SGS North America Inc. - newtowncreek.infonewtowncreek.info/docs2/2 Remedial Investigation/Phase 2...

SGS North America Inc. Standard Operating Procedure

of 47 SGS SOP ID: Dioxin/Furan

Date Implemented: 7/18/2013 Document Number: DC_364

Revision History

• The following change was made effective March 7, 2013: o Updated method reference section to include 8290A in section 17.7

• The following change was made effective July 16, 2013: o Added cycle information to section 9.2.2 o Added sample volume information to section 9.2.3 o Added IDOC and MDL requirements to section 12.1




TABLE OF CONTENTS 1.0 Scope and Application ............................................................................................ 4

2.0 Summary of Method ............................................................................................... 4

3.0 Safety Precautions ................................................................................................... 5

4.0 Interferences and Preventive Measures .................................................................. 5

5.0 Sample Collection, Preservation and Storage ......................................................... 6

6.0 Equipment and Supplies ......................................................................................... 6

7.0 Standards and Reagents .......................................................................................... 7

8.0 Calibration............................................................................................................... 8

9.0 Sample Preparation ............................................................................................... 10

10.0 Analytical Procedure ............................................................................................. 15

11.0 Details of Calculations .......................................................................................... 19

12.0 Quality Control Requirements .............................................................................. 21

13.0 Data Review and Reporting Requirements ........................................................... 24

14.0 Preventative Maintenance ..................................................................................... 30

15.0 Tables .................................................................................................................... 30

16.0 Definitions............................................................................................................. 45

17.0 References ............................................................................................................. 47




1.0 Scope and Application

1.1 This SOP covers all analyses of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs) in all matrices. It incorporates, standardizes, and expands on EPA Methods 8290, 1613, 23, 0023A,TO-9A, and the EPA DLM. SGS-AP provides two options for quality control. SGS-AP offers, and strongly recommends, an enhanced quality control option based on a document dubbed internally Method 8223. To ensure we meet client needs, we also use the traditional QC model, as detailed in published EPA methods, upon client request. A unified approach is taken to all analyses using these methods, and procedural details vary only according to matrix.

1.2 The provisions of this SOP may be overridden by client request and represent the default instructions to follow. In all cases, SGS-AP strives to provide top-notch advice to clients in the preparation of such documents. Such client-specific needs will be communicated to the laboratories through the use of special instructions entries in the LIMS or by use of paperwork developed specifically for the project in question.

1.3 This method is applicable to all matrices.

1.4 The limits of detection (LODs) and limits of quantitation (LOQs) in various matrices are available in the Performance Commitment document. Note that a separate listing of LODs and LOQs is provided for analyses of samples for the Department of Defense (DOD).

1.5 This method is intended to apply to all PCDD/F extractions performed at SGS Analytical Perspectives, regardless of matrix. It is consistent with the Final Rule on the Methods Innovation Rule 40 CFR Parts 63, 268, et al.; June 14 th, 2005; pp. 34538-34592. The lists of target analytes can be found in Table 4.

1.6 Our Statement of Performance thoroughly documents the method’s performance and includes the results of proficiency testing studies, audit samples, and international inter-laboratory calibration studies.

2.0 Summary of Method

2.1 For solid samples, the percent solids is determined. The samples are conditioned appropriately, depending on matrix. The samples are fortified with extraction standards and then extracted by a number of means, depending on which is appropriate for the matrix in question. Following extraction, additional conditioning may be necessary, after which the sample extracts are concentrated and fortified with additional standards for clean-up. A variety of clean-up procedures, usually including gravity-fed column chromatography with acid and base impregnated silica are




employed. Extracts are then concentrated, fortified with injection standards, and undergo final preparations for HRGC/HRMS. The extracts are analyzed by HRGC/HRMS.

3.0 Safety Precautions

3.1 Gloves must be worn when handling samples, standards, and reagents.

3.2 Safety glasses must be worn at all times when working in the laboratory.

3.3 A lab coat must be worn while dealing with samples, standards, and reagents.

3.4 In the interest of safety and pollution prevention, all spills must be cleaned up immediately.

4.0 Interferences and Preventive Measures

Three types of interferences may cause problems with this method. The three kinds of interferences are specific, non-specific, and quantitative, and each is considered below.




4.1 Specific Interferences: Specific interferences are PCDD/Fs that appear in a final extract but did not originate with that sample. Two primary sources of specific interferences are cross-contamination from other samples and, less commonly, chemical reactions that form PCDD/Fs during sample preparation procedures (e.g., extraction).

4.2 Non-specific interferences: Non-specific interferences are compound present in the final extract that create false positive signals for PCDD/Fs (e.g., polychlorinated diphenyl ethers). Ideally, sample preparation removes these interferences, but analysts should always be wary of them. An exhaustive list of non-specific interferences may be found in – Environmental Carcinogens: Methods of Analysis and Exposure Measurement Volume 11: Polychlorinated Dioxins and Dibenzofurans (IARC Scientific Publications), C. Rappe (Editor), H. R. Buser (Editor), B. Dodet (Editor), I. K. O'Neill (Editor).

4.3 Quantitative Interferences: Quantitative interferences cause alterations in the measured instrumental response for one or more target analyte instrument responses. These interferences result from the presence of compounds that in the final extract that either suppress ionization or upset the lock mass correction of the instrument. For more detail on these types of interferences, see "Hybrid HRGC/MS/MS Method for the Characterization of Tetrachlorinated Dibenzo-p-dioxins in Environmental Samples." Y. Tondeur, W.J. Niederhut, S.R. Missler, and J.E. Campana, Mass Spectrom. 14, 449-456 (1987).

5.0 Sample Collection, Preservation and Storage

5.1 Tissue is stored at -20 ° C; all other samples are stored at 4 ° C, extracted within 30 days and completely analyzed within 45 days of extraction. Note that for PCDD/F analysis, exceedance of the stated extraction and analysis time frames does not necessarily indicate inferior data quality. Stability of these compounds in the received matrix, if in question and deemed critical to generation of valid results, should ideally be assessed through stability studies conducted over the time period necessary to establish valid data. SGS-AP freezes non-tissue samples only upon client request and strongly recommends against the practice.

5.2 Minimize use of solvents as much as possible.

5.3 Waste is disposed of in appropriate waste containers in accordance with the local, state and federal regulations. No sample or reagent is ever disposed of down a drain or in the trash. See document MI_278 for waste disposal and pollution prevention.

6.0 Equipment and Supplies




6.1 Autosampler

6.2 DEC Alpha Station with Opus Quan Data System or PC with MassLynx

6.3 Neslab Water Coolers or equivalent

6.4 Waters/Micromass Autospec Magnetic Sector High-Resolution Mass Spectrometer

6.5 Pipet, disposable, serological, 10mL

6.6 Amber glass bottles, 1 L (Teflon-lined screw cap)

6.7 2L separatory funnels

6.8 Teflon boiling chips

6.9 Glass chromatographic columns

6.10 N-Evaporator or equivalent

6.11 Conical autosampler vials, ~2mL

6.12 Glass fiber filters

6.13 Glass wool plugs

6.14 Glass Funnels

6.15 Dean-Stark Trap, condenser and flask.

6.16 Rotary Evaporator or Vacuum Centrifuge

6.17 Round bottom flasks, 500mL

6.18 Top-Loader Balance

6.19 Oven

6.20 DB-5MS (Agilent J&W GC Column p/n 122-5562 or equivanent)

6.21 Hydromatrix™

6.22 15mL test tubes

6.23 Meat grinder (Hobart or equivalent)

6.24 Tissuemizer

6.25 Spatula

6.26 Glass beakers

6.27 Glass thimble

7.0 Standards and Reagents

7.1 Reagents

7.1.1 Sulfuric acid, concentrated.




7.1.2 Silica gel. Highest purity grade.

7.1.3 Celite 545.

7.1.4 Water, distilled.

7.1.5 Florisil.

7.1.6 Prepurified nitrogen gas.

7.1.7 Anhydrous sodium sulfate.

7.1.8 Sodium Hydroxide. Highest available purity.

7.2 Solvents

7.2.1 Methylene chloride. Highest available purity.

7.2.2 Hexane. Highest available purity.

7.2.3 Tetradecane. Highest available purity.

7.2.4 Acetone. Highest available purity.

7.2.5 Methanol. Highest available purity.

7.2.6 Toluene, Highest available purity.

7.3 Standards

7.3.1 Analytical standards, in toluene, (Cambridge Isotope Laboratory, Wooburn, MA, Wellington Laboratories, or other qualified vendor). See Table 4 for details on composition of standard solutions.

8.0 Calibration

8.1 Initial HRMS Calibration (ICAL)

8.1.1 The ICAL establishes reference relative response factors for continuing calibration and, secondarily, demonstrates linear response over a given detector response range at a single point in time. The components of the ICAL solutions are summarized in Table 4.

8.1.1.1 Supply the reference compound -- e.g., perfluorokerosene (PFK) -- to the HRMS system. The reference compound provides the required lock masses and is used for tuning the mass spectrometer.

8.1.1.2 The lock-mass ion from the reference compound depends on the masses of the ions monitored within each m/z descriptor. Monitor each m/z descriptor in succession as a function of GC retention time to ensure that all PCDD/PCDFs are detected.




8.1.1.3 Monitor an appropriate QC Check Ion for each mass descriptor and use it to monitor system performance throughout the respective retention time windows. Evaluate specific situations where “deflections” are detected on a case-by-case basis with emphasis on the impact on the reliability of the analyte data.

8.1.1.4 Inject all initial calibration solutions in a single analytical sequence, along with an opening and closing solvent blank spiked (SBS). Due to special customer requirements, you may inject a truncated ICAL (e.g., dropping the highest and lowest calibration point) in addition to the full ICAL. In either case, proceed with processing and reporting of the ICAL in the same manner. The same acceptance criteria will apply regardless of the number of injections.

8.2 Continuing Calibration

8.2.1 Continuing calibration demonstrates the stability of the HRMS system and, more importantly – with the BCS3 system – provides a means to correct systematic errors.

8.2.2 Unless the client specified an OPR, follow the BCS3 requirements.

8.2.2.1 Analyze the BCS3 at the beginning and at the end of a (not to exceed 12 h) analytical sequence during which samples are analyzed.

8.2.2.2 Using the same operating conditions used for the ICAL, inject 1 µL of the batch control spike (BCS3) to demonstrate adequate GC resolution and sensitivity, response factor reproducibility, establish the PCDD/PCDF retention time windows and isomer-specificities, and to validate the standards used and the spiking technique.

8.2.2.3 For air samples returning with multiple trap preparations (in other words, multiple spiking events are involved) for a given project, select one of the BCS3s (i.e., the one representing the largest number of traps) and use it for continuing calibration. However, use the ICAL RRFs to process all the samples associated with such a project if you determine that the data reliability is not adversely affected. The laboratory will make every effort to communicate to the stack-sampling firm the significance of the trap preparation in




light of the QA/QC.

8.2.3 When the client requires traditional EPA QC requirements (i.e., OPR), follow the procedures below.

8.2.3.1 Inject a mid-range standard from the initial calibration curve (CS3) at the beginning of every (not to exceed 12h) analytical sequence.

9.0 Sample Preparation

9.1 Pre-Extraction Conditioning

9.1.1 Sample Examination

9.1.1.1 When first opening a sample, examine the sample for any qualities that may affect reliable analysis. A strong chemical odor may indicate that a sample will contain particularly high concentrations of PCDD/Fs or laboratory contaminants. It may also indicate potential interference problems, such as high fuel content. Note any large pebbles, rocks, twigs, shells and other matter that interferes with representative subsampling. For aqueous samples, note color and solids content. Record any observations you make that you believe may affect the analysis.

9.1.1.2 If you have reason to suspect that a sample may contain unusually high concentrations of PCDD/Fs, make note of this fact in the documentation accompanying the sample and consult the Technical Director for instructions on how to proceed.

9.1.2 Determine the percent solids on all soil, sediment, and aqueous samples the day before extraction. Perform the percent solids determination in triplicate if the sample amount submitted permits doing so. Percents solids determinations are not necessary for air matrices. For other matrices, use your best scientific judgment to determine if a percent solids determination is necessary. Consult with the Technical Director as necessary.

9.1.2.1 Weigh out an appropriate mass or volume of sample. For soils and sediments, target approximately 2-3g. For aqueous samples, target 10 mL.

9.1.2.2 Dry the sample overnight in a 110 °C oven and re-weigh the remaining residue.

9.1.2.3 Determine percent solids using the calculations shown




section 11.3.

9.1.3 Drying and grinding of soil and sediment samples is expressly forbidden. Only freeze soils and sediments upon client instruction to do so. Involve the Technical Director or SGS-SEAL if the customer requests drying or grinding of soils or sediments.

9.2 Extraction

9.2.1 All samples are extracted on a wet weight basis. For solid samples where a percent solids determination has been made, adjust the sample amount for extraction to yield a dry weight equivalent (DWE) based on the percent solids that will fit the purpose of the analysis. Default to 10g DWE when you have no fitness-for-purpose information and cannot obtain any.

9.2.2 Soils & Sediments

9.2.2.1 Weigh the samples in clean glass beakers. Perform the following steps for each sample.

9.2.2.2 Add several mL of 5% H2SO4.

9.2.2.3 Dilute the ES spike in acetone.

9.2.2.4 Add the spike to the beaker.

9.2.2.5 Stir the beaker’s contents to thoroughly mix the spike with the sample.

9.2.2.6 Add sufficient Hydromatrix™ to disperse the sample across the surface of the inert material. This step will have the secondary effect of absorbing the liquid present in the sample.

9.2.2.7 Transfer the beaker contents to a thimble for Soxhlet Dean-Stark (SDS) extraction. Rinse the beaker twice with extraction solvent (toluene, unless otherwise specified as a project-specific instruction) and transfer the rinses to the thimble.

9.2.2.8 Extract the samples by SDS using toluene (or other solvent specified as a project special instruction) for a length of time sufficient to effect full recovery of extraction standards. By default, the extraction time is 16 hours, although shorter times may be used if they have been previously demonstrated through validation to provide adequate performance.

9.2.2.9 The solvent must cycle through the system five time each hour.




9.2.3 Aqueous or Effluent Samples (nominal volume of 1 liter)

9.2.3.1 If the sample contains >1.0 percent solids, treat the sample as a solid (i.e., isolate a filter cake and extract it as a solid).

9.2.3.2 Aliquot an amount of the aqueous sample that will yield a 10-g dry weight equivalent.

9.2.3.3 Filter this aliquot thru a 1.0 mm filter using the filtering setup.

9.2.3.4 Transfer the filter and cake inside the thimble.

9.2.3.5 Spike the appropriate standards using acetone as a carrier.

9.2.3.6 Add ~10 g Hydromatrix™ and mix the solids from the filter with the Hydromatrix™.

9.2.3.7 If solids are still present in the water phase, filter the water through 0.7 µm and 0.45 µm filters, removing and placing them in the same glass thimble. Extract with SDS using toluene, as with other solid samples.

9.2.3.8 If the solid load is less than 1.0 percent, the sample may be extracted using the TALEX, CLLE, or the SPE-C approaches.

9.2.3.9 The details for these procedures are given in the relevant SOPs. The decision to follow one particular approach over the others is dictated by the client’s requirements, the type of samples undergoing analysis, and the expertise of the laboratory staff.

9.2.3.10 All paper mill effluent samples should be filtered and the isolated solids extracted by SPE and SDS.

9.2.4 Oil Samples

9.2.4.1 Dilute the sample with ~500 mL of hexane.

9.2.4.2 Perform an acid partition and cleanup the sample as detailed in section 9.3.

9.2.5 Air Samples

9.2.5.1 Collect all the sample components.

Note: Usually, samples being prepared by EPA Method 23 are comprised of a filter, an XAD-2 resin trap, an acetone rinse for both the front and back halves, a toluene rinse for both the front and back halves, and sometimes a methylene chloride rinse for




both the front and back halves. Method 0023A samples contain the same components, but the method requires the filter and front half rinses to be extracted and analyzed as a separate sample from the XAD-2 resin trap and back half rinses. For ambient air samples (Method TO-9A), a PUF and filter combination is typical.

9.2.5.2 For a Method 23 analysis (only), photograph the filter and XAD-2 resin trap together for documentation purposes prior to combination. Line up components next to the corresponding thimbles and take a group photograph (i.e., of all samples and thimbles), and then take a photograph of each sample’s components immediately before unpacking and combining in the thimble. Follow the steps below as written. If the analysis is by Method 0023A, follow the steps below except for combining the resin and filter – these components (with their respective rinses) constitute two separate samples.

9.2.5.3 Combine resin and filter in a glass SDS thimble.

9.2.5.4 If FS was not used in the field, spike the acetone rinse(s) with FS.

9.2.5.5 If FS was used in the field, spike the acetone rinse(s) with TS.

9.2.5.6 Concentrate the rinses to less than 5 mL using a rotovap or by other appropriate means (i.e., fit for the purpose of the analysis).

9.2.5.7 Add the concentrated solvent rinses to the resin/filter in the thimble.

9.2.5.8 Into the thimble, spike the appropriate amount, as indicated in the project documentation, of PCDD/F Extraction Standards.

9.2.5.9 Use the glasswool that was packed into the XAD-2 trap to collect any stray XAD-2 particles and solvent and place the glasswool plug in the thimble on top of the resin.

9.2.5.10 Extract using SDS with toluene.

9.2.5.11 Concentrate the residue to ~6 mL, solvent exchange to hexane, quantitatively transfer the extract to a 15 mL test tube, and bring to a volume of 12 mL in hexane.

9.2.5.12 Split the extract per the project instructions, and spike




the portion destined for PCDD/F analysis with AS, per project instructions.

9.2.5.13 Store the remaining extract at room temperature in the test tube with a Teflon™-lined screw top.

9.2.6 Tissue Samples: Tissue samples received as whole material are ground and homogenized. Others can be used as received.

9.2.6.1 Grind fish (or other tissue received as whole tissues) using the Hobart meat grinder. Collect the ground tissue directly on cleaned aluminum foil.

9.2.6.2 Using Tissuemizer blend meat into a creamy paste consistency.

9.2.6.3 Using a spatula, mix well.

9.2.6.4 Aliquot 25 g for analysis. Use the requested or necessary aliquot size if a larger aliquot is requested by the customer or necessary to meet customer data quality objectives.

9.2.6.5 Place two more aliquots of the ground tissue inside two 4-oz jars, label the jars and store in a freezer.

9.2.6.6 Fortify the sample with ES directly using acetone as the carrier solvent.

9.2.6.7 Mix the sample with 20-25 g of Hydromatrix™, stir until a dry, free-flowing consistency is achieved, adding more Hydromatrix™ as needed. Be careful not to exceed an amount that can fit into the thimble for the sample size used.

9.2.6.8 Soxhlet Dean-Stark using hexane for 16h (or other duration previously demonstrated to be adequate to meet fitness-for-purpose requirements).

9.2.6.9 For samples exceeding 50g, implement the following additional steps:

9.2.6.9.1 AFTER spiking the sample AND thoroughly integrating the spike, split the sample across enough thimbles such that no thimble holds more than 50g of tissue.

9.2.6.9.2 Use a mixture of 30% acetone and 70% toluene rather than hexane

9.2.6.9.3 Extend the extraction time from 16h to 60h.

9.2.6.9.4 Combine the extracts after the extraction is




complete but before concentrating the extracts for cleanup.

9.2.6.10 In case the tissue contains sediment particles, it is recommended to follow the hexane(-acetone) extraction with a toluene-acetone extraction.

9.2.7 If necessary, and not already done, solvent-exchange any extracts not in hexane to hexane.

9.2.8 Combine (solvent-exchanged) extracts, if necessary

9.2.9 Concentrate the extract(s).

9.2.10 Add the CS.

9.2.11 Cleanup the extract. (see below).

9.3 CLEANUPS

9.3.1 If not already done, add the Cleanup Standards and/or Alternate Standards.

9.3.2 Optionally, acid/base partitioning ("dirty" samples only)

9.3.3 Acid/Base Silica Column (see SOP DC_365)

9.3.4 Florisil Column (see SOP DC_365)

9.3.5 Optionally, use a carbon column if significant residue remains in the extract, such as oil. (see SOP DC_365)

9.4 TRANSFER

9.4.1 Concentrate the extract resulting from the clean-up process to volume of approximately 1.5 mL.

9.4.1.1 Concentrate using a rotovap or in a vacuum oven

9.4.1.2 Optionally, collect the cleaned-up extract in a glass vessel containing 20 µL tetradecane and allow the extract to concentrate to 20 µL. You must reconstitute the sample within 24 hours of beginning the concentration if you concentrate the extract in this manner.

9.4.2 Spike the extract with 2ng of injection standards (JS).

9.4.3 Concentrate the extract to a volume of 20 µL using nitrogen blow down. Use caution to avoid over-concentrating the extract at this stage, as you may cause losses of tetra-PCDD/Fs if the extract is over-concentrated.

10.0 Analytical Procedure




10.1 Establish the necessary operating conditions. The following GC operating conditions are for guidance, and adjustments may be required.

10.1.1 The reference compound provides the required lock masses. Use it for tuning the mass spectrometer (see instrument manuals for details on tuning).

10.1.2 The instrument is tuned to the minimum required resolving power of 10,000 (10 percent valley or 5 percent crossing over) at an appropriate reference signal close to the mass-to-charge ratios of interest. Document the resolving power using the mass peak profile display. Obtain the latter is at the beginning and the end of each analytical sequence of 12 hours or less.

10.1.3 The QC Check Ion from the reference compound is dependent on the masses of the ions monitored within each descriptor. Each descriptor will be monitored in succession as a function of GC retention time to ensure that all PCDD/PCDFs are detected.

10.1.4 Set up the analytical run following this sequential injection pattern: CPSM, BCS3 (or CS3, if an OPR was required), Solvent Blank Spiked (SBS), Method Blank, OPR (if required), Samples, BCS3 (if using BCS3 system).

10.1.5 If the analytical sequence will exceed 12 hours, ensure a BCS3 (or CS3, if an OPR was required) injection occurs before 12 hours from the sequence start expires.

10.1.6 Use the column performance standard mix (CPSM) to verify retention time windows and the retention times of individual congeners and co-eluting clusters of congeners.

10.1.7 Whenever practical, determine the acceptability of the continuing calibration sample (BCS3 or CS3) before injecting the remainder of the sequence. Do the same for the SBS and LMB, if practical.

10.1.8 Qualitative Determination

10.1.8.1 Identify a chromatographic peak as a PCDD or PCDF (either a native or a labeled compound) using the following criteria:

10.1.8.1.1 The signals for the two exact m/zs being monitored must be present and must maximize within ±2 seconds of one another.

10.1.8.1.2 The signal-to-noise ratio (S/N) of each of the two native exact m/zs must be greater than or equal to 2.5:1 for a sample extract, and greater than or equal to 10:1 for a calibration




standard. The S/N for labeled compounds must always be at least 10:1. If an interference diminishes the signal for one native m/z to the point that it is not detected and all other criteria are met, the peak may be considered a PCDD/F and quantified using the one m/z available. The chromatographic peak will be annotated to indicate that this choice was made.

10.1.8.1.3 The ratio of ion abundances for the two m/zs must lie within the limits established for the homologous series. If the ratio for a native compounds lies outside these limits while all other criteria are met, the peak is still considered to be a PCDD/F but will be reported as an Estimated Maximum Possible Concentration (EMPC).

10.1.8.1.4 The absolute retention times for all congeners must be within 2 seconds of those obtained in the CPSM. If an interference causes retention time shifting, use the overlay and/or predicted retention time features of UTP to determine the shifted retention times of the individual congeners.

10.1.8.1.5 The absolute retention times for native (only) 2,3,7,8-substituted congeners must be within 0 to +2 seconds of the isotopically-labeled standard.

Note: These limits are derived from over 500 data points obtained from MM1 and MM3 over a 17-week period worth of BCS3 and CS3/CS0 injections. The 99% confidence intervals suggested 0 to 1 sec.; we decided to use 2 sec. to cover all analytes. These limits are more representative of our level of performance. Weak peaks will continue to be treated as before with the call made by the chemist.

Observe the following information when identifying the following potential co-eluters in PCDD/F processing:




23478/12489-PeCDF

234678/123689-HxCDF

123789/123489-HxCDF

10.1.8.1.5.1 Although our processing software allows each of these to be treated as peak pairs, you should not use this option.

10.1.8.1.5.2 23478/12489-PeCDF: This pair occasionally shows a point of inflection that would allow a vertical to be dropped distinguishing the two separate peaks (with the appropriate PR flag), but, in most cases, you should treat this peak as the specific only (i.e. just 23478-PeCDF) so that it does contribute to the TEQ. Use a PR flag on the chromatogram even if the peak is designated as 23478-PeCDF, but the co-eluter is clearly also present.

10.1.8.1.5.3 234678/123689-HxCDF: This pair is never resolved (on a *-5ms column), so you should always quantify it as the specific (i.e. 234678-HxCDF).

10.1.8.1.5.4 123789/123489-HxCDF (the “last” HxCDF eluter): This pair presents a more difficult issue since, in samples, the peak is often actually due to the non-specific 123489-HxCDF. We can determine that the peak represents the non-specific from the notable shift in retention time relative to the CPSM/BCS3 (or indeed, the RRT relative to the ES-123789-HxCDF). Therefore, the analyst must decide which peak it is assigned as and whether a PR flag should be used, but it is particularly important to be consistent across all data. The method actually allows the specific to be -1/+3 seconds relative to the ES, and this can also be used as guidance.

10.1.9 Quantitative Determination

10.1.9.1 For peaks that meet the criteria listed above, quantitate the PCDD and PCDF peaks from the mean BCS3 (or




CS3) RRFs relative to the appropriate Extraction Standard established, unless they fail to meet acceptance criteria (see acceptance criteria below) and the appropriate corrective action involves using other RRFs.

10.1.9.2 For peaks that meet all the criteria above except the ratio of ion abundances, report the peak as an EMPC.

10.1.9.3 For circumstances in which an EMPC cannot be reported, we report any out-of-ratio criterion peak eluting meeting all other identification criteria as a non-detect using the peak concentration as the detection limit.

10.1.9.4 Report results in picogram per gram, picogram per liter or picogram per sample.

10.1.9.5 A correction for the contribution at m/z 322 from the CS ( 37Cl4-2,3,7,8-TCDD) is applied automatically in our calculations. This correction subtracts 0.06% of the CS peak area from the m/z 322 ion for 2,3,7,8-TCDD (only).

10.1.9.6 Calculate a sample- and analyte-specific estimated detection limit for each congener.

10.1.9.7 Report the data.

11.0 Details of Calculations

11.1 Calculate the concentrations for the PCDD or PCDF compounds using the formula:

Where

= Concentration of unlabeled PCDD/PCDF congeners (or group of coeluting isomers within an homologous series) in pg/g.

= Sum of the integrated ion abundances of the quantitation ions for unlabeled PCDD/PCDFs.




= Quantity, in pg, of the Extraction Standard added to the sample before extraction.

Sum of the integrated ion abundances of the quantitation ion ( See Table 6 of EPA Method 8290) for the labeled Extraction Standards.

= Weight of the sample (solid or liquid).

= Calculated relative response factor for the analyte.

11.2 The detection limits for each absent PCDD/PCDF can be calculated using the following formula:

= Estimated detection limit for PCDD/PCDFs.

= Noise height (peak to peak).

= Peak height of the Extraction Standard.

= Quantity, in pg, of the Extraction Standard added to the sample before extraction.

= Weight of the sample (solid or liquid).

= Calculated relative response factor for the analyte.

11.3 Percent solids:

)()(%−+−+

=boatofweightboatofweightsampleofweightboatofweightboatofweightresidueofweightsolids

11.4 Lipid content:




100)(

)(Re% ×=

gSampleWetofWeightgsidueofWeight

Lipids

12.0 Quality Control Requirements

12.1 A method detection limit (MDL) study determination is required by this method. This study need only be performed once unless changes are made to an instrument that would be expected to alter its detection limits.

12.2 Initial Demonstration of Proficiency – New staff are required to demonstrate initial proficiency by generating data of acceptable accuracy and precision for target analytes in a clean matrix.

12.3 (Laboratory) Method Blank (LMB): The method blank enables monitoring of lab processes to ensure that the laboratory introduces no target analytes or specific interferences during routine operations and to quantify them should they be found. The method blank further assesses the performance of the chosen extraction method on a quality control matrix using the sample personnel, equipment, reagents, etc. that were used for the samples it accompanies. The method blank may consist of sand, Hydromatrix™, Na2SO4, or distilled water known to be free of native analytes. The laboratory prepares and analyzes the method blank using the same procedures followed for the rest of the analytical batch. The results of the method blank are considered in the performance-based criteria analysis, permitting assessment of whether any laboratory background observed causes data to be unfit for its intended purpose.

12.3.1 Include a method blank with every analytical batch or 20 field samples (whichever is less) per matrix type.

12.3.2 The extraction standard recoveries of the method blank indicate extraction efficiency performance for a quality control sample (thus eliminating the need for that portion of an OPR).

12.3.3 Use Table 1 to select the appropriate quality control matrix for the method blank: Select the weight or volume of method blank matrix to approximately match that used in the samples, leaning towards the largest amount used in any sample.

12.3.4 For air samples, prepare the LMB at the same time as the air traps/PUFs and spike them with the same amount of sampling standard that is spiked into the traps/PUFs. When preparing a large number of traps/PUFs (e.g., 10+), consider preparing more than one LMB. Refrigerate these LMBs until the traps/PUFs




return to the laboratory for analysis.

12.4 On-Going Precision & Accuracy (OPR = Laboratory Control Spike): The OPR assesses extraction efficiency performance for a quality control sample and, ostensibly, native analyte recovery. However, since, for PCDD/F analysis, all of the compounds required to be present in an OPR are analyzed using true isotope dilution, the OPR does not provide native analyte recovery information. Worse, the OPR results can be confounded by spiking errors it fails to detect. Thus, the OPR fails to fulfill its purpose. For these reasons, SGS-AP does not recommend its use, but SGS-AP will use it upon client request and/or program requirements.

12.4.1 Use a known amount of native PCDD/Fs for spiking.

12.4.2 When requested, analyze an OPR with every analytical batch or 20 field samples (whichever is less) per matrix type.

12.4.3 Apply the same guidelines that apply to the method blank for choosing the matrix -- and the weight or volume thereof -- to make the same choices for the OPR.

12.4.4 Do not prepare a batch control spike (BCS3, vide infra sec. 12.5) when an OPR is selected by the customer

12.5 Matrix Spike (MS/MSD): The purpose of the matrix spike and matrix spike duplicate samples is to assess the ability of the analysis to accurately quantify the target analytes in the presence of matrix interferences. However, as with the OPR, since all specific target analytes are analyzed using true isotope dilution, this sample, prepared and analyzed in the usual way, fails to fulfill its purpose. Rather, each sample, spiked with the carbon-labeled analogs of all 17 target analytes, serves as its own matrix spike. Therefore, SGS-AP not only recommends that MS/MSDs not be used, but SGS-AP considers MS/MSD samples to be billable samples.

12.5.1 Use known amount of native PCDD/Fs for spiking.

12.5.2 Unless specified by the customer, choose the sample to use for an MS(/MSD) at random from those samples being analyzed in the batch.

12.5.3 Analyze MS/MSDs at a frequency specified by the customer.

12.5.4 Apart from spiking native PCDD/Fs into the sample, analyze the sample in the same manner as all other samples.

12.6 Laboratory Duplicate Samples: Laboratory duplicate samples are two separate subsamples taken from a well-homogenized sample. We recommend that customers provide these to the laboratory as blind duplicates – each with its own identity – to prevent laboratory bias.

12.6.1 Analyze duplicate samples independently to assess laboratory




precision.

12.6.2 Analyze the duplicate samples in the same manner as all other samples in the batch.

12.7 Batch Control Spike (BCS3): The BCS3, in tandem with the LMB, fulfills the purpose of the OPR. The BCS3 also acts as a physical record of the spiking of the batch, as a continuing calibrant, and as a chromatographic resolution demonstration sample. Many benefits of the BCS3 exist but are beyond the scope of this SOP.

12.7.1 Prepare a batch control spike as the default option. Do not prepare a batch control spike if the customer has specified the use of an OPR.

12.7.2 Prepare the BCS3 at the same concentration as the CS3 of the ICAL

12.7.3 Prepare the BCS3 —inside a 4oz. glass jar containing the appropriate volume (20 µL if no split is involved) of tetradecane —in stages at the same time as the batch of field samples; i.e., at each phase involving the addition of the ES, CS or AS, JS to the samples. See 12.5.6 for details of how to handle an analysis with a split.

12.7.4 For air samples, initiate the BCS3 at the same time as the preparation of the air sampling modules before the sampling session, using the same amount of Sampling Standards. Keep the jar at room temperature and away from light. When preparing a large number of air traps (e.g., 10+), consider preparing multiple BCS3s.

12.7.5 Prepare one BCS3 per batch of 20 samples or less -- regardless of the matrix type, going through the same spiking scheme with the same spiking solutions, same analyst, same delivery system, and at the same time as the field samples. The laboratory must ensure preparation of sufficient BCS3s to provide front- and back-end calibration verifications for all the samples as well as re-injections when the latter are necessary.

12.7.6 For air samples where a split factor is involved -- i.e., the sample extract is split and a portion is archived as backup -- do not subject the BCS3 to an actual physical division. Simulate the division by adding an appropriate volume of the same solvent as for the ICAL and the samples (e.g., if the split factor is 2, then, the BCS3 needs to be diluted two fold before analysis to allow the analytes to be at the same concentration as for the ICAL CS3).




12.7.7 Ensure that the following condition is met during the preparation of the BCS3: The lab uses either two different spike delivery techniques or two different pieces of equipment (using the same delivery technique). Observing this stricture ensures that the lab detects any systematic error in the volume delivered by any one piece of equipment.

12.8 Data outside of QC limits referenced in section 13 may be addressed by one or more of the following options:

12.8.1 Re-preparation and re-analysis of sample

12.8.2 Client notification

12.8.3 Discussion and qualification of data by case narrative

12.8.4 Re-sampling and reanalysis (client decision)

12.9 Data outside of QC limits referenced in section 13 may be reported if directed by the client. It must be qualified by a case narrative detailing the QC problems with advice on the usability of the data.

13.0 Data Review and Reporting Requirements

13.1 This method is performance-based (see Federal Register: October 6, 1997 (Volume 62, Number 193), Page 52098-52100). Thus, the acceptance criteria (i.e., target objective guidelines) are ideally determined by the project needs through a process of negotiation with the client and any other ultimate end-user of the data. These criteria should ensure that the performance of the method is adequate for the purposes for which it is intended. The criteria will likely not resemble those used in the past. If negotiations with the client yield no criteria, statistically determined criteria or the default criteria below may be used.

13.2 A failure to meet all criteria does not necessarily result in a particular course of action (see section 13.11 for details). Data assessment includes a thinking process that assesses the impact of a particular analytical problem and develops the most appropriate response, in the context of the data user’s needs (when known). Data that do not meet all criteria for all target analytes may still be reported with relevant qualifiers and notes in the case narrative, if appropriate.

13.3 GC/MS Analysis Performance Acceptance Criteria

13.3.1 The mass resolving power (10% valley or 5% crossing over definition) must be at least 10,000, over the full mass range used. This criterion applies to both the beginning and ending mass resolving power checks.

13.3.2 The chromatographic peaks for 2,3,7,8-TCDD and the closest




eluting isomer must resolve with a valley of 25% or less, although valleys above 10% indicate a need to perform column maintenance.

13.3.3 (BCS3 system only) Should the valley exceed 25%, the analyst may accept the data if the RRFs for 2,3,7,8-TCDD meet acceptance criteria.

13.3.4 The chromatographic peaks for 2,3,7,8-TCDF and the closest eluting isomer must resolve with a valley of 40% or less.

13.3.5 (BCS3 system only) Should the valley exceed 40%, the analyst may accept the data if the RRFs for 2,3,7,8-TCDF meet acceptance criteria.

13.4 ICAL Acceptance Criteria

13.4.1 The percent RSD for the mean response factors must be ≤10% for the native standards and ≤20% for extraction standards.

13.4.2 In addition, the RRFs must come to ±20% of historical averages.

13.4.3 All S/N must be at least 10:1.

13.4.4 Ion-abundance ratios must come to ±15% of the theoretical ratios.

13.5 Batch Control Spike (BCS3) Acceptance Criteria

13.5.1 In order to use the front- and back-end BCS3s averaged RRFs to process the samples, the individual front- and back-end chromatographic peaks need to meet a number of requirements (independent verification, RPD, and PD or bias).

13.5.1.1 The first and last PCDD/F eluters are verified to be within the eight homologue retention time windows.

13.5.1.2 The signal to noise ratio (S/N) exceeds 10:1 for all compounds spiked and monitored,

13.5.1.3 The ion abundance ratio measurements are within ± 15 percent of the theoretical ratio,

13.5.1.4 The first and last eluters in each homolog group are verified to be within the eight homologue retention time windows.

13.5.2 The RRFs Percent Differences (PDs) relative to the ICAL should remain within established criteria. These criteria may be established statistically or on a per-project basis, according to the data user's needs. If no criteria are established through negotiation with the client or statistically, the following criteria are recommended.




13.5.2.1 ±20% for the unlabeled compounds

13.5.2.2 ±30% for the labeled ES compounds

13.5.2.3 ±20% for air’s labeled SS, and

13.5.2.4 ±30% for non air’s labeled CS compounds

13.5.2.5 Other requirements are shown in Tables 3-6.

13.5.3 The RRFs Relative Percent Differences (RPDs) between the front and back BCS3 injections should also meet the following criteria.

13.5.3.1 ±10% for the unlabeled compounds

13.5.3.2 ±20% for the labeled compounds

13.6 CS3 Acceptance Criteria

13.6.1 The signal to noise ratio (S/N) exceeds 10:1 for all compounds spiked and monitored,

13.6.2 The ion abundance ratio measurements are within ± 15 percent of the theoretical ratio for all the compounds spiked and monitored.

13.6.3 The RRFs must be within the criteria established in Method 1613.

13.7 Method Blank (LMB) Acceptance Criteria

13.7.1 Ideally, the LMB analysis will return only non-detects. Occasionally, an amount of OCDD just above the detection limit may be found.

13.7.2 If you detect any PCDD/Fs in the LMB, use the performance-based criteria analysis (PBCA) to determine whether the LMB contents are tolerable or not.

13.7.3 If the client specified the use of an OPR (thus disabling PBCA), any hits in the LMB should be below the reporting limit or 10% of the sample contents, whichever is greater.

13.8 OPR Acceptance Criteria

13.8.1 The acceptance criteria for the OPR depend on the scope of target analytes, as detailed below

13.8.2 If only TCDD and/or TCDF is(are) analyzed, the acceptance criteria are given in Table 6A.

13.8.3 The acceptance criteria are given in Table 6 if the analysis includes the full list of 17 2,3,7,8-PCDD/F containing analytes.

13.9 MS/MSD Acceptance Criteria




13.9.1 The laboratory considers MS/MSDs in which the recovery of the spike is within the range 80-120% acceptable, provided the matrix used does not contain PCDD/Fs at levels impairing the measurements.

13.9.2 In assessing duplicate matrix spikes, the laboratory considers the duplicates acceptable if the results plus the expanded measurement uncertainties for each congener overlap, provided the matrix used does not contain PCDD/Fs at levels impairing the measurements and is not so inhomogenous as to make representative subsampling practically impossible.

13.9.3 If a customer requires relative percent difference criteria for acceptance of duplicate matrix spikes, RPDs between MS/MSD samples should be ≤30%, provided the matrix used does not contain PCDD/Fs at levels impairing the measurements and is not so inhomogenous as to make representative subsampling practically impossible.

13.10 Laboratory Duplicate Acceptance Criteria

13.10.1 In assessing duplicates, the laboratory considers the duplicates acceptable if the results plus the expanded measurement uncertainties for each congener overlap, provided the matrix used does not contain PCDD/Fs at levels impairing the measurements and is not so inhomogenous as to make representative subsampling practically impossible,

13.10.2 If a customer requires relative percent difference criteria for acceptance of duplicate samples, RPDs between the duplicate sample should be ±30 percent, provided the matrix used does not contain PCDD/Fs at levels impairing the measurements and is not so inhomogenous as to make representative subsampling practically impossible.

13.11 CORRECTIVE ACTIONS

13.11.1 Corrective actions are taken whenever needed, regardless of acceptance criteria. It is not always the case that data meeting all acceptance criteria should be accepted nor that data not meeting all criteria should be rejected. The end user data quality needs guide the determination of whether and what corrective action may be required in any given situation.

13.11.2 When the BCS3 fails to meet acceptance criteria, one must endeavor to understand the information the failure brings to the surface. Consider the following in the search for that understanding.




13.11.2.1 First, distinguish between a critical error and a minor one. A critical error means erroneous data resulting from a seriously flawed spiking technique (e.g., wrong amount of NS added under the BCS3 system) while other minor errors can provide useful information or feedback on the measurement step (e.g., instrumentation variation). The interpretation of the information obtained from the analysis of the BCS3 is best handled when done contextually. This analytical protocol does not claim that it offers a comprehensive analysis but merely puts forward guidelines to help the analyst in assessing the quality and reliability of the data.

13.11.2.2 A failure on the “PD” requirements may be indicative of an instrumentation difficulty or spiking error. The latter can be of Level PD-1 (i.e., at the standard solution level) or Level PD-2 (i.e., at the spiking operation level). A third Level PD-3 is associated with instrumentation. An error at the NS standard solution level (Level PD-1) constitutes, under the BCS3 system, the most serious failure and requires that a new set of standard solutions be prepared and independently validated before repeating the sample extraction and analysis, if called for by the decision flow chart (“Thinking Method”). Note that the decision flow charts make a distinction between short- and long-term actions. When appropriate, a new initial calibration may be required before analyzing the BCS3 and the samples. However, if the error is a Level PD-2 error, a re-extraction and analysis is the most suitable action after correcting the flawed spiking technique. As customarily done, a new BCS3 is prepared with a Level PD-2 error. Here again, the decision flow charts should be used since the text herein is only for illustration purposes. Distinction between Levels PD-1 and PD-2 can be accomplished contextually by examination of the initial independent validation study and control charts (showing for instance a trend suggesting a degradation of the ES solution), behavior over time charts, and/or using the matrices shown in Figure 3. The Level PD-3 error is associated with instrumentation when an out-of-calibration situation is present or a temporary or localized instrumentation variation is operative. Depending on the severity of the Level PD-3 error, a




new calibration (either ICAL or rerunning the BCS3 and all the affected samples) following a new “tuning” of the instrumentation may be required (see Decision Flow Charts for “Thinking Method”). Note that coincident Level PD-2 and Level PD-3 errors may occur in the same BCS3 analysis, causing determination of the root cause of the failure to be complex. Analyzing the BCS3 back-to-back with a CS3 helps resolve the situation in these cases.

13.11.2.3 A failure on the “RPD” requirements may be indicative of instrumentation instability or inability to sustain the instrumentation’s performance over a 12-H period. Again, two levels are possible. Level RPD-1 is strictly associated with instrumentation difficulties that are unrelated to the samples under analysis. A re-analysis (i.e., re-injection) of the BCS3 and of the samples can be considered as a corrective action after correction of the source of the instrumentation’s shortfall (see Decision Flow Charts for “Thinking Method”). If however, the re-analysis of the BCS3 fails again, and there are indications that the spiking procedure is questionable (Level PD-1 or PD-2), the associated extraction batch may need to undergo re-extraction and analysis with the preparation of a new associated BCS3 as discussed above for the BCS3 PD deviations. A Level RPD-2 BCS3 failure may be found with the analysis of samples presenting special challenges (i.e., highly complex matrices that do not cleanup well under the various options offered by this protocol). Depending on the severity of the deviation, additional cleanup or other appropriate actions may be required before re-analysis of the samples and associated BCS3. If such action proves to be ineffective, the data should be qualified accordingly.

13.11.2.4 The BCS3 “PD Requirements” are summarized in Tables 5A-5D. In addition to the traditional RRFs, another set of “pseudo-RRFs” is computed from the BCS3 data to help with the validation of the ES and RRFs used to report the sample analytes. The pseudo-RRFs are used to further differentiate the various “A” to “C” types PD Requirements. Use Table 5C (non-air) or Table 5D (air) for departing-from-the-norm groups of analytes (e.g., all 17 Ax or the five SS show a




deviation similar in “sign” and “amplitude”). It is also recommended to examine the data contextually (e.g., using QC charts)

13.12 See Data review SOP (MI_141) for additional information.

14.0 Preventative Maintenance

15.0 Tables

Table 1 Sample Matrix LMB Matrix

XAD-2 resin M23 XAD-2 resin XAD-2 & Filter for

M0023A Filter for Front Half & XAD-2 for Back

Half PUF for TO-9A PUF

Aqueous DI water Soil/Sediment & Tissue Hydromatrix™

Chemical Sodium sulfate or other as deemed appropriate – document choice

Oil Hexane




Figure 1 Sample Preparation Flow Chart for Air Matrices




Figure 2 Sample Analysis Flow Chart (example given for air matrices)




Table 2

Column Flow Rate 1.3 mL/min Injector temperature:

Interface temperature:

Initial time:

Temperature program:

280 °C

280 °C

2 minutes at 190°C

190 to 220 ° C, at 5 °C/min

220 °C for 14.5 minutes

220 to 290 ° C, at 5 °C/min

290 °C for 1 minute

290 to 305 ° C, at 14 °C/min

305°C for 10 min




Figure 3 Decision Flow Charts for “Thinking Method”




Table 3 Theoretical Ion Abundance Ratios and Control Limits for PCDDs and PCDFs. (a) Used for 13C-HxCDF (IS) only. (b)Used for 13C-HpCDF (IS) only

Number of Theoretical Control Limits

Chlorine Atoms Ion Type Ratio Lower Upper

4

0.77 0.65 0.89

5

1.55 1.32 1.78

6

1.24 1.05 1.43

6a

0.51 0.43 0.59

7b

0.44 0.37 0.51

7

1.04 0.88 1.20

8

0.89 0.76 1.02




TABLE 4: PRIMARY HIGH-RESOLUTION CONCENTRATION CALIBRATION SOLUTIONS (Regular Initial Calibration)




Table 5A: Batch CS3 PD Requirements Based on Traditional RRFs

(“A” to “C”Types)

Type Analytes Requirement Failure Possible Causea,b

Failure Level

Suggested

Corrective Action A AX vs ES + 20% 1. Calibration

out

2. Spiking error

1. PD-3

2. PD-1/PD-2

3. New Calibration

4. New Standards/New Extraction

B ES vs JS + 30% non air

+ 30% for air c

1. Calibration out

2. Spiking error

5. PD-3

6. PD-1/PD-2

7. New Calibration

8. New Standards/New Extraction

C1 CS vs JS + 30% 1. Calibration

out

2. Spiking error

9. PD-3

10. PD-1/PD-2

11. New Calibration

12. Affects other Types

C2 SS vs ES

(air)

+ 20% 1. Calibration out

2. Spiking error

13. PD-3

14. PD-1/PD-2

15. New Calibration

16. New Standards/New Extraction/New Samplingd




a. Calibration out = usually when one (localized) or several/all analytes are affected; instrumental source. b. Spiking error = when all analytes are affected with the same “sign” and “amplitude”; must be considered contextually; i.e., using historical data or other information on the set of standards such as the “pseudo-RRFs”. Situations when selected analytes degrade are rare but should not be excluded from consideration. c. By design for air matrices, the amounts of ES and JS added during the preparation of the Batch CS3 are different. Thus, an additional error is introduced, which can deceive the analyst’s interpretation. In this case, the QC emphasis is shifted towards the “C2” type PD requirement.

Because of the nature of an “air” sample, there is no additional sample volume available to repeat the extraction. The laboratory is required to qualify the data by estimating and documenting accordingly the “error” associated with the reported measurements. If such documentation is not possible, and/or the information points toward a seriously flawed ES addition (as opposed to a spiking error associated with the SS), the data can be rejected and re-sampling efforts may be necessary. See the “Air Spiking Related Error Matrix” tables for an alternative approach whereby the Ax vs SS RRFs are used to determine the analyte’s concentrations (Table Insert 4).

Table 5B : Batch CS3 PD Requirements Based on Pseudo-RRFs

(“D” to “G”Types)

Type Analytes Requirement D Ax vs CS/SS +/- 25%

E1 Ax vs JS

(non air)

+/- 35%

E2 Ax vs JS

(air)

+/- 35%

F1 ES vs CS

(non air)

+/- 20%

G2 SS vs JS

(air)

+/- 50%




Table 5C:

“Non-Air” Spiking Related PD Errors

“PD Requirements Decision Matrix” – Normal Configuration (use BCS3 RRFs)

Ax ES CS JS Ax - Y Y Y ES - - Y Y CS - - - Y

“PD Requirements Decision Matrix” – Defective Ax Spiking = Fatal (If Ax spiking is shown to be in error, and if no additional or replacement sample is available, use ICAL RRFs if

CS3 is acceptable)

Ax ES CS JS Ax - N N N ES - - Y Y CS - - - Y

“PD Requirements Decision Matrix” – Defective JS Spiking (use BCS3 RRFs)

percent recovery measurements for CS & ES affected, not the analytes

Ax ES CS JS Ax - Y Y N ES - - Y N CS - - - N

“PD Requirements Decision Matrix” – Defective CS Spiking (use BCS3 RRFs)

percent recovery measurements for CS affected, not the analytes

Ax ES CS JS Ax - Y N Y ES - - N Y CS - - - N

“PD Requirements Decision Matrix” – Defective ES Spiking (use BSC3 RRFs; see Levels PD-1 or PD-2)

Ax ES CS JS Ax - N Y Y ES - - N N CS - - - Y




Table 5D:

“Air” Spiking Related PD Errors

“PD Requirements Decision Matrix” – Normal Configuration (use BCS3 RRFs)

Ax ES SS JS Ax - Y Y Y ES - - Y Y SS - - - Y

“PD Requirements Decision Matrix” – Defective Ax Spiking = Fatal (If Ax spiking is shown to be in error, and if no additional or replacement)

Ax ES SS JS Ax - N N N ES - - Y Y SS - - - Y

“PD Requirements Decision Matrix” – Defective JS Spiking (use BCS3 RRFs)

percent recovery measurements for ES affected, not the analytes or the SS

Ax ES SS JS Ax - Y Y N ES - - Y N SS - - - N

“PD Requirements Decision Matrix” – Defective SS Spiking (use BCS3 RRFs)

percent recovery measurements for SS affected, not the analytes

Ax ES SS JS Ax - Y N Y ES - - N Y SS - - - N

“PD Requirements Decision Matrix” – Defective ES Spiking (use BCS3 RRFs; see Levels PD-1 or PD-2; for air samples only, consider using the Ax vs SS RRFs)

Ax ES SS JS Ax - N Y Y ES - - N N SS - - - Y




TABLE 8: Acceptance Criteria for Tetra-Only Analyses and Labeled Compound Recoveries in Samples for Analyses with all 17 2,3,7,8-PCDD/Fs




16.0 Definitions

16.1 Definitions (and more thorough explanations) of concepts unique to SGS Analytical Perspectives may be found in the Performance Commitment document.

16.2 See Document DC_139 Appendix D for a full list of definitions.

16.3 AS = alternate standard

16.4 Ax = target analyte

16.5 BCS3 = batch control spike

16.6 CAA = Clean Air Act

16.7 COC = chain of custody

16.8 CPSM = column performance standard mixture

16.9 CS3 = calibration solution corresponding to the middle point of the initial calibration curve

16.10 CS = cleanup standard

16.11 CWA = Clean Water Act

16.12 dfa Technology = dioxin-furan array (fingerprinting reminiscent of DNA plates)

16.13 DOC = documentation of capability

16.14 DUP = duplicate

16.15 EDL = estimated detection limit (sample- and analyte-specific)

16.16 ES = extraction standard (isotopically labeled standard added before the extraction)

16.17 FS = field concentration standard (used for monitoring in-field concentration of solvents and may be used for monitoring of other non-routine processes)

16.18 GC = gas chromatography

16.19 ICAL = initial calibration

16.20 ID-HRMS = comprehensive and stable isotope-dilution high-resolution mass spectrometry

16.21 Ind. Val. = independent validation

16.22 JS = injection standard

16.23 LOD = limit of detection

16.24 LOQ = limit of quantitation




16.25 MB = lab method blank

16.26 MDL = method detection limit (not sample-specific)

16.27 M/∆M = mass spectrometer resolving power

16.28 MIR = Methods Innovation Rule

16.29 ML = minimum level (equivalent to lowest point on the calibration curve)

16.30 MS/MSD = matrix spike/matrix spike duplicate

16.31 ND = not detected

16.32 NS = native spike

16.33 NELAC = National Environmental Laboratory Accreditation Conference

16.34 OPR = on-going precision & recovery (equivalent to LCS or Lab Control Spike)

16.35 PAH = polynuclear aromatic hydrocarbon

16.36 PBMS = performance-based measurement system

16.37 PCB = polychlorinated biphenyl

16.38 PCDD/F = polychlorinated dibenzo-p-dioxin and dibenzofuran

16.39 PD = percent difference

16.40 PE = performance evaluation sample

16.41 QI = quantitative interference

16.42 QuanTIC = selected SVOAs by full-scan GC/MS (ID-HRMS)

16.43 RCRA = Resources Recovery Act

16.44 RL = reporting limit (usually represents the lowest point on the calibration curve or ML)

16.45 RPD = relative percent difference

16.46 RRF = relative response factor

16.47 RSD = relative standard deviation

16.48 RTW = retention time defining window solution

16.49 SDWA = Safe Drinking Water Act

16.50 SICP = selected ion current profile

16.51 S/N = signal-to-noise ratio

16.52 SRM = standard reference material

16.53 SS = sampling standard

16.54 TEQ = toxic equivalency quotient




16.55 TS = field concentration standard (used for monitoring in-field concentration of solvents and may be used for monitoring of other non-routine processes)

16.56 TSCA = Toxic Substances Control Act

16.57 U-SVOA = ultra-semi volatile analyte (selected SVOAs by SIR by ID-HRMS)

16.58 VER = continuing calibration verification (equivalent to ConCal)

16.59 WHO-2 / WHO-2S = ID-HRMS assay for the 29 World Health Organization target analytes (S = serum)

17.0 References

17.1 SGS document DC_139, Appendix D Definitions.

17.2 SGS document MI_278, Waste Disposal and Pollution Prevention.

17.3 SGS document MI_141, Review of Analytical Data.

17.4 USEPA Method 8290, Revision 0, Dated September 1994.

17.5 USEPA Method 8290B, Dated July 2002.

17.6 USEPA Method 1613, Revision B, Dated October 1994.

17.7 USEPA Method 8290A; Revision 1, Dated February 2007.

SGS North America Inc. - newtowncreek.infonewtowncreek.info/docs2/2 Remedial Investigation/Phase 2...

Documents

Transcript of SGS North America Inc. - newtowncreek.infonewtowncreek.info/docs2/2 Remedial Investigation/Phase 2...