EPA Document # EPA 815-R-03-007
METHOD 326.0 DETERMINATION OF INORGANIC OXYHALIDE DISINFECTION
              BY-PRODUCTS IN DRINKING WATER USING ION
              CHROMATOGRAPHY INCORPORATING THE ADDITION OF A
              SUPPRESSOR ACIDIFIED POSTCOLUMN REAGENT FOR TRACE
              BROMATE ANALYSIS
                                Revision 1.0

                                 June 2002
Herbert P. Wagner and Barry V. Pepich, Shaw Environmental Inc
Daniel P. Hautman and David J. Munch, US EPA, Office of Ground Water and Drinking
Water
E. Salhi and Urs von Gunten, Swiss Federal Institute for Environmental Science and
Technology, EAWAG, CH-8600, Dubendorf, Switzerland
                      TECHNICAL SUPPORT CENTER
            OFFICE OF GROUND WATER AND DRINKING WATER
               U. S. ENVIRONMENTAL PROTECTION AGENCY
                         CINCINNATI, OHIO 45268
                                  326.0-1

-------
                                  METHOD 326.0
 DETERMINATION OF INORGANIC OXYHALIDE DISINFECTION BY-PRODUCTS
 IN DRINKING WATER USING  ION CHROMATOGRAPHY INCORPORATING THE
ADDITION OF A SUPPRESSOR ACIDIFIED POSTCOLUMN REAGENT FOR TRACE
                              BROMATE ANALYSIS

1.     SCOPE AND APPLICATION

      1.1    This method covers the determination of inorganic oxyhalide disinfection by-
             product anions in reagent water, surface water, ground water, and finished
             drinking water. In addition, bromide can be accurately determined in source or
             raw water and it has been included due to its critical role as a disinfection by-
             product precursor. Bromide concentration in finished water can differ due to
             numerous variables which can influence the concentration. Since this method,
             prior to the addition of the postcolumn reagent (PCR), employs the same hardware
             as EPA Method 300.1(1), the analysis of the common anions (using EPA Method
             300.1, Part A1) can be performed using this instrument setup with the postcolumn
             hardware attached but "off-line" and with the appropriate smaller sample loop.
Inorganic Disinfection By-products by Conductivity Detection
Analyte
Bromate
Bromide
Chorite
Chloride
Comments
report
report
report
range
report
range
values
values
values
values
> 15.0 ug/L
from source
> Minimum
> Minimum
*
and raw waters only
Reporting Level in calibration
Reporting Level in calibration
Inorganic Disinfection By-products by Absorbance Detection
Analyte
Bromate
Comments
report
values
> Minimum Reporting Level to 15.0 ug/L *
             * the concentrations reported for bromate assume both detectors to be running
             simultaneously.
                                       326.0-2

-------
      1.2    The single laboratory reagent water Detection Limits (Sect. 3.15) for the above
             analytes are listed in Table 1.  The Detection Limit 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 Detection Limit differs from, and is
             lower than the Minimum Reporting Level (MRL) (Sect. 3.16).  The Detection
             Limit for a specific matrix may differ from  those listed, depending upon the
             nature of the sample and the specific instrumentation employed.

             1.2.1   In order to achieve comparable detection limits on the conductivity
                    detector, an ion chromatographic system must utilize suppressed
                    conductivity detection, be properly maintained and must be capable of
                    yielding a baseline with no more than 5 nanosiemen (nS) noise/drift per
                    minute of monitored response over the background conductivity.

             1.2.2   In order to achieve acceptable detection  limits on the postcolumn
                    absorbance detector, the postcolumn reagent must be delivered
                    pneumatically and some form of software signal filtering or smoothing of
                    the absorbance signal from the absorbance detector must be incorporated.

      1.3    This method is recommended for use only by or under the supervision of analysts
             experienced in the use of ion chromatography and in the interpretation of the
             resulting ion chromatograms.

      1.4    When this method is used to analyze unfamiliar samples for any of the above
             anions, anion identification  should be supported by the use of a fortified sample
             matrix covering the anions of interest. The fortification procedure is described in
             Section 9.8.

      1.5    Users of the method must demonstrate the ability to generate acceptable results
             with this method, using the procedures described in  Section 9.

2.    SUMMARY OF METHOD

      2.1    The development of this method was based upon the work of several investigators
             as summarised elsewhere.(3) A volume of sample, approximately 225 |_iL (see
             Note), is introduced into an  ion chromatograph (1C)  which includes a guard
             column, analytical column, suppressor devices,  conductivity detector, a
             postcolumn reagent delivery system (pneumatically controlled), a heated
             postcolumn reaction coil, and a ultraviolet/visible (UV/Vis) absorbance detector
             (see Figure 1). After separation and  suppression of the eluent, the oxyhalide
             anions chlorite, chlorate, bromate >15.0 |-ig/L and bromide are measured using
             conductivity detection. To facilitate  low-level detection of bromate, the
             suppressed effluent from the conductivity detector is combined with an acidic
                                        326.0-3

-------
             solution of potassium iodide containing a catalytic amount of molybdenum VI.
             The mixture is heated at 80 C (to facilitate complete reaction) where the bromate
             reacts with iodide to form the tri-iodide ion which is measured by its UV
             absorption at 352 nm.

             NOTE:  A 225 uL sample loop can be made using approximately 111 cm (44
             inches) of 0.02 inch i.d. PEEK tubing. The volume should be verified to be
             within 5% by weighing the sample loop empty, filling the loop with deionized
             water and re-weighing the loop assuming the density of water is  1 mg/uL.

3.      DEFINITIONS

       3.1    ANALYSIS BATCH - A sequence of samples, which are analyzed within a 30
             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. The required QC samples include:
                          Laboratory Reagent Blank (LRB)
                          Continuing Calibration Check Standards (CCCs)
                          Laboratory Fortified Blank (LFB)
                          Laboratory Fortified Sample Matrix (LFSM), and
                          Either a Field Duplicate (FD) or a Laboratory Fortified  Sample
                          Martix Duplicate (LFSMD).

       3.2    SURROGATE ANALYTE (SUR) -  A pure analyte, which chemically resembles
             target analytes and is extremely unlikely to be found in any sample. This analyte
             is added to a sample aliquot in known amount(s) before filtration or other
             processing and is measured with the same procedures used to measure other
             sample components. The purpose of the SUR is to monitor method performance
             with each sample.

       3.3    LABORATORY REAGENT BLANK (LRB) - An aliquot of reagent water or
             other blank matrix that is treated exactly as a sample including exposure to all
             glassware, equipment, solvents and reagents, sample preservatives,  and
             surrogates that are used in the analysis batch. The LRB is used to determine if
             method analytes or other interferences are present in the laboratory environment,
             the reagents, or the apparatus.

       3.4    LABORATORY FORTIFIED BLANK (LFB) - An aliquot of reagent water or
             other blank matrix to which known quantities of the method analytes and all the
             preservation compounds are added in the  laboratory. The LFB  is analyzed exactly
             like a sample, and its purpose is to determine whether the methodology is in
             control, and whether the laboratory is capable of making accurate and precise
             measurements.
                                       326.0-4

-------
3.5   LABORATORY FORTIFIED SAMPLE MATRIX (LFSM) - An aliquot of an
      environmental sample to which known quantities of the method analytes and all
      the preservation compounds are added in the laboratory. The LFSM is analyzed
      exactly like a sample, and its purpose is to determine whether the sample matrix
      contributes bias to the analytical results. The background concentrations of the
      analytes in the sample matrix must be determined in a separate aliquot and the
      measured values in the LFSM corrected for background concentrations.

3.6   LABORATORY FORTIFIED SAMPLE MATRIX DUPLICATE (LFSMD) - A
      second aliquot of the field sample used to prepare the LFSM fortified, processed
      and analyzed identically. The LFSMD is used instead of the Field Duplicate to
      access method precision when the occurrence of target analytes is low.

3.7   LABORATORY DUPLICATES (LD1 and LD2) - Two aliquots of the same
      sample taken in the laboratory and analyzed separately with identical procedures.
      Analyses of LD1 and LD2 indicate precision associated with laboratory
      procedures, but not with sample collection, preservation, or storage procedures.

3.8   FIELD DUPLICATES (FD1 and FD2) - Two separate samples collected at the
      same time and place under identical circumstances, and treated exactly the same
      throughout field and laboratory procedures. Analyses of FD1 and FD2 give a
      measure of the precision associated with sample collection, preservation, and
      storage, as well as laboratory procedures.

3.9   STOCK STANDARD SOLUTION (SSS) - A concentrated solution containing
      one or more method analytes prepared in the laboratory using assayed reference
      materials or purchased from a reputable commercial source.

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

3.11  CALIBRATION STANDARD (CAL) - A solution prepared from the primary
      dilution standard solution and/or stock standard solution, and the surrogate
      analytes. The CAL solutions are used to calibrate the instrument response with
      respect to analyte concentration.

3.12  INSTRUMENT PERFORMANCE CHECK SOLUTION (PC) - A solution of
      one or more method analytes, surrogates, or other test substances used to evaluate
      the performance of the instrument system with respect to a defined set of criteria.
                                 326.0-5

-------
       3.13   CONTINUING CALIBRATION CHECK (CCC) - A calibration standard
             containing the method analytes and surrogates (s), which is analyzed periodically
             to verify the accuracy of the existing calibration for those analytes.

       3.14   QUALITY CONTROL SAMPLE (QCS) - A solution of method analytes and
             surrogate(s) of known concentrations that is obtained from a source external to the
             laboratory and different from the source of calibration standards. It is used to
             check standard integrity.

       3.15   DETECTION LIMIT - 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 of precision
             (Sect. 9.2.4), and accurate quantitation is not expected at this level.(2)

       3.16   MINIMUM REPORTING LEVEL (MRL) - The minimum concentration that can
             be reported as a quantitated value for a target analyte in a sample following
             analysis.  This defined concentration can be no lower than the concentration of the
             lowest continuing calibration standard for that analyte and can only be used if
             acceptable quality control criteria for this standard are met.

       3.17   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.

4.     INTERFERENCES

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

             4.1.1  A direct chromatographic coelution may be solved by changing columns,
                    eluent strength, modifying the eluent with organic solvents (if compatible
                    with 1C columns), changing the detection systems, or selective removal of
                    the interference with pretreatment.  Sample dilution will have little to no
                    effect. The analyst must verify that these changes do not induce any
                    negative affects on method performance by repeating and passing all the
                    QC criteria as described  in Section 9.
                                         326.0-6

-------
       4.1.2   Sample dilution may resolve some of the difficulties if the interference is
              the result of either concentration dependant coelution or ionic character
              displacement, but it must be clarified that sample dilution will alter your
              Minimum Reporting Limit (MRL) by a proportion equivalent to that of the
              dilution. Therefore, careful consideration of project objectives should be
              given prior to performing such a dilution. An alternative to sample
              dilution, may be dilution of the eluent as outlined in Section 11.2.6.

       4.1.3   Pretreatment cartridges can be effective as a means to eliminate certain
              matrix interferences.  With any proposed pretreatment, the analyst must
              verify that target analyte(s) are not affected by monitoring recovery after
              pretreatment.  With advances in analytical separator column technology
              which employ higher capacity anion exchange resins, the need for these
              cartridges has been greatly reduced.

4.2    Method interferences may be caused by contaminants in the reagent water,
       reagents, glassware, and other sample processing apparatus that lead to discrete
       artifacts or elevated baselines in an ion chromatogram.  These interferences can
       lead to false positive results for target analytes as well as reduced detection limits
       as a consequence of elevated baseline noise.

4.3    Samples that contain particles larger than 0.45 microns  and reagent solutions
       require filtration to prevent damage to instrument columns and flow systems.

4.4    Close attention should be given to the potential for carry over peaks from one
       analysis which will affect the proper detection of analytes of interest in a second
       or subsequent analysis. Normally, in this analysis, the elution of sulfate (retention
       time of 17.5 min.) indicates the end of a chromatographic run, but in the ozonated
       and chlorine dioxide matrices,  a small response (200 nS baseline rise) was
       observed for a very late eluting unknown peak following the response for sulfate.
       Consequently, a run time of 25 minutes is recommended to allow for the proper
       elution of any potentially interfering late peaks.  It is the responsibility of the user
       to confirm that no late eluting peaks have carried over into a subsequent analysis
       thereby compromising the integrity of the analytical results.

4.5    Any residual chlorine dioxide present in the sample will result in the formation of
       additional chlorite prior to analysis.  If residual chlorine dioxide is suspected in
       the  sample, the sample must be sparged with an inert gas (helium, argon or
       nitrogen) for approximately five minutes. This sparging must be conducted prior
       to ethylenediamine preservation and at the time of sample collection.

4.6    The presence of chlorite can interfere with the quantitation of low concentrations
       of bromate on the postcolumn UV/Vis absorbance detector. In order to accurately
                                   326.0-7

-------
             quantify bromate concentrations in the range 0.5 - 15.0 |-ig/L in this postcolumn
             system, the excess chlorite must be removed prior to analysis as outlined in
             Section 11.1.4.

5.      SAFETY

       5.1    The toxicity or carcinogenicity of each reagent used in this method has not been
             precisely defined; each chemical compound should be treated as a potential health
             hazard, and exposure to these chemicals should be minimized.  The laboratory is
             responsible for maintaining a current awareness file of OSHA regulations
             regarding the safe handling of the chemicals specified in this method. A reference
             file of MSDSs should also be made available to all personnel involved in the
             chemical analysis.  Additional references to laboratory safety are available/4"7'

       5.2    Pure standard materials and stock standards of these compounds should be
             handled with suitable protection to skin and eyes. Care should  be taken not to
             breathe the vapors or ingest the materials.

       5.3    The following chemical has the potential to be highly toxic or hazardous. The
             Material Safety Data Sheet (MSDS) should be consulted.

             5.3.1   Sulfuric acid - used to prepare regenerant solution (Sect. 7.1.8) for the
                    second suppressor (Dionex AMMS or Ultra ASRS-1 used in the chemical
                    mode) and  to for pretreatment of the samples for chlorite removal (Sect.
                    7.1.7,11.1.4)

6.      EQUIPMENT AND SUPPLIES

       6.1    ION CHROMATOGRAPH - Analytical  system complete with ion
             chromatographic pump and all required accessories including syringes, analytical
             columns, compressed  gasses, suppressors, conductivity detector,  mixing "tee",
             postcolumn reagent delivery system, reaction coil, reaction coil heater, UV/Vis
             absorbance detector (Figure 1) and a computer-based data acquisition and control
             system.

             NOTE: Because the KI PCR solution is susceptible to oxidation, resulting in a
             yellow colored solution, the PCR MUST be flushed from the suppressor, reaction
             coil and detector cell with reagent water upon completion of the final analysis and
             prevented from draining through the reaction coil by gravity once the system is
             shut down. This can be accomplished either manually or by incorporating a
             column switching valve in combination with a reagent water flush.
                                         326.0-8

-------
6.1.1   ANION GUARD COLUMN - Dionex AG9-HC  4 mm (Cat.#: 51791 or
       equivalent). This column functions as a protector of the separator column.
       If omitted from the system the retention times will be shorter.

6.1.2   ANION SEPARATOR COLUMN - Dionex AS9-HC column, 4 mm
       (Cat.#:  51786 or equivalent, see Note). The AS9-HC, 4-mm column using
       the conditions outlined in Table 1 produced the separations shown in
       Figures 2 and 3.

       NOTE: The use of 2-mm columns is not recommended. A 50-uL sample
       loop would be required with the 2-mm columns.  This reduced injection
       volume would decrease the "on-column" bromate and negatively affect
       PCR reactivity and the subsequent absorbance response. As well, the 2-
       mm columns require a flow rate approximately 4 times less than the 4-mm
       columns. At the lower flow rates, band broading may become an issue and
       it would be difficult to accurately maintain the appropriate reduced flow
       rate for the PCR.

6.1.3   ANION SUPPRESSOR DEVICES - The data presented in this method
       were generated using a Dionex Ultra-1 Anion Self Regenerating
       Suppressor (4 mm ASRS, Cat.#: 53946) for electrolytic suppression of the
       eluent and  a second Ultra -1 ASRS was used in the chemical mode  to
       acidify  the PCR just prior to addition to the mixing tee. Equivalent
       suppressor devices maybe utilized providing comparable conductivity
       detection limits are achieved and adequate baseline stability is attained as
       measured by a combined baseline drift/noise of no more than 10 nS per
       minute  over the background conductivity.  Alternative suppressor
       evaluations subsequent to this method development work have indicated
       that improved detection limits and precision and accuracy can be obtained
       for bromate by conductivity detection(8).  If conductivity analytes below
       10 ug/L are to be reported, the combined baseline drift/noise will be
       required to be no greater than 5 nS per minute over the background
       conductivity. The suppressor must be able to withstand approximately 80
       -120 psi back pressure which results from connecting the postcolumn
       hardware to the "eluent out" port of the suppressor. The suppressor used
       to acidify the PCR must be capable of continuous operation using 150 mN
       sulfuric acid as the regenerant.

       6.1.3.1  - The conductivity suppressor was set to perform electrolytic
       suppression at a current setting of 100 mA using the external water mode.
       Insufficient baseline  stability was observed on the conductivity detector
       using an ASRS in recycle mode.
                           326.0-9

-------
      6.1.4  CONDUCTIVITY DETECTOR - Conductivity cell (Dionex CD20 or
             equivalent) capable of providing data as required in Section 9.2.

      6.1.5  ABSORBANCE DETECTOR - Absorbance detector (Dionex AD20 or
             equivalent with 10-mm cell pathlength, equipped with a deuterium source
             bulb, or equivalent and capable of measuring absorbance at 352 nm)
             capable of providing data as required in Section 9.2.

      6.1.6  POSTCOLUMN REAGENT DELIVERY SYSTEM -Delivery system
             (Dionex PC-10 or equivalent) capable of pneumatically delivering the
             postcolumn reagent to the "eluent in" port of the suppressor to acidify the
             PCR prior to entering the mixing tee (see Note). The pressure settings will
             need to be established on an individual basis for each specific instrument
             configuration and at a level which yields the prescribed PCR flow rates.

             NOTE: Since KI is photosensitive, the KI/Mo PCR solution was observed
             to develop a light yellow color with time, even when stored under helium
             in the opaque plastic PC-10 delivery container inside the PC-10
             pressurization vessel. Purging the KI/Mo solution with helium
             immediately after preparation to remove all oxygen did not completely
             eliminate the problem. Consequently, in order to facilitate overnight (24
             hours) operation, the external wall of the PC-10 plastic pressurization
             vessel was wrapped with an opaque tape, or other light impervious
             material, to prevent any light exposure to the KI/Mo PCR (care must be
             exercised to leave about  1/16th of an inch at both the top and bottom of the
             vessel free of tape to allow for proper sealing of the top and bottom.  The
             generation of the tri-iodide ion is pH dependant and the second suppressor
             is used to acidify the PCR just before entering the reaction coil. Prior to
             initiating any analysis batch, to ensure that the pH of the reaction mixture
             is below 2, the effluent from the absorbance detector should be monitored
             using pH test strips.

      6.1.7  REACTION COIL -  500-uL internal volume, knitted, potted and
             configured to fit securely in the postcolumn reaction coil heater.  (Dionex
             Cat.#: 39349 or equivalent).

      6.1.8  POSTCOLUMN REACTION COIL HEATER - Capable of maintaining a
             temperature of 80 C (Dionex PCH-2 or equivalent).

6.2   DATA SYSTEM - The Dionex Peaknet Data Chromatography Software was used
      to generate all the data in the attached tables. Other computer-based data systems
      may achieve approximately the same Detection Limits, but the user must
      demonstrate this by the procedure outlined in Section 9.2.
                                 326.0-10

-------
       6.3    ANALYTICAL BALANCE - Used to accurately weigh target analyte salts for
             stock standard preparation (0.1 mg sensitivity).

       6.4    TOP LOADING BALANCE - Used to accurately weigh reagents to prepare
             eluents (10 mg sensitivity).

       6.5    WEIGH BOATS - Plastic, disposable, used to weigh eluent reagents.

       6.6    SYRINGES - Plastic, disposable, 10 mL, used during sample preparation.

       6.7    PIPETS - Pasteur, plastic or glass, disposable, graduated, 5 mL and 10 mL.

       6.8    BOTTLES - Opaque, high density polyethylene (HDPE) or amber glass, 30 mL,
             125 mL, 250 mL - used for sample collection and storage of calibration solutions.
             Opaque bottles are required due to the photoreactivity of the chlorite anion.

       6.9    MICRO BEAKERS - Plastic, disposable, used during sample preparation.

       6.10   PARTICULATE FILTERS - Gelman ion chromatography Acrodisc 0.45 urn
             (Cat.#: 4485 or equivalent) syringe filters.  These cartridges are used to remove
             particulates and  [Fe(OH)3(s)] which are formed during the oxidation-reduction
             reaction between Fe (IT) and C1O2" (see Sect.  11.1.4).

       6.11   HYDROGEN CARTRIDGES - Dionex OnGuard-H cartridges (Cat.#: 039596 or
             equivalent).  These cartridges are conditioned according to the manufacturer's
             directions and are used to protect the analytical column and the suppressor
             membrane by removing excess ferrous iron [Fe (n)]. The ferrous iron is added to
             field samples to reduce chlorite levels prior to analysis of chlorine dioxide
             disinfected water samples for trace levels of bromate (see Sect. 11.1.4).

7.      REAGENTS AND STANDARDS

       7.1    REAGENTS AND SOLVENTS - Reagent grade or better chemicals should be
             used. Unless otherwise indicated, it is intended that all reagents shall conform to
             the specifications of the Committee on Analytical Reagents of the American
             Chemical Society, 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 - Distilled or deionized water 18 M Q or better, free
                    of the anions of interest.  Water should contain particles no larger than
                    0.20 microns.
                                        326.0-11

-------
7.1.2  ELUENT SOLUTION - Sodium carbonate (CAS#: 497-19-8) 9.0 mM.
      Dissolve 1.91 g sodium carbonate (NajCOg) in reagent water and dilute to
      2L.

      7.12.1 This eluent solution must be sparged for 10 minutes with helium
             prior to use to remove dissolved gases which may form micro
             bubbles in the 1C compromising system performance and adversely
             effecting the integrity of the data. Alternatively, an in-line degas
             apparatus may be employed.

7.1.3  ETHYLENEDIAMINE (EDA) PRESERVATION SOLUTION (100
      mg/mL) - Dilute 2.8 mL of ethylenediamine (99%) (CAS#: 107-15-3) to
      25 mL with reagent water.  Prepare fresh monthly.

7.1.4  AMMONIUM MOLYBDATE SOLUTION - A 2.0-mM solution of
      ammonium molybdate tetrahydrate [(NH 4)6Mo7O244H2O, CAS#:
      12027667, Fluka Cat.#: 09878 or equivalent] is prepared by dissolving
      0.247 g in 100 mL of reagent water. This reagent is stored in an opaque
      plastic storage bottle and prepared fresh monthly.

7.1.5  POSTCOLUMN REAGENT (0.26 M KI, 43 uM ammonium molybdate
      heptahydrate) -The postcolumn reagent is prepared by adding 43.1 g of
      potassium iodide (KI, CAS#: 7681110, Fluka Cat.#: 60400  or equivalent)
      to a 1-L volumetric flask containing about 500 mL of reagent water. Two
      hundred and fifteen |jL of the ammonium molybdate solution (Sect. 7.1.4)
      is added to the volumetric flask and diluted to volume with reagent water.
      The PCR is sparged with helium  for 20 minutes to remove all traces of
      dissolved oxygen and immediately placed in the PC-10 delivery vessel and
      pressurized with helium. The reagent is stable for 24 hours if properly
      protected from light (Sect. 6.1.6).

7.1.6  FERROUS IRON SOLUTION [1000 mg/L Fe(II)] - Dissolve 0.124 g
      ferrous sulfate heptahydrate (FeSO4.7H2O, CAS#: 7782630, Sigma Cat. #:
      F-7002 or equivalent) in approximately  15 mL reagent water containing
      6 uL of concentrated nitric acid and dilute to 25 mL with reagent water in
      a volumetric flask (final pH ~2).  The Fe (II) solution must be prepared
      fresh every two days.

7.1.7  SULFURIC ACID (0.5 N) - Dilute 1.4 mL of concentrated  sulfuric acid
      (Fisher Scientific Certified ACS Plus, A 300-500) to 100 mL.

7.1.8  SULFURIC ACID (0.15 N) - Dilute 8.5 mL of concentrated sulfuric acid
      (Fisher Scientific Certified ACS Plus, A 300-500) to 2000 mL.
                          326.0-12

-------
7.2    STANDARD SOLUTIONS - Standard Solutions may be prepared from certified,
       commercially available solutions or from solid compounds.  Compounds used to
       prepare solutions must be 96% pure or greater and the weight may be used
       without correction for purity to calculate the concentration of the stock standard.
       Solution concentrations listed in this section were used to develop this method
       and are included as an example. Even though stability times for standard
       solutions are suggested in the following sections, laboratories should use
       standard QC practices to determine when Standard Solutions described in
       this section need to be replaced.

       7.2.1   ANALYTE STANDARD  SOLUTIONS (1000 mg/L) - Stock standard
             solutions may be purchased as certified solutions or prepared from ACS
             reagent grade, potassium or sodium salts as listed below, for most
             analytes. Certified chlorite standard solutions are commercially available.
             If these are not used, chlorite requires careful consideration as outlined
             below in Section 7.2.1.1.4.

             7.2.1.1 ANALYTE STOCK STANDARD SOLUTIONS - Individual
                   Analyte Stock Standards Solutions are prepared as described
                   below.

                   7.2.1.1.1   Bromide (Br) 1000 mg/L - Dissolve 0.1288 g sodium
                              bromide (NaBr, CAS#: 7647-15-6) in reagent water and
                              dilute to 100 mL in a volumetric flask.

                   7.2.1.1.2   Bromate (BrO3') 1000 mg/L -  Dissolve 0.1180 g of
                              sodium bromate (NaBrO3, CAS#: 7789-38-0) in reagent
                              water and dilute to 100 mL in  a volumetric flask.

                   7.2.1.1.3   Chlorate (CIO,') 1000 mg/L - Dissolve 0.1275 g of
                              sodium chlorate (NaClO3, CAS#: 7775-09-9) in
                              reagent water and dilute to 100 mL in a volumetric
                              flask.

                   7.2.1.1.4   Chlorite (C1O2') 1000 mg/L -  Prepare from
                              commercially available standards  or as described below.
                              If the amperometric titration of the technical grade
                              sodium chlorite (NaClO2), as  specified in the Note
                              below, had indicated the purity of the salt to be 80.0 %
                              NaClO2, the analyst would dissolve 0.1676 g of sodium
                              chlorite (NaClO2, CAS#: 7758-19-2) in reagent water
                              and dilute to 100 mL in a volumetric flask.
                                 326.0-13

-------
                 Note: High purity sodium chlorite (NaCIO 2) is not
                 currently commercially available due to its potential
                 explosive instability.  Recrystallization of the technical
                 grade (approx. 80%) can be performed but it is labor
                 intensive and time consuming. The simplest approach
                 is to determine the exact purity of the NaCIO 2 using the
                 iodometric titration procedure.(9)  Following titration,
                 an individual component standard of chlorite must be
                 analyzed to determine if there is any significant
                 contamination (greater than 1% of the chlorite weight)
                 from chlorate, bromate or bromide (as other method
                 target anions) in the technical grade chlorite standard.

7.2.1.2 ANALYTE PRIMARY DILUTION STANDARD (Analyte PDS)
       SOLUTION - Prepare two Analyte PDSs by diluting the Analyte
       Stock Standard Solutions with reagent water containing EDA (at a
       final concentration of 50 mg/L) in volumetric glassware. The
       dilutions used to prepare these solutions during the method
       development studies are provided below as an example. Prior to
       using mixed standards for calibration or spiking solutions, ensure
       that the individual Analyte Stock Standard Solutions do not contain
       any appreciable concentrations of the other target analytes.
       Dilutions of these Analyte PDSs, referred to as Solution A and B
       below, are used to prepare the calibration  solutions (Sect. 7.2.3)
       and the continuing calibration check solutions (Sect. 10.3) for both
       detectors.

                           Analyte PDS Solution A
Analyte
Chlorite
Bromide
Chlorate
Initial Cone.
(mg/L)
1000
10000
1000
Volume
(mL)
2.5
0.25
2.5
Final Volume
(mL)
25
25
25
Final Cone.
(mg/L)
100
100
100
                            Analyte PDS Solution B
Analyte
Bromate
Initial Cone.
(mg/L)
1000
Volume
(mL)
1.0
Final Volume
(mL)
100
Final Cone.
(mg/L)
10
                    326.0-14

-------
7.2.2   SURROGATE ANALYTE (SUR) SOLUTION, DICHLOROACETATE
       (DCA, CAS#: 19559-59-2)

       7.2.2.1 SURROGATE STOCK SOLUTION (0.50 mg/mL) - Prepare a
             surrogate stock solution by dissolving 0.065 g of dichloroacetic
             acid, potassium salt (C12CHCO2K) in reagent water and diluting to
             100 mL in a volumetric flask.  This solution is used to fortify all
             field samples, QC samples and calibration standards by adding a
             20-uL aliquot of the Surrogate Stock Solution to 10 mL of the
             sample. This solution must be prepared fresh every 3 months or
             sooner if signs of degradation are present.

             7.2.2.1.1  Dichloroacetate is potentially present in treated drinking
                       waters as the acetate of the organic disinfection
                       byproduct, dichloroacetic acid (DCAA). Typical
                       concentrations of DCAA rarely exceed 50 |-ig/L,  which
                       would represent only a five percent increase in the
                       observed response over the fortified concentration of
                       1.00 mg/L. Consequently, the upper recovery limit for
                       the surrogate (90% to 115%) has been increased to
                       allow for this potential background.

             7.2.2.1.2  If the analyst is exclusively interested in monitoring
                       trace bromate using the PCR and the UV/VIS
                       absorbance detector, suppression of the eluent prior to
                       reaction with the PCR MUST be incorporated. In
                       addition, the surrogate must also be included and meet
                       the QC requirements as outlined in Section 9.7.1.

       7.2.2.2 SURROGATE PRIMARY DILUTION STANDARD - A
             Surrogate PDS is not  prepared since the Surrogate Stock (Sect
             7.2.2.1) is used to fortify samples.

7.2.3   CALIBRATION STANDARDS (CAL) - At least 5 calibration
       concentrations are required to prepare the initial calibration curve (Sect.
       10.2) for each detector. Prepare the calibration standards over the
       concentration range of interest from dilutions of the Analyte PDSs A and
       B in reagent water containing EDA (50 mg/L).  The lowest concentration
       calibration standard must be  at or below the MRL, which may depend
       upon system sensitivity. The calibration standards for the development of
       this method were prepared by adding aliquots of the two Analyte PDSs
       (Analyte PDS Solution A and B described above in Sect. 7.2.1.2) as shown
       in the tables below to a volumetric flask, containing the listed volume of
                          326.0-15

-------
EDA solution, and diluting to volume with reagent water. These standards
may be also be used as CCCs.

Conductivity Detector CAL and CCC Standards
Cal
Std.
1*
2
3*
4
5*
Stock A
(HL)
10
25
75
200
500
Stock B
(HL)
100
250
500
750
1000
EDA
Volume
(uL)
50
50
50
50
50
Final
Volume
(mL)
100
100
100
100
100
C1O2, Br, C1O3
Final Cone.
(l-ig/L)
10
25
75
200
500
Bromate
Final Cone.
(l-ig/L)
10
25
50
75
100
*Prepared in larger volume and used as CCCs
       Absorbance Detector CAL and CCC Standards
Cal
Std.
1*
2
3
4*
5
6*
Stock B
(jiL)
5
10
20
50
100
150
EDA
Volume
(uL)
50
50
50
50
50
50
Final
Volume
(mL)
100
100
100
100
100
100
Bromate
Final Cone.
(l-ig/L)
0.5
1.0
2.0
5.0
10.0
15.0
*Prepared in larger volume and used as CCCs

7.2.3.1 Fortify each CAL or CCC standard by adding a 20 uL aliquot of
       the Surrogate Stock Standard Solution (Sect. 7.2.2.1) to a 20 mL
       disposable plastic micro beaker containing 10.0 mL of the
       calibration standard (or CCC) and mix.  These volumes may be
       adjusted to meet specific laboratory autosampler volume
       requirements.
                   326.0-16

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

       8.1    SAMPLE COLLECTION

             8.1.1   Samples should be collected in opaque plastic or amber glass bottles.  All
                    bottles must be thoroughly cleaned and the volume collected should be
                    sufficient to ensure a representative sample, allow for replicate analysis
                    and laboratory fortified matrix analysis, if required, while minimizing
                    waste disposal.

             8.1.2   When collecting a field sample from a treatment plant employing chlorine
                    dioxide, the field sample must be sparged with an inert gas (helium or
                    nitrogen) prior to addition of the EDA preservative at time of sample
                    collection.  The sample should be collected in a clean wide mouth flask
                    (such as an Erlenmeyer flask). The sparging gas can be obtained by using
                    a lecture bottle of nitrogen or helium fitted with a regulator and connected
                    to a disposable glass Pasteur pipette with PVC tubing.  The gas flow
                    should be adjusted to produce a steady flow of bubbles. After 10-15
                    minutes of sparging, all traces of chlorine dioxide should be removed from
                    the sample. It can then be poured from the flask into the sample bottle that
                    contains the ethylenediamine (EDA) preservative. In order to eliminate
                    potential cross contamination problems, it is recommended that a clean
                    Erlenmeyer flask and a new disposable pipette be used at each sampling
                    point.

             8.1.3   Add a sufficient volume of the EDA preservation solution (Sect. 7.1.3)
                    such that the final concentration is 50 mg/L in the sample.  This would be
                    equivalent to adding 0.5 mL of the EDA preservation solution to 1 L of
                    sample.

       8.2    SPECIAL SAMPLING REQUIREMENTS AND PRECAUTIONS FOR
             CHLORITE

             8.2.1   Sample bottles used for chlorite analysis must be opaque plastic or amber
                    glass to protect the sample from light.

             8.2.2   When preparing the  LFSM, be aware that chlorite is an oxidant and may
                    react with the natural organic matter in an untreated drinking water matrix
                    as a result of oxidative demand. If untreated water is collected for chlorite
                    analysis, and subsequently used for the LFSM, EDA preservation will not
                    control this demand  and reduced chlorite recoveries may be observed.
                                        326.0-17

-------
       8.3
             8.2.3  Chlorite is susceptible to degradation both through catalytic reactions with
                    dissolved iron salts and reactivity towards free chlorine which exists as
                    hypochlorous acid/hypochlorite ion in most drinking water as a residual
                    disinfectant/10' EDA serves a dual purpose as a preservative for chlorite by
                    chelating iron as well as any other catalytically destructive metal cations
                    and removing hypochlorous acid/hypochlorite ion by forming an
                    organochloramine. EDA preservation of chlorite also preserves the
                    integrity of chlorate which can increase in unpreserved samples as a result
                    of chlorite degradation. EDA also preserves the integrity of bromate
                    concentrations by binding with hypobromous acid/hypobromite ion which
                    is an intermediate formed as a by-product of the reaction of either ozone or
                    hypochlorous acid/hypochlorite ion with bromide ion.  If hypobromous
                    acid/hypobromite ion is not removed from the matrix, further reactions
                    may form bromate ion.
SAMPLE SHIPMENT AND STORAGE - All samples must be chilled during
shipment and must not exceed 10 C during the first 48 hours after collection.
Samples must be confirmed to be at or below 10 C when they are received at the
laboratory. Samples stored in the lab must be held at or below 6 C and protected
from light until analysis. Samples should not be frozen.  Sample preservation and
holding times for the anions are as follows:
Analyte
Bromate
Chlorate
Chlorite
Bromide*
Preservation
50 mg/L EDA, store at < 6 C
50 mg/L EDA, store at < 6 C
50 mg/L EDA, store at <6 C
50 mg/L EDA, store at <6 C
Holding Time
28 days
28 days
14 days
28 days
              *Source and raw water only.

9.      QUALITY CONTROL

       9.1     Each laboratory using this method is required to operate a formal quality control
              (QC) program. The requirements of this program consist of an initial
              demonstration of laboratory capability (IDC), and subsequent analysis in each
              Analysis Batch (Sect. 3.1) of a Laboratory Reagent Blank (LRB), Continuing
              Calibration Check Standards (CCCs), Laboratory Fortified Blank (LFB),
              Instrument Performance Check Standard (IPC), Laboratory Fortified Sample
              Matrix (LFSM) and either Laboratory Fortified Sample Matrix Duplicate
              (LFSMD) or a Field Duplicate (FD) Sample.  This section details the specific
                                        326.0-18

-------
       requirements for each of these QC parameters for both the conductivity and
       absorbance detectors used in this application. Although the Detection Limits and
       MRLs may differ, the QC requirements and acceptance criteria are the same for
       both detectors. The QC criteria discussed in the following sections are
       summarized in Section 17, Tables 4 and 5.  These criteria are considered the
       minimum acceptable QC criteria, and laboratories are encouraged to institute
       additional QC practices to meet their specific needs.

9.2    INITIAL DEMONSTRATION OF CAPABILITY (IDC) - Requirements for the
       Initial Demonstration of Capability are described in the following sections and
       summarized in Section 17, Table 4.

       9.2.1  INITIAL DEMONSTRATION OF LOW SYSTEM BACKGROUND -
             Before any field samples are analyzed, and any time a new set of reagents
             is used, it must be demonstrated that a laboratory reagent blank is
             reasonably free of contamination and that the criteria in Section 9.4 are
             met.

       9.2.2  INITIAL DEMONSTRATION OF ACCURACY - Prior to the analysis of
             the IDC samples, verify calibration accuracy with the preparation and
             analysis of a mid-level QCS as defined in Section 9.11. If the analyte
             recovery is not + 15% of the true value, the accuracy of the method is
             unacceptable.  The source of the problem must be identified and  corrected.
             After the accuracy of the calibration has been verified, prepare and analyze
             7 replicate LFBs fortified at a recommended concentration of 20  ug/L for
             the conductivity detector or near the mid-range of the initial calibration
             curve. For the absorbance detector, prepare 7 replicate LFBs fortified at a
             recommended concentration of 2.0 ug/L bromate. Sample preservatives as
             described in Section 8.1.3 must be added to all LFBs. The average
             recovery of the replicate values must be within  15%  of the true value.

       9.2.3   INITIAL DEMONSTRATION OF PRECISION - Using the same set of
             replicate data generated for  Section 9.2.2, calculate the standard deviation
             and percent relative standard deviation of the replicate recoveries.  The
             percent relative standard deviation (%RSD) of the results of the replicate
             analyses must be < 20%.

       9.2.4  DETECTION LIMIT DETERMINATION - Prepare and analyze at least 7
             replicate LFBs at a concentration estimated to be near the Detection Limit
             over at least 3 days using the procedure described in Section 11.  This
             fortification level may be estimated by selecting a concentration with a
             signal of 2 to 5 times the noise level.  The appropriate  concentration will
             be dependent upon the sensitivity of the 1C system being used. Sample
                                 326.0-19

-------
             preservatives as described in Section 8.1.3 must be added to these
             samples. Calculate the Detection Limit using the equation

                    Detection Limit = St(n. ^  ^. alpha = 099)

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

             NOTE: Calculated Detection Limits need only be less than V3 of the
             laboratory's MRL to be considered acceptable. Do not subtract blank
             values when performing Detection Limit calculations. The Detection
             Limit is a statistical determination of precision only.(2) No precision and
             accuracy criteria are specified.

9.3     MINIMUM REPORTING LEVEL (MRL) - The MRL is the threshold
       concentration of an analyte that  a laboratory can expect to accurately quantitate in
       an unknown sample. The MRL should not be established at an analyte
       concentration that is less than either three times the Detection Limit or a
       concentration which would yield a response less than a signal-to-noise (S/N) ratio
       of five. Depending upon the study's data quality objectives it may be set at a
       higher concentration. The lowest calibration standard must be at or below the
       MRL and therefore, the MRL must never be established at  a concentration
       lower than the lowest calibration standard.

9.4     LABORATORY REAGENT BLANK (LRB) - A LRB is required with each
       Analysis Batch (Sect. 3.1) of samples to determine any background system
       contamination. If within the retention time window of any analyte, the LRB
       produces a peak that would prevent the determination of that analyte, determine
       the source of contamination and eliminate the interference before processing
       samples.  Background contamination must be reduced to an acceptable level
       before proceeding.  Background from method analytes or contaminants that
       interfere with the measurement of method analytes must be below  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 in the analysis batch.

       9.4.1.  EDA must be added to the LRB  at 50 mg/L.  By including EDA in the
             LRB, any potential background contamination from the EDA will be
             identified.
                                 326.0-20

-------
       9.4.2  When the PCR method is used for low level bromate analysis on samples
             from public water systems (PWSs) which employ chlorine dioxide
             disinfection, the matrix must be pretreated to remove the potentially
             interfering chlorite anion (Sect. 11.1.4).  When these types of pretreated
             samples, or any type of pretreatment is applied to field samples included as
             part of an analysis batch, a second LRB must be prepared, pretreated and
             analyzed to confirm no background effects of the pretreatment are present.
             If the analysis batch contains only pretreated samples, then only a
             pretreated LRB is required.

9.5    CONTINUING CALIBRATION CHECK (CCC) - CCCs are prepared in the
       same manner as the Calibration Standards (Sect. 7.2.3), using reagent water and
       EDA as described in Section 8.1.3.  They are analyzed during an analysis batch at
       a required frequency to confirm that the instrument meets initial calibration
       criteria.  See Section 10.3 for concentration requirements, frequency requirements,
       and acceptance criteria.

9.6    LABORATORY FORTIFIED BLANK (LFB) - A LFB is required with each
       analysis batch to confirm acceptable method accuracy.  Since calibration solutions
       are prepared in large volumes and can be used over an extended period of time,
       the integrity of the concentration of the solution used to fortify the LFSM is
       checked by preparing the LFB using the same Analyte Stock Standard Solutions
       used to prepare the LFSM  fortification solution. The fortified concentration of the
       LFB should be rotated between, low, medium, and high concentrations from batch
       to batch. The low concentration LFB must be as near as practical to, but no more
       than two times the MRL. Similarly, the high concentration should be near the
       high end of the calibration range established during the initial calibration (Sect.
       10.2). The recovery of all analytes  fortified at the low concentration must be 75-
       125% of the true value, and 85-115% when fortified at the medium and high
       concentrations.  If the LFB recovery for an analysis batch does not meet these
       recovery criteria, the data are considered invalid, and the source of the problem
       must be identified and resolved before continuing with analyses.
LFB Fortified Concentration Range
MRL to 2 x MRL
2 x MRL to highest calibration level
LFB Recovery Limits
75-125 %
85-115%
9.7    SURROGATE RECOVERY - The surrogate standard is fortified into all samples,
       blanks, CCCs, QDCSs, LRBs, and LFSMs and LFSMDs prior to analysis. It is
                                 326.0-21

-------
       also added to the calibration curve and calibration check standards. The surrogate
       is a means of assessing chromatographic method performance.

       9.7.1   Surrogate recoveries must fall between 90-115% for proper instrument
              performance and analyst technique to be verified.  The recovery range for
              the surrogate is extended to 115% to allow for the potential contribution of
              trace levels of dichloroacetate as ahalogenated organic disinfection by-
              product (DBF) of dichloroacetic acid (DCAA).  Background levels of this
              organic DBF are rarely observed above 50 |-ig/L (0.05 mg/L) which
              constitutes only 5% of the 1.00 mg/L recommended fortified
              concentration.

       9.7.2   When surrogate recovery from a sample, blank, or CCC is less than 90%
              or greater than 115%, check (1) calculations to locate possible errors, (2)
              standard solutions for degradation, (3) contamination, and (4) instrument
              performance.  If those steps do not reveal the cause of the problem,
              reanalyze the sample.

       9.7.3   If the reanalysis meets the surrogate recovery criteria, report only data for
              the reanalyzed  sample.

       9.7.4   If the sample reanalysis fails the 90-115% surrogate recovery criteria, the
              analyst should  check the calibration byre-injecting the most recently
              acceptable calibration standard. If the calibration standard fails the criteria
              of Section 10.3, recalibrate as described in Section 10.2. If the calibration
              standard is acceptable, preparation and analysis of the sample should be
              repeated provided the sample is still within the holding time. If this
              sample reanalysis also fails the recovery criteria, report all data for that
              sample as suspect  due to surrogate recovery.

       9.7.5   If a laboratory chooses to monitor exclusively for trace bromate using PCR
              and the UV/VIS absorbance detector, suppression of the eluent MUST be
              used and the surrogate added and monitored on the conductivity detector
              and the appropriate QC criteria for the surrogate as outlined in Section
              9.7.1 must be met.
9.8    LABORATORY FORTIFIED SAMPLE MATRIX (LFSM) - Analysis of LFSMs
       are required in each analysis batch and are used to determine that the sample
       matrix does not adversely affect method accuracy. Additional LFSM
       requirements, as described in Section 9.8.4, apply when the PCR system is used
       for low level bromate in waters disinfected with chlorine dioxide. If the
       occurrence of target analytes in the samples is infrequent, or if historical trends are
                                  326.0-22

-------
unavailable, a second LFSM, or LFSMD (Sect. 9.9), must be prepared, and
analyzed from a duplicate of the field sample used to prepare the LFSM to assess
method precision. Analytical batches that contain LFSMDs will not require the
analysis of a Field Duplicate (Sect. 9.9).  If a variety of different sample matrices
are analyzed regularly, for example, drinking water from groundwater and surface
water sources, method performance should be established for each.  Over time,
LFSM data should be documented for all routine sample sources for the
laboratory.

9.8.1   Within each analysis batch, a minimum of one field sample is fortified as a
       LFSM for every 20 samples processed. The LFSM is prepared by spiking
       a sample with an appropriate amount of the appropriate Analyte PDS
       (Sect. 7.2.1.2). Select a spiking concentration at least twice the matrix
       background concentration, if known.  Use historical data or rotate through
       a range of concentrations when selecting a fortifying concentration.
       Selecting a duplicate bottle of a sample that has already been analyzed aids
       in the selection of appropriate spiking levels.

9.8.2   Calculate the  percent recovery (%R) for each analyte using the equation


                     %R=  (*"B)
       where
       A = measured concentration in the fortified sample
       B = measured concentration in the unfortified sample, and
       C = fortification concentration.

9.8.3   Analyte recoveries may exhibit a matrix bias. For samples fortified at or
       above their native concentration, recoveries should range between 75 -
       125%.   If the accuracy of any analyte falls outside the designated range,
       and the CCC performance for that analyte is shown to meet the acceptance
       criteria, 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.

9.8.4   When the PCR method is used for low level bromate analysis on field
       samples from PWSs which employ chlorine dioxide disinfection and
       consequently contain chlorite, a LFSM must be prepared, exclusively for
       trace bromate, for each of these field samples.  Initially, the field sample is
       analyzed and chlorite, chlorate and bromide levels are determined. Then,
       a second aliquot of field sample is pretreated to remove chlorite, as
                           326.0-23

-------
             described in Section 1 1 .1 .4, and analyzed to determine native bromate
             concentration.  A third aliquot of the field sample then must be fortified
             with bromate, pretreated as described in Section 1 1.1.4 to remove chlorite,
             and analyzed to assess bromate recovery from that matrix.  This additional
             QC is required to rule out matrix effects and to confirm that the laboratory
             performed the chlorite removal step (Sect. 11.1.4) appropriately. This
             LFSM should be fortified with bromate at concentrations close to but
             greater than the level determined in the native sample. Recoveries are
             determined as described above (Sect. 9.8.2). Samples that fail the LFSM
             percent recovery criteria of 75 -  125% must be reported as suspect/matrix.

9.9    FIELD DUPLICATE OR LABORATORY FORTIFIED SAMPLE MATRIX
       DUPLICATE (FD or LFSMD) - Within each analysis batch, a minimum of one
       Field Duplicate (FD) or Laboratory Fortified Sample Matrix Duplicate (LFSMD)
       must be analyzed.  Duplicates check the precision associated with sample
       collection, preservation, storage, and laboratory procedures. If target analytes are
       not routinely observed in field samples, a LFSMD  should be analyzed rather than
       aFD.

       9.9. 1  Calculate the relative percent difference (RPD) for duplicate
             measurements (FD1 and FD2) using the equation


                       RPD-
                               (FD1+FD2)/2
       9.9.2  If a LFSMD is analyzed instead of a Field Duplicate, calculate the relative
             percent difference (RPD) for duplicate LFSMs (LFSM and LFSMD) using
             the equation


                    RPD-   \LFSM-LFSMD\   ^QQ
                            (LFSM+ LFSMD)/2

       9.9.3  RPDs for FDs and duplicate LFSMs should fall in the range presented in
             the table below for samples fortified at or above their native concentration.
             Greater variability may be observed when LFSMs are spiked near the
             MRL.
                                 326.0-24

-------
Concentration Range
MRL to 5 x MRL
5 x MRL to highest calibration level
RPD Limits
 20 %
 10 %
             If the accuracy of any analyte falls outside the designated range, and the
             laboratory performance for that analyte is shown to meet the acceptance
             criteria in the LFB, the recovery for that analyte 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

9.10   INSTRUMENT PERFORMANCE CHECK - The low-level CCC Standard is
       evaluated in each analytical batch in order to confirm proper instrument
       performance (Sect. 10.3). This analysis confirms the MRL and demonstrates
       proper chromatographic performance at the beginning of each analysis batch.
       Chromatographic performance is judged by calculating the Peak Gaussian Factor
       (PGF), which is a means to measure peak symmetry and monitor retention time
       drift in the surrogate peak over time. The PGF, as determined below, must fall
       between 0.80 and 1.15, and the retention time for the surrogate must be at least
       80% of the initial retention time when the 1C column was new.  If these criteria
       are not met, corrective action must be performed prior to analyzing additional
       samples. Major maintenance such as replacing columns requires repeating the
       IDC determination (Sect. 9.2).

       9.10.1 The PGF is calculated using the equation
                         PGF =
             where
             W(l/2) is the peak width at half height, and
             W (V10) is the peak width at tenth height.

             NOTE: Values for WO/2) and W (V10) for each peak can be attained
             through most data acquisition software packages.

       9.10.2 Small variations in retention time can be anticipated when a new solution
             of eluent is prepared but if sudden shifts of more than 5% are observed in
             the surrogate retention time, some type of instrument problem should be
                                 326.0-25

-------
                    suspected. Potential problems include improperly prepared eluent,
                    erroneous method parameters such as flow rate or some other system
                    problem. The chromatographic profile (elution order) of the target anions
                    following an ion chromatographic analysis should closely resemble the
                    profile displayed in the test chromatogram that was shipped when the
                    column was purchased. As a column ages, it is normal to see a gradual
                    shift and shortening of retention times, but if after several years of use,
                    extensive use over less than a year, or use with harsh samples, this
                    retention time has noticeably shifted to any less than 80% of the original
                    recorded value, the column requires cleaning or replacement,  especially if
                    resolution problems are beginning to occur between previously resolved
                    peaks. A laboratory should retain a historic record of retention times for
                    the surrogate and all the target anions to provide evidence of an analytical
                    column's efficiency.

              9.10.3 If a laboratory chooses to monitor exclusively for trace bromate using PCR
                    and the UV/VIS absorbance detector, suppression of the eluent MUST be
                    used and the surrogate added and monitored on the conductivity detector
                    and the appropriate QC criteria for the surrogate as outlined in Section
                    9.7.1 must be met.

       9.11    QUALITY CONTROL SAMPLE (QCS) - Each time new Calibration Standards
              (Sect. 7.2.3) are prepared, or at least quarterly, analyze a QCS from a source
              different from the source of the calibration standards.  The acceptance criteria for
              the QCS is 85-115% of the true value. If measured analyte concentrations are not
              of acceptable accuracy, check the entire analytical procedure to locate and correct
              the problem source.

10.    CALIBRATION AND STANDARDIZATION

       10.1    Demonstration of acceptable initial calibration is  required prior to performing the
              IDC and before any samples are analyzed. It is also required intermittently
              throughout sample analysis to meet required QC performance criteria summarized
              in Tables 4 and 5.  Initial calibration verification is performed using a QCS (Sect.
              9.11) as well as with each analysis batch using Continuing Calibration Check
              Standards. The procedure for establishing the initial  calibration curve is
              described in Section 10.2. The procedure to verify the calibration with each
              analysis batch is described in Section 10.3.

       10.2    INITIAL CALIBRATION

              10.2.1 Establish ion chromatographic configuration and operating parameters
                    equivalent to those indicated in Table 1 and Figure 1.
                                        326.0-26

-------
10.2.2  Estimate the calibration range over which the instrument response is
       linear.  On the conductivity detector for the four target analytes (chlorite,
       bromate, bromide and chlorate) the linear range should cover the expected
       concentration range of the field samples and should not extend over more
       than two orders of magnitude in concentration. The method development
       data were collected on single linear calibrations that spanned 5 to 500 ug/L
       for chlorite, bromide and chlorate and 5 to 100 ug/L bromate for the
       conductivity detector and 0.5 to 15.0 ug/L for bromate on the absorbance
       detector.

       10.2.2.1 If quantification is desired over a larger range, then two or more
               separate calibration curves must be prepared.

       10.2.2.2 A minimum of five Calibration Standards (Sect. 7.2.3) should be
               prepared for each calibration. It is recommended that at least
               four of the Calibration Standards are at a concentration > the
               MRL. Because high concentrations of chlorite can interfere with
               the postcolumn analysis of low levels of bromate, the
               conductivity and absorbance detectors must be calibrated
               separately.

       10.2.2.3 When quantitated using the initial calibration curve, each
               calibration point, except the lowest point, for each analyte should
               calculate to be 85-115% of its true value.  The lowest point
               should calculate to be 75-125% of its true value.  Failure to meet
               this criteria may indicate future difficulty in meeting CCC QC
               requirements during the analysis batch.

       10.2.2.4 Since the concentration ranges in actual field samples by
               conductivity detection for chlorite, bromide and chlorate are
               expected to cover two orders of magnitude, the use of calibration
               standards in the range 5 - 500 |-ig/L is recommended.

       10.2.2.5 Bromate concentrations are expected to be significantly lower. It
               is suggested that the conductivity detector be calibrated using
               bromate calibration standard levels in the range 5-100 |-ig/L.
               Additionally, report values for bromate by conductivity ONLY
               when they are measured by the PCR above 15.0 ug/L. The
               conductivity detector may exhibit a response for bromate at
               concentrations below  15.0 ug/L, but these should not be reported.
               When using both detectors, PCR results for bromate in this range
               (5-15 ug/L) will have far better precision and accuracy.
                           326.0-27

-------
       10.2.3 Prepare a set of at least 5 calibration standards as described in Section
             7.2.3. The lowest concentration calibration standard must be at or below
             the MRL, which may depend on system sensitivity.

       10.2.4 Inject 225 |_iL of each calibration standard and tabulate peak area responses
             against the concentration for the four target analytes, the surrogate from
             the conductivity detector, and bromate from the postcolumn absorbance
             detector. Prepare calibration curves using linear regression analysis for
             each analyte on the conductivity detector and using a quadratic polynomial
             function for bromate on the absorbance detector.

             10.2.4.1  Use of peak areas are strongly recommended since they have
                      been found to be more consistent, in terms of quantitation, than
                      peak heights.  Peak height can tend to be suppressed as a result of
                      high levels of common anions in a given matrix which can
                      compete for exchange sites leading to peak broadening.
                      However, poorly drawn baselines can have a more significant
                      influence on peak areas than peak heights. It is the analyst's
                      responsibility to review all chromatograms to ensure accurate
                      baseline integration of target analyte peaks.

       10.2.5 After establishing (or re-establishing) calibration curves, the accuracy of
             this calibration must be verified through the analysis of a QCS or an
             externally prepared second source standard. The QCS should be prepared
             at a concentration near the middle of the calibration range. As specified in
             Section 9.11, determined concentrations must fall within  15% of the
             stated values.

10.3   CONTINUING CALIBRATION CHECK (CCC) - Initial calibrations may be
       stable for extended periods of time.  Once the calibration curves have been
       established for both the conductivity and absorbance detectors, they must be
       verified for each analysis batch prior to conducting any field sample analyses
       using CCCs. The first CCC each day must be at or below the MRL in order to
       verify instrument sensitivity prior to any analyses. Subsequent CCCs must be run
       after every 10 field samples  and should alternate between a mid- and high-level
       CCC. LRBs, CCCs, LFSMs and LFSMDs are not counted as field samples.

       10.3.1 A low-level CCC must be determined to be valid each day prior to
             analyzing any samples by injecting an aliquot of the appropriate CCC
             under the same instrumental conditions used to collect the initial
             calibration.  Since two detectors are incorporated in this method, this must
             be accomplished by using a mixed calibration check standard for the four
             conductivity analytes and a separate low-level bromate CCC for the
                                 326.0-28

-------
                    absorbance detector.  The low-level CCC for both detectors must be at or
                    below the MRL. Percent recovery for the low-level CCC must be in the
                    range of 75 - 125% before the analyst is allowed to analyze samples.

              10.3.2 Additional CCC standards must be analyzed after every tenth field sample
                    and at the end of the analysis batch. If more than 10 field samples are
                    included in an analysis batch, the analyst should alternate between the
                    mid- and high-level CCC Standards. Percent recovery for the mid- and
                    high-level CCCs must be in the range of 85 - 115%.

              10.3.3 If the calibration verification criteria listed above are not met, or the
                    retention times shift more than  2% from the last acceptable initial or
                    continuing calibration check standard for any analyte, then all samples
                    analyzed after the last acceptable calibration check standard are considered
                    invalid and must be reanalyzed.  The source of the problem must be
                    identified and resolved before reanalyzing the samples or continuing with
                    the analyses.

                    10.3.2.1 In the case where the end calibration failed to meet performance
                            criteria, but the initial and middle calibration check standards
                            were acceptable, the samples bracketed by the acceptable
                            calibration check standards may be reported. However, all field
                            samples between the middle and end calibration check standards
                            must be reanalyzed.
11.    PROCEDURE
       11.1    SAMPLE PREPARATION
              11.1.1 For refrigerated or field samples arriving at the laboratory cold, ensure the
                    samples have come to room temperature prior to conducting sample
                    analysis by allowing the samples to warm on the bench for at least 1 hour.

              11.1.2 Prepare a 10.0-mL aliquot of surrogate fortified sample which can be held
                    for direct manual injection or used to fill an autosampler vial.  This is done
                    by adding 20 |_iL of the surrogate solution (Sect. 7.2.2) to a 20-mL
                    disposable plastic micro beaker. Next, place a 10.0-mL aliquot of sample
                    in the micro beaker and mix. These volumes may be adjusted to meet
                    specific laboratory autosampler volume requirements provided the
                    fortified surrogate concentration is at the prescribed concentration of 1.0
                    mg/L.  The sample is now ready for analysis.
                                        326.0-29

-------
      NOTE:  The less than 1% dilution error introduced by the addition of the
      surrogate is considered insignificant.  If a laboratory chooses to monitor
      exclusively for trace bromate using PCR and the UV/VIS absorbance
      detector, suppression of the eluent MUST be used and the surrogate added
      and monitored on the conductivity detector and the appropriate QC criteria
      for the surrogate as outlined in Section 9.7.1 must be met.

11.1.3 Using a Luer lock, plastic 10-mL syringe, withdraw the sample from the
      micro beaker and attach a 0.45-|_im particulate filter (demonstrated to be
      free of ionic contaminants) directly to the syringe. Filter the sample into
      an autosampler vial (if vial is not designed to automatically filter) or
      manually load the injection loop injecting a fixed amount of filtered, well
      mixed sample. If using a manually loaded injection loop, flush the loop
      thoroughly between sample analysis using sufficient volumes of each new
      sample matrix.

11.1.4 CHLORINE DIOXIDE - TREATED  WATERS CONTAINING
      CHLORITE - Treatment plants that use chlorine dioxide as part of their
      treatment process can produce high levels of chlorite in samples.  Since
      chlorite can interfere with the postcolumn quantitation of low levels of
      bromate as described in Section 4.6, chlorite must be removed from these
      samples prior to analysis/11'  The oxidation-reduction reaction between
      ferrous iron and chlorite(12) is used to remove chlorite without any adverse
      affects on the bromate concentration/13'

       11.1.4.1  Place a 10-mL aliquot of sample in a 20-mL micro beaker and add
               35 uL of 0.5 N sulfuric acid (Sect. 7.1.7).  After mixing, verify the
               pH is between 5 and 6 using pH test strips, add 40 uL of ferrous
               iron solution (Sect. 7.1.6), mix and allow to react for 10 minutes.
               Filter the reaction mixture using a 0.45 micron particulate filter
               (Sect. 6.10) attached to a 10-mL syringe into the barrel of a second
               syringe to which a pre-conditioned hydrogen cartridge (Sect. 6.11)
               is attached. Pass the solution through a hydrogen cartridge at a
               flow rate of approximately 2 mL per minute. Discard the first 3
               mL, and collect an appropriate volume (depending on autosampler
               vial  size) for analysis.  Add the respective volume of surrogate
               solution, depending on the volume collected. The sample is ready
               for analysis (Sect. 11.2).

               NOTE: Pretreated samples  can be held for no more than 30 hours
               after initial pretreatment. If this time has expired, the pretreatment
               steps must be repeated on a second aliquot of both the field sample
               matrix and the respective LFSM.
                           326.0-30

-------
              11.1.4.2 In order to ensure data quality, all samples from PWSs which
                      utilize chlorine dioxide which have been pretreated to remove
                      chlorite, MUST also be used to prepare a pretreated LFSM specific
                      to trace bromate. This LFSM should be fortified with bromate at
                      concentrations close to but greater than the level determined in the
                      native sample. Initially, the field sample is analyzed and chlorite,
                      chlorate and bromide levels are determined.  Then, a second
                      aliquot of field sample is pretreated to remove chlorite, as
                      described above and analyzed to determine native bromate
                      concentrations.  A third aliquot of the field sample then must be
                      fortified with bromate, pretreated to remove chlorite, and analyzed
                      to assess bromate recovery from that matrix.  This additional QC
                      is required to  rule out matrix effects and to confirm that the
                      laboratory performed the chlorite removal  step appropriately. If
                      the bromate recovery falls outside the acceptance range of 75 -
                      125% (Sect. 9.8), that particular sample should be reported as
                      suspect/matrix.

              11.1.4.3 All samples from PWSs that utilize chlorine dioxide, which have
                      been pretreated  to remove chlorite, MUST also include an
                      additional pretreated LRB in the analytical batch (Sect. 9.4.2).

              11.1.4.4 Suppressor devices which have had long term exposure to iron
                      cations may have reduced method performance in other
                      applications, such as the determination of certain common
                      inorganic anions. If reduced peak response is observed,
                      particularly for fluoride or phosphate, the suppressor should be
                      cleaned according to the manufacturer's recommendations.

11.2   SAMPLE ANALYSIS

       11.2.1  Table 1 summarizes  the recommended operating conditions for the ion
              chromatograph and delivery of the postcolumn reagent. Included in this
              table are the actual retention times and Detection Limits that were
              determined during the development of this method.  Other columns or
              chromatographic conditions maybe used if the requirements of Section 9 are
              met.

       11.2.2  Establish a valid initial calibration as described in Section 10.2 and complete
              the IDC (Sect. 9.2).  Check system calibration by analyzing a low-level
              CCC (Sect. 10.3.1) as part of the initial QC for the analysis batch and, if
              required, recalibrate  as described in Section 10.2.
                                  326.0-31

-------
             11.2.3 Inject 225 |_iL of each sample. Use the same size loop for standards and
                    samples. An automated constant volume injection system may also be used.

             11.2.4 The width of the retention time window used to make identifications should
                    be based upon measurements of actual retention time variations of standards
                    measured over several days.  Three times the standard deviation of retention
                    time can be used to calculate a suggested window size for each analyte.
                    However, the experience of the analyst should weigh heavily in the
                    interpretation of chromatograms.

             11.2.5 If the response of a sample analyte exceeds the calibration range, the sample
                    must be diluted with an appropriate amount of EDA fortified reagent water
                    and reanalyzed.

             11.2.6 Should more complete resolution be needed between any two coeluting
                    peaks, the eluent (Sect. 7.1.2) can be diluted. This will extend the run,
                    however, and will cause late eluting anions to be retained even longer.  The
                    analyst must verify that this dilution does not negatively affect performance
                    by repeating the IDC (Sect. 9.2), and by reestablishing a valid initial
                    calibration curve (Sect. 10.2). As a specific precaution, upon dilution of the
                    carbonate eluent, a peak for bicarbonate may be observed on the
                    conductivity detector within the retention time window for bromate which
                    will negatively impact the analysis.

                    11.2.6.1 Eluent dilution will reduce the overall response of an anion due to
                            chromatographic band broadening which will be evident by
                            shortened and broadened peaks.  This  will adversely effect the
                            Detection Limit for each analyte.

11.3   AUTOMATED ANALYSIS WITH METHOD 326.0

       11.3.1        Laboratories conducting analyses on large numbers of samples often prepare
                    large analysis batches that are run in an automated manner.  When
                    conducting automated analyses, careful attention must be paid to all
                    reservoirs to be certain sufficient volumes are available to sustain extended
                    operation. Laboratories must ensure that all QC performance criteria are
                    met as described in preceding sections to ensure their data are of acceptable
                    quality.
                    11.3.1.1  Special attention must be paid when the PCR reservoir is refilled.
                            The PCR is stable for only 24 hours and consequently the reservoir
                            must be regularly filled with freshly prepared reagent. Since this is
                            a pneumatically driven system, the baseline will require a
                                         326.0-32

-------
                             minimum often minutes to restabilize after the reservoir has been
                             refilled and the bottle repressurized.

              11.3.2 Because this method has two detectors that require independent calibration,
                    analysis sequences must be carefully constructed to meet required QC
                    specifications and frequency (Sect. 17, Table 5). To help with this task, an
                    acceptable sequence for a sample analysis batch, with all the method-
                    required QC, is shown in Table 6. This schedule is included only as an
                    example of a hypothetical analysis batch where the analyst desires to collect
                    data using both detectors. Within the analysis batch, references to exact
                    concentrations for the CCCs are for illustrative purposes only. The analyses
                    for sample #14 provides an example of the QC requirements for a complete
                    conductivity and trace bromate PCR analysis of a sample from a PWS
                    employing chlorine dioxide disinfection.

12.    DATA ANALYSIS AND CALCULATIONS

       12.1    Identify the method analytes in  the sample chromatogram by comparing the
              retention time of the suspected analyte peak to the retention time of a known analyte
              peak in a calibration standard. If analyte retention times have shifted (generally
              towards shorter times) since the initial calibration, but are still within acceptance
              criteria and are reproducible during the analysis batch, the analyst should use the
              retention time in the daily calibrations to confirm the presence or absence of target
              analytes.

       12.2    Compute sample concentration  using the initial calibration curve generated in
              Section 10.2.

       12.3    Report ONLY those values that fall between the MRL and the highest calibration
              standard.  Samples with target analyte responses exceeding the highest standard
              must be diluted and reanalyzed.  When this is not possible the alternate calibration
              procedures described in Section 11.2.5  must be followed.

              12.3.1 Report bromate concentrations using the postcolumn UV/Vis absorbance
                    detector when they fall between the MRL and 15.0 ug/L. When bromate
                    concentrations exceed 15.0 ug/L, as detected by UV/Vis absorbance, either
                    report by conductivity, calibrate the postcolumn UV/Vis absorbance detector
                    to a higher bromate concentration, or dilute the sample.

       12.4    Report analyte concentrations in |_ig/L (usually with two significant figures).

       12.5    Software filtering of the postcolumn UV/Vis absorbance signal is recommended to
              improve the precision of peak measurements, minimize non-random noise and
                                         326.0-33

-------
             improve peak appearance, ensuring that all QC requirements for the method are
             met. Olympic smoothing (25 points, 5 seconds with 1 iteration) was chosen using
             peak area for quantitation because it was determined to have minimal effect on peak
             height and/or area.(14) The use of alternate smoothing routines is acceptable
             providing all QC criteria are met.

13.    METHOD PERFORMANCE

       13.1   Table 1 lists the standard conditions, typical retention times and single laboratory
             Detection Limits in reagent water, as determined for each of the inorganic oxyhalide
             DBFs and bromide.

       13.2   Table 2 shows the precision and accuracy of the trace bromate measurement,
             evaluated on both detectors, at two fortified concentrations, in reagent water (RW),
             a simulated high ionic strength water (HIW) and a simulated high organic (HOW)
             content water. The mean recovered bromate concentration (accuracy relative to the
             fortified level) and the precision (expressed as %RSD of the replicate analyses) are
             tabulated.  The HIW was designed to simulate a high ionic strength field sample and
             the HOW designed to simulate a high organic content field sample. The HIW was
             prepared from reagent water which was fortified with the common anions of
             chloride at 100 mg/L, carbonate at 100 mg/L, nitrate at 10.0 mg/L as nitrogen,
             phosphate at 10.0 mg/L as phosphorous, and sulfate at 100 mg/L.(1) The HOW was
             prepared from reagent water fortified with 1.0 mg/L humic acid.(1)

       13.3   Table 3 summarizes the  single laboratory accuracy (%Recovery) and precision (%
             RSD) for each anion included in the method in a variety of waters for the standard
             conditions identified in Table 1.

14.    POLLUTION PREVENTION

       14.1   For information about pollution prevention that may be applicable to laboratories
             and research institutions, consult "Less is Better: Laboratory Chemical Management
             for Waste Reduction," available from the American Chemical Society's Department
             of Government Regulations and Science Policy,  1155  16th Street N.W.,
             Washington D.C. 20036, (202) 872-4477.

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.  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
                                         326.0-34

-------
             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, consult "The Waste Management Manual for
             Laboratory Personnel" also available from the American Chemical Society at the
             address in Section 14.1.

16.    REFERENCES

      1.     U.S. EPA Method 300.1. "Determination of Inorganic Anions in Drinking Water by
             Ion Chromatography". EPA Document number: EPA/600/R-98/118. NTIS number
             PB98-169196INZ.

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

      3.     Wagner, H..P., Pepich, B.V., Hautman, D.P. and Munch, D.J. "US Environmental
             Protection Agency Method 326.0, a New Method for Monitoring Inorganic
             Oxyhalides and Optimization of the Postcolumn Derivatization for the Selective
             Determination of Trace Levels of Bromate." L  Chro. A, 2002, 956, 93-101.

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

      5.     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.

      6.     "Carcinogens-Working with Carcinogens," Publication No. 77-206, Department of
             Health, Education, and Welfare, Public Health Service, Center for Disease Control,
             National Institute of Occupational Safety and Health, Atlanta, Georgia, August
             1977.

      7.     "Safety In Academic Chemistry Laboratories," 3rd Edition, American Chemical
             Society Publication,  Committee on Chemical Safety, Washington, D.C., 1979.
       8.     Wagner, H..P., Pepich, B.V., Hautman, D.P. and Munch, D.J. "Improving the
             Performance of EPA Method 300.1 for Drinking Water Compliance Monitoring."
            J. Chrom. A.. Special Edition of the 2002 International Ion Chromatography
            Smyposium (in press).
                                       326.0-35

-------
9.      Standard Methods for the Examination of Water and Wastewater, "Method 4500-
       C1O2,C Amperometric Method I (for the determination of Chlorine Dioxide)," 19th
       Edition of Standard Methods (1995).

10.    Hautman, D.P. & Bolyard, M. "Analysis of Oxyhalide Disinfection By-products and
       other Anions of Interest in Drinking Water by Ion Chromatography."_L Chrom. A,
       1992, 602,65-74.

11.    Wagner, H..P., Pepich, B.V., Hautman, D.P. and Munch, D.J. "Analysis of 500 ppt
       Levels of Bromate in Drinking Waters Using Direct Injection Suppressed Ion
       Chromatography with a Single, Pneumatically Delivered Postcolumn Reagent."_L
       Chrom. A, 1999, 850, 119-129.

12.    latrou, A. and Knocke, W.R. "Removing Chlorite by the Addition of Ferrous Iron".
       Journal of the AWWA. Research and Technology, (November, 1992), 63-68.

13.    Wagner, H..P., Pepich, B.V., Hautman, D.P. and Munch, D.J. "Eliminating the
       Chlorite Interference in US Environmental Protection Agency Method 317.0 Permits
       the Analysis of Trace Bromate Levels in all Drinking Water Matrices." L  Chrom..
       A, 2000, 882, 309-319.

14.    Schibler, J.A., "  Improving Precicion and Accuracy with Software-based Signal
       Filtering". American Laboratory. (December, 1997), 63-64.
                                  326.0-36

-------
17.  TABLES. DIAGRAMS. FLOWCHARTS AND VALIDATION DATA

TABLE 1.     CHROMATOGRAPHIC CONDITIONS AND METHOD DETECTION LIMITS IN
              REAGENT WATER FOR THE INORGANIC OXYHALIDE DISINFECTION BY-
              PRODUCTS AND BROMIDE.

Standard Conditions and Equipment(a):
Ion Chromatograph:
Sample Loop:
Eluent:
Eluent Flow:
Columns :
Typical System Backpressure:
Conductivity Suppressor:
PCR Suppressor:
Detectors:
Postcolumn Reagent Flow:
Postcolumn Reactor Coil:
Postcolumn Heater:
Postcolumn Regenerant
Total analysis time:
Dionex DX500
225 jiL
9.0 mM NajCOg
1.3 mL/min
Dionex AG9-HC / AS9-HC, 4 mm
2300 psi
ASRS-1, external water mode, 100 mA current for conductivity
ASRS-1 used with sulfuric acid regenrant to acidify the PCR
Dionex CD20 suppressed conductivity detector, background
conductivity: 24 [iS
Dionex AD20 Absorbance Detector, 10 mm cell path length, set at
352 nm (deuterium lamp)
0.4 mL/min
knitted, potted for heater, 500 uL internal volume
80 C
150 mN H2SO4, 2.5 mL/min, effluent pH < 2
25 minutes
Analyte Retention Times and Detection Limits :
Analyte
Chlorite
Bromate (c)
Bromate (d)
Surrogate: DCA
Bromide
Chlorate
Retention Time 
-------
TABLE 2.     SINGLE LABORATORY PRECISION IN VARIOUS MATRICES FOR
             BROMATE BY ABSORBANCE DETECTION.

Matrix Detection
Reagent Absorbance
Water
Absorbance
High Ionic Absorbance
Water
Absorbance
High Absorbance
Organic
Water Absorbance
PRECISION
Fortified
Cone.
(ng/L)
1.0
5.0
1.0
5.0
1.0
5.0
# of Reps.
8
8
8
8
7
8
Mean
(ng/L)
1.1
5.2
1.1
5.2
1.1
5.2
%
RSD
4.4
2.1
4.2
2.0
3.4
3.2
   Standard Conditions: Same as listed in Table 1.
                                   326.0-38

-------
TABLE 3.     SINGLE-LABORATORY PRECISION AND RECOVERY FOR THE
             INORGANIC DISINFECTION BY-PRODUCTS, BROMIDE AND
             SURROGATE.
Analyte
Chlorite



Bromate by
Conductivity



Bromide



Chlorate



Surrogate


Matrix
RW
HIW
HOW

RW
HIW
HOW

RW
HIW
HOW

RW
HIW
HOW

RW
HIW
HOW
Fortified
Cone.
(ug/L)
100
500
100
500
100
500

10.0
25.0
10.0
25.0
10.0
25.0

10.0
25.0
10.0
25.0
10.0
25.0

100
500
100
500
100
500

1.00
1.00
1.00
#of
Replicates
8
8
8
8
8
8

8
8
8
8
8
8

8
8
8
8
8
8

8
8
8
8
8
8

8
8
8
8
8
8
Mean
% Recovery
107
108
102
106
99.3
107

102
99.8
103
92.9
101
97.6

97.5
104
108
104
104
99.7

111
104
99.0
100
101
105

108
106
103
105
108
108
%RSD
3.0
1.2
2.0
0.71
2.7
0.49

4.6
4.3
3.8
11
8.1
5.9

5.3
5.1
4.8
5.0
6.1
3.7

1.7
0.97
2.3
0.66
2.8
1.1

6.1
4.7
1.3
2.1
4.4
4.0
     RW = Reagent Water; HIW = High Ionic Strength Water; HOW = High Organic Water
                                   326.0-39

-------
TABLE 4.      INITIAL DEMONSTRATION OF CAPABILITY QC REQUIREMENTS.
  Reference
 Requirement
     Specification and Frequency
 Acceptance Criteria
  Sect. 9.2.1
   and 9.4
Initial
Demonstration
of Low System
Background
Analyze a method blank (LRB) and
determine that all target analytes are
below l/2 of the proposed MRL prior to
performing the IDC
The LRB
concentration must be
< 1/3 of the proposed
MRL
  Sect. 9.2.2
Initial
Demonstration
of Accuracy
(IDA)
Run mid-level QCS and determine
recovery.

Conductivity: analyze 7 replicate LFBs
recommend fortify at 20 ug/L
Absorbance: analyze 7 replicate LFBs
recommend fortify with bromate at 2.0
ug/L Calculate average recovery of IDA
replicates
QCS recovery must be
 15% of true value.

Mean % recovery for
IDA replicates  must
be 15% of true
value.
  Sect. 9.2.3
Initial
Demonstration
of Precision
(IDP)
Calculate the %RSD of the IDA
replicates.
%RSDmustbe<20%
  Sect. 9.2.6
Detection Limit
Determination
Select a fortifying level at 3-5 times the
estimated instrument detection limit at or
lower than the MRL. Analyze 7
replicate LFBs
Calculate over at least 3 days using
equation in Section 9.2.6 - do not
subtract blank
Detection Limit must
be < 1/3 the MRL
                                           326.0-40

-------
TABLE 5.  QUALITY CONTROL REQUIREMENTS (SUMMARY).
 Reference
  Requirement
   Specification and Frequency
   Acceptance Criteria
  Sect. 8.3
Sample Holding
Time /
Preservation
Bromate   28 days, refrig. at <6 C /
          EDA Preservation
Bromide   28 days, EDA Permitted
Chlorate   28 days, refrig. at <6 C  /
          EDA Preservation
Chlorite   14 days, refrig. at <6 C  /
          EDA Preservation
Holding time and
temperature must not be
exceeded. EDA added to all
samples
  Sect. 9.4
Laboratory
Reagent Blank
(LRB)
Include LRB with every analysis
batch (up to 20 samples)
Analyze prior to analyzing field
samples
All analytes must be
< 1/3 MRL
  Sect. 9.4.2
  (specific
  to PCR)
PRETREATED
Laboratory
Reagent Blank
REQUIRED in any analysis batch
which includes samples which have
been pretreated to remove chlorite
prior to PCR measurement of trace
bromate.
PCR measured bromate
< 1/3 MRL
  Sect. 9.6
Laboratory
Fortified Blank
(LFB)
Laboratory must analyze LFB in
each analysis batch following the
first CCC. Calculate %REC prior to
analyzing samples
LFB recovery fortified at:
 >MRLto5XMRL
     = 75 - 125%
 >5X MRL to highest CCC
   = 85-115%
Sample results from batches
that fail LFB are invalid
  Sect. 10.2
Initial
Calibration
Conductivity: generate calibration
curve using at least 5 standards
Absorbance: generate calibration
curve using at least 5 bromate
standards
The lowest calibration
standard MUST be at or
below the MRL
4 CAL standards should be
above the MRL
  Sect. 9.5
  and Sect.
  10.3
Continuing
Calibration
Check (CCC)
Verify initial calibration by
analyzing a low level CCC prior to
analyzing samples. CCCs are then
injected after every 10 samples and
after the last sample, rotating
concentrations to cover the
calibrated range of the instrument.
Recovery for each analyte
must be 85-115% of the true
value for all but the lowest
level of calibration. The
lowest calibration level CCC
must be 75-125% of the true
value
All acceptable data MUST
be bracketed by valid CCCs
                                            326.0-41

-------
TABLE 5.  QUALITY CONTROL REQUIREMENTS (SUMMARY CONTINUED).
  Reference
 Requirement
    Specification and Frequency
  Acceptance Criteria
   Sect. 9.7
Surrogate
Dichloroacetate is added to all blanks,
samples and standards
Surrogate recovery must
be 90-115%.
Samples that fail
surrogate recovery must
be reanalyzed. If second
analysis fails label result
as suspect/matrix
   Sect. 9.8
  Sect.
  11.1.4.3
Laboratory
Fortified
Sample Matrix
(LFSM)
Must add known amount of each target
analyte to a minimum of 5% of field
samples or at least one within each
analysis batch for both detectors
LFSM must be fortified above the
native level and at no greater than 5 x
the highest field sample concentration
Calculate target analyte recovery using
formula (Sect. 9.8.2)
When field samples from chlorine
dioxide plants which contain chlorite
are pretreated prior to the PCR
measurement of trace bromate, an
additional LFSM must be prepared for
each pretreated field sample (Sect.
9.8.4)
Recovery should be
75 - 125%

If fortified sample fails
the recovery criteria,
label both as
suspect/matrix.
   Sect. 9.9
Field Duplicate
(FD)
or
Laboratory
Fortified
Sample Matrix
Duplicate
(LFSMD)
Analyze either a FD or LFSMD for a
minimum of 5% of field samples or at
least one within each analysis batch for
both detectors.

Calculate the relative percent difference
(RPD) using formula in Section 9.9.1
The RPD for
concentrations at MRL
to 5 x MRL should be 
20% on both detectors,
and  10% on both
detectors for
concentrations at 5 x
MRL to highest CCCs.  If
this range is exceeded,
label both as
suspect/matrix
   Sect. 9.10
Instrument
Performance
Check (IPC)
Calculate Peak Gaussian Factor (PGF)
using equation (Sect. 9.10.1) and
monitor retention time for surrogate in
the initial CCC each day
PGF must fall between
0.80 and 1.15
Ret. Time (RT) for
surrogate must remain
80% of initial RT when
column was new
                                            326.0-42

-------
TABLE 6.  EXAMPLE SAMPLE ANALYSIS BATCH WITH QUALITY CONTROL
         REQUIREMENTS
Injection
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Sample
Description
Laboratory reagent blank (LRB)
ICCS conductivity detector (5.0 |j,g/L)
ICCS absorbance detector (0.5 ng/L)
Laboratory Fortified Blank (LFB) -
conductivity detector
LFB - absorbance detector
Field sample 1
Field sample 1 - Laboratory Duplicate (LD) (a)
Field sample 2
Field sample 2 - Laboratory Fortified Sample Matrix
(LFSM) (a) at concentrations specific for conductivity
detector
Field sample 2 - LFSM specific for trace bromate on the
absorbance detector
Field sample 3
Field sample 4
Field sample 5
Field sample 6
Field sample 7
Field sample 8
Field sample 9
Field sample 1 0
CCCS conductivity detector (75.0 jig/L)
CCCS absorbance detector (5.0 ug/L)
Field sample 1 1
Acceptance
Criteria
< 1/3 MRL
3.75 to 6.25 ng/L
0.375 to 0.625 (ig/L
 25 % fortified level
 25 % fortified level

 15%RPD

 25% fortified level
 25% fortified level








63.8 to 86.3 (ig/L
4.25 to 5.75 ng/L

                                  326.0-43

-------
22
23
24
25
26
27
28
29
30
31
32
33
34
Field sample 12
Field sample 1 3
Field sample 14 - (finished water from PWS using chlorine
dioxide)
Pretreate (Sect. 9.3.1.2) using the acid/Fe(II) chlorite
removal procedure (Sect. 11.1.4)
Field sample 14 (b) - (finished water from PWS using
chlorine dioxide) pretreated with acid/Fe(II) (Sect. 1 1.1.4)
Field sample 14 - (finished water from PWS using chlorine
dioxide) LFSM specific for trace bromate on the
absorbance detector, pretreated with acid/Fe(II) (Sect.
11.1.4.2)
Field sample 1 5
Field sample 1 6
Field sample 1 7
Field sample 1 8
Field sample 19(b)
ECCS conductivity detector (500.0 ug/L)
ECCS absorbance detector (15.0 ug/L)



< 1/3 MRL

 25% fortified level





425 to 575 ug/L
12.8 to 17.3 ug/L
(a)   If no analytes are observed above the MRL for a sample, an alternate sample which contains
    reportable values should be selected as the laboratory duplicate. Alternately, the LFSM can be
    selected and reanalyzed as the laboratory matrix duplicate ensuring the collection of QC data for
    precision.

(b)   Field sample #19 was the final field sample permitted in this batch but 20 total field samples were
    analyzed.
    Field sample #14 was analyzed both initially and as a acid/Fe (II) pretreated sample, therefore, it
    accounted for two "field sample analyses" toward the maximum of twenty in an analysis batch (Sect.
    3.1).
                                             326.0-44

-------
                   Sample loop
     1C System
    (9.0mMCOj)
                          1C Guard & Analytical Columns
                                             Conductivity
                                              Suppressor
             Autos ampler
                     PCR
                  Suppressor
    PCR
  Delivery
  (Pneumatic)
 Regenerant
(150mNH2SQ0
             Regenerant to waste
                                             Conductivity
                                               Detector
                                      Reaction coil @
                           Absorb ance
                            Detector
                             (352 nm)
Effluent to waste
Figure 1:   Schematic detailing the configuration of postcolumn hardware addition to an ion
          chromatograph.  Mention of trade names or commercial products does not constitute
          endorsement or recommendation for use. If the requirements found in Section 9 are
          met, equivalent products or hardware can be employed.

          NOTE: In a typical Method 300.1 hardware configuration, a backpressure coil is
          included after the conductivity cell as part of the waste stream when this
          manufacture's equipment is used. These backpressure coils are not required when the
          Method 326.0 instrument configuration is employed since the additional PCR system
          components, placed in-line, function in the same capacity and provide sufficient
          backpressure.
                                     326.0-45

-------
             0.500-r
             0.400-
             0.300-
             0.200-
             0.100-
                0-
             -0.100-
                  Suppressed Conductivity Detection
             -0.200
                           r
                 0     2.5     5.0     7.5    10.0    12.5    15.0    17.5    20.0
                                           Minutes
          8.OOxlOT -
          4.00x10" -
          2.00x10- -
                  UV/Vis Absorbance Detection
                            BrO,
                              \i
                       2.5      5.0     7.5     10.0     12.5     15.0    17.5     20.0
                                             Minutes
Figure 2:   Reaent water fortified with bromate at 10 ug/L on both detectors.
                                         326.0-46

-------
                    Suppressed Conductivity Detection
0.500-
0.400-
0.300-
0.200-
C/l
i
0.100-
0-
-0.100-
-0.200-
(





|Br,(
u-ic\fii
Hp
1
^

L
1 ao3
;VliMLJl 	

3 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
Minutes
UV/Vis Absorbance Detection
8.00xl03-r
e.ooxio3-

4.00x1 03 -
2.00xl03-
o-
2. OOxl O3-
0
C1O2
[
^E


r-v-^

rr\
I I V_^Q
' I

11

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
                                           Minutes
Figure 3:   Reagent water fortified with inorganic oxyhalide disinfection by-products and
           bromide at 20.0 ug/L and bromate at 10 ug/L on both detectors.
                                         326.0-47

-------