First- and Second-Dimension Chromatograms of 5.0 ug/L perchlorate fortification in
Reagent Water.
  1.20
 1.00
 0.80
 0.60
      MS
 0.40-
 0.20
-0.00
                                                                                 40.0
45.0
                                             314.2-41

-------
Figure 7. Typical First- and Second-Dimension Chromatograms of a municipal surface water
disinfected with chlorine.  The native level of perchlorate was 0.055 ug/L.
         MS
   1.20
   1.00-
   0.80-
   0.60-
   0.40-
   0.20-
  -0.00
  -0.20i i " |
      0.0
10.0
20.0
30.0
                                                                                      mm
40.0      45.0
                                             314.2-42

-------