7.2.5 Blanks: Requirements in the methods must be followed. A laboratory reagent blank should be carried through the full analytical procedure with every sample batch. In general, results from laboratory reagent blanks should not exceed the laboratory's Minimum Reporting Limit, the lo west concentration of standard used for quantitation. ( MRL) 7.2.6 Laboratory Fortified Blanks: Requirements in the methods must be followed. LFBs should be nalyzed at the level specified in the method. Some methods require that a laboratory fortified blank at ten times the MDL or a mid level concentration be analyzed with each batch of samples. Precision and accuracy data reporting limit for each analyte by analyzing a laboratory fortified blank at the minimum reporting leve/ um should be documented for this determination. In addition, the analyst should routinely verify the 7. 2.7 Laboratory Fortified Sample Matrix: Laboratory fortified sample matrix requirements in the methods must be met. If there are no laboratory fortified sample matrix requirements in the method, the following are guidelines to be used. The laboratory should add a known quantity of analytes to a percentag (to be described in the approved SoP)of the routine samples to determine sample matrix interference. The fortified concentration should not be less than the concentration of the sample selected for fortification unless specified by the method. If the sample concentration is unknown or less than detectable, the analyst should choose an appropriate concentration(e.g, a percentage of the mCl or mid point in the calibration range). Over time, samples from all routine sample sources should be fortified. The procedure should be described in the SoP. If any of these checks are not within the criteria specified in the method or control limits specified in 7.2.7, and the laboratory performance is in control, the result for that sample should be flagged to inform the data user that the results are suspect due to matrix effects. 7.2.8 Control Charts: Control charts for accuracy and precision, generated from laboratory fortified blanks (LFBs)should be maintained and used by the laboratory. Until sufficient data are available from the laboratory usually a minimum of 20 to 30 test results on a specific analysis, the laboratory should use the control limits specified in the methods. If there are no control limits specified in the method, the limits may be statistically calibrated using the procedure below When sufficient data become available, the laboratory should develop LFB control charts from the mean percen ecovery(x) and the standard deviation (S)of the percent recovery for the qc checks specified above(se Standard Methods for the Examination of Water and Wastewater, part 1020B, or similar QC reference texts for further information). These data are used to establish upper and lower control limits as follows upper control limit=x+ 3s(upper warning limit 2S) lower control limit =x-3s (lower warning limit- 2S) After each five to ten new recovery measurements, new control limits should be calculated using the most recent 20-30 data points. These calculated control limits should not exceed those established in the method. If any of these control limits are tighter than the method specifications, the laboratory should use the tighter criteria 7.2.9 Initial Demonstration of Capability: Requirements in the methods must be followed. Before beginning the analysis of compliance samples, an initial demonstration of capability (IDC)must be performed for each method as required in the method. The IdC includes a demonstration of the ability to achieve a low ackground the precision and accuracy required by the method, and determination of the method detection limit (MDL)(see below). An IDC should be performed for each instrument. It is also recommended that an IdC b performed by each analyst. In addition, it is lended that the IdC also address the variability introduced if more than one sample preparation technician is used. Precision, accuracy and mdl should be similar for made which could affect the precision or accuracy or sensitivity. Minor changes should prompt a chect,'p each technician. The analyst should recalculate IDCs when a change in the method, analyst or instrument ascertain that the precision, accuracy and sensitivity have been maintained ⅣV-7
IV- 7 7.2.5 Blanks: Requirements in the methods must be followed. A laboratory reagent blank should be carried through the full analytical procedure with every sample batch. In general, results from laboratory reagent blanks should not exceed the laboratory's Minimum Reporting Limit, the lowest concentration of standard used for quantitation. (MRL). 7.2.6 Laboratory Fortified Blanks: Requirements in the methods must be followed. LFBs should be analyzed at the level specified in the method. Some methods require that a laboratory fortified blank at ten times the MDL or a mid level concentration be analyzed with each batch of samples. Precision and accuracy data should be documented for this determination. In addition, the analyst should routinely verify the minimum reporting limit for each analyte by analyzing a laboratory fortified blank at the minimum reporting level. 7.2.7 Laboratory Fortified Sample Matrix: Laboratory fortified sample matrix requirements in the methods must be met. If there are no laboratory fortified sample matrix requirements in the method, the following are guidelines to be used. The laboratory should add a known quantity of analytes to a percentage (to be described in the approved SOP) of the routine samples to determine sample matrix interference. The fortified concentration should not be less than the concentration of the sample selected for fortification unless specified by the method. If the sample concentration is unknown or less than detectable, the analyst should choose an appropriate concentration (e.g., a percentage of the MCL or mid point in the calibration range). Over time, samples from all routine sample sources should be fortified. The procedure should be described in the SOP. If any of these checks are not within the criteria specified in the method or control limits specified in 7.2.7, and the laboratory performance is in control, the result for that sample should be flagged to inform the data user that the results are suspect due to matrix effects. 7.2.8 Control Charts: Control charts for accuracy and precision, generated from laboratory fortified blanks (LFBs) should be maintained and used by the laboratory. Until sufficient data are available from the laboratory, usually a minimum of 20 to 30 test results on a specific analysis, the laboratory should use the control limits specified in the methods. If there are no control limits specified in the method, the limits may be statistically calibrated using the procedure below. When sufficient data become available, the laboratory should develop LFB control charts from the mean percent recovery (0) and the standard deviation (S) of the percent recovery for the QC checks specified above (see Standard Methods for the Examination of Water and Wastewater, part 1020B, or similar QC reference texts for further information). These data are used to establish upper and lower control limits as follows: upper control limit = 0 + 3S (upper warning limit + 2S) lower control limit = 0 - 3S (lower warning limit - 2S) After each five to ten new recovery measurements, new control limits should be calculated using the most recent 20-30 data points. These calculated control limits should not exceed those established in the method. If any of these control limits are tighter than the method specifications, the laboratory should use the tighter criteria. 7.2.9 Initial Demonstration of Capability: Requirements in the methods must be followed. Before beginning the analysis of compliance samples, an initial demonstration of capability (IDC) must be performed for each method as required in the method. The IDC includes a demonstration of the ability to achieve a low background, the precision and accuracy required by the method, and determination of the method detection limit (MDL)(see below). An IDC should be performed for each instrument. It is also recommended that an IDC be performed by each analyst. In addition, it is recommended that the IDC also address the variability introduced if more than one sample preparation technician is used. Precision, accuracy and MDL should be similar for each technician. The analyst should recalculate IDCs when a change in the method, analyst or instrument is made which could affect the precision or accuracy or sensitivity. Minor changes should prompt a check to ascertain that the precision, accuracy and sensitivity have been maintained
7.2.10 Quantitation of Multicomponent Organic Analytes (toxaphene, chlordane and PCBs) The quantitation of multicomponent analytes requires professional judgment on the part of the analyst. This is required due to the complex nature of the chromatography involved, sample weathering, degradation and interferences that may be present in the samples. The pattern of peaks found in the sample should be examined carefully and compared to a standard. The peaks in the sample that match the peak ratios in the standard can be used in quantitation. Peaks that have obvious interferences(such as pesticides or phthalates or peak exhibiting poor peak shape)or appear to have been degraded or weathered should not be used for quantitatie A representative number (5-9)of peaks is suggested. Peak area should be used for quantitation and the analyst should ensure that the samples and standards have been integrated in the same manner. Quantitation can b lone by using the total peak area or height(comparing the area of the 5-9 peaks used for quantitation of the sample to the area of the standard)or by calculating each peak separately (using area) and taking the average concentration of the 5-9 peaks. Because of factors such as peak shape and baseline rise, the most accurate quantitation is obtained when the concentration of the sample closely matches that of the standard(e.g, within 20% of the standard). See EPA Method 8081, Organochlorine Pesticides and PCBs as Aroclors by gas Chromatography: Capillary Column Technique, (EPA Sw 846 Test Methods for Evaluating Solid Waste Physical/Chemical Methods, Third Edition) for a more detailed discussion of quantitation of multicomponent lote: PCBs are qualitatively identified as Aroclors and measured for compliance purposes as decachlorobiphenyl. Chlordane is regulated as technical chlordane, a mixture of at least 11 major components and 30 minor 7.2.11 MDL Calculation: Requirements in the methods must be followed. Most methods require initial MDL calculations for all analytes and certification officers should require the laboratories to calculate their detection limits for all regulated contaminants. If there is no procedure to determine the detection limits in the method, it should be determined in accordance with the procedure given in 40 CFR 136. Appendix B. the CFR, at14124(f(17)(ii(C) requires an MDL of 0.0005 mg/L be attained for VOCs, and 141.89(a(1(iii) requires an MDL of 0.001 mg/L be attained for lead if the lab will ba he MCL must be attained for ompositing [CFR 141. 23(a)(4) and [CFR 141.24(f(10). voCs should not be composited. The SoC detection limits listed at CFR(141.24(h)(18)are required to reduce monitoring(CFR 141.24(f(I1(iv). Table TV-8 lists the MCLs, MCLGs and MDls, for VOCs which are in the drinking water regulations. Table Iv-9 lists the SOC MCLs, MCLGs and Monitoring Triggers Sample preparation and analyses for the mdl calculation should be made over a period of at least three days to include day-to-day variation as an additional source of error. The analyst should determine Mdls initially, when any change is made which could affect the MDls, or more frequently if required by the method. (Inorganic methods may require MDls to be determined differently, and in all cases the methods must be followed. In addition, the analyst must demonstrate low level capability on an ongoing basis through an MDL determination or repeated low level analyses(MRL). The calculation of MDLs by the CFR procedure may not be adequate for toxaphene and chlordane because they require pattern or peak profile recognition for identification. Presently, no standard procedure exists, so it is recommended that the mdl be defined as the lowest concentration for which pattern recognition is possible Pattern recognition is used for qualitative identification of PCBs as Aroclors. Quantitation of PCBs is achieved by conversion of PCBs to decachlorobiphenyl ( dCB) 7. 2.12 Low Level Quantitation: The laboratory s minimum reporting limits(MRL) should be reported to the client along with the data. The reporting limit must be below the MCL. Laboratories should run a lFB at their MRL every analysis day and should not report contaminants at levels less than the level at which they routinely analyze their lowest standard. While this is a scientifically sound practice, whether it is an acceptable ractice will depend on State and Federal reporting requirements. It is important for users of data to understand the statistical and qualitative significance of the data. Laboratories may be required by the States to achieve a specific MDL or quantitation limit more stringent than that required by epa 8
IV- 8 7.2.10 Quantitation of Multicomponent Organic Analytes (toxaphene, chlordane and PCBs) The quantitation of multicomponent analytes requires professional judgment on the part of the analyst. This is required due to the complex nature of the chromatography involved, sample weathering, degradation and interferences that may be present in the samples. The pattern of peaks found in the sample should be examined carefully and compared to a standard. The peaks in the sample that match the peak ratios in the standard can be used in quantitation. Peaks that have obvious interferences (such as pesticides or phthalates or peaks exhibiting poor peak shape) or appear to have been degraded or weathered should not be used for quantitation. A representative number (5-9) of peaks is suggested. Peak area should be used for quantitation and the analyst should ensure that the samples and standards have been integrated in the same manner. Quantitation can be done by using the total peak area or height (comparing the area of the 5-9 peaks used for quantitation of the sample to the area of the standard) or by calculating each peak separately (using area) and taking the average concentration of the 5-9 peaks. Because of factors such as peak shape and baseline rise, the most accurate quantitation is obtained when the concentration of the sample closely matches that of the standard (e.g., within 20% of the standard). See EPA Method 8081, Organochlorine Pesticides and PCBs as Aroclors by Gas Chromatography: Capillary Column Technique, (EPA SW 846 Test Methods for Evaluating Solid Waste Physical/Chemical Methods, Third Edition) for a more detailed discussion of quantitation of multicomponent analytes. Note: PCBs are qualitatively identified as Aroclors and measured for compliance purposes as decachlorobiphenyl. Chlordane is regulated as technical chlordane, a mixture of at least 11 major components and 30 minor ones. 7.2.11 MDL Calculation: Requirements in the methods must be followed. Most methods require initial MDL calculations for all analytes and certification officers should require the laboratories to calculate their detection limits for all regulated contaminants. If there is no procedure to determine the detection limits in the method, it should be determined in accordance with the procedure given in 40 CFR 136, Appendix B. The CFR, at141.24(f)(17)(ii)(C) requires an MDL of 0.0005 mg/L be attained for VOCs, and 141.89(a)(1)(iii) requires an MDL of 0.001 mg/L be attained for lead if the lab will be processing source water composite samples. For inorganics and SOCs, a method detection limit of 1/5 of the MCL must be attained for compositing [CFR 141.23(a)(4)] and [CFR 141.24(f)(10)]. VOCs should not be composited. The SOC detection limits listed at CFR (141.24(h)(18) are required to reduce monitoring (CFR 141.24(f)(11)(iv). Table IV-8 lists the MCLs, MCLGs and MDLs, for VOCs which are in the drinking water regulations. Table IV-9 lists the SOC MCLs, MCLGs and Monitoring Triggers. Sample preparation and analyses for the MDL calculation should be made over a period of at least three days to include day-to-day variation as an additional source of error. The analyst should determine MDLs initially, when any change is made which could affect the MDLs, or more frequently if required by the method. (Inorganic methods may require MDLs to be determined differently, and in all cases the methods must be followed.) In addition, the analyst must demonstrate low level capability on an ongoing basis through an MDL determination or repeated low level analyses (MRL). The calculation of MDLs by the CFR procedure may not be adequate for toxaphene and chlordane because they require pattern or peak profile recognition for identification. Presently, no standard procedure exists, so it is recommended that the MDL be defined as the lowest concentration for which pattern recognition is possible. Pattern recognition is used for qualitative identification of PCBs as Aroclors. Quantitation of PCBs is achieved by conversion of PCBs to decachlorobiphenyl (DCB). 7.2.12 Low Level Quantitation: The laboratory's minimum reporting limits (MRL) should be reported to the client along with the data. The reporting limit must be below the MCL. Laboratories should run a LFB at their MRL every analysis day and should not report contaminants at levels less than the level at which they routinely analyze their lowest standard. While this is a scientifically sound practice, whether it is an acceptable practice will depend on State and Federal reporting requirements. It is important for users of data to understand the statistical and qualitative significance of the data. Laboratories may be required by the States to achieve a specific MDL or quantitation limit more stringent than that required by EPA
8. Records and Data Reporting 8.1 Legal Defensibility: Compliance monitoring data should be made legally defensible by keeping thorough and accurate records. The Qa plan and/or SOPs need to(EPa Order 5360. 1)describe the policies and procedures used by the facility for record integrity, retention and storage. If samples are expected to become part of a legal action, chain of custody procedures should be used (See Appendix A) 8.2 Maintenance of Records: Public Water Systems are required to maintain records of chemical analyses of compliance samples for 10 years(40 CFR 141.33)and lead and copper for 12 years(40 CFR 141.91). The laboratory should maintain easily accessible records for five years or until the next certification data audit is complete, whichever is longer. Changes in ownership, mergers, or closures of laboratories do not eliminate these requirements. The client water system should be notified before disposing of records so they may request copies if needed. This includes all raw data, calculations, and quality control data. These data files may be either hard copy, microfiche or electronic. Electronic data should always be backed up by protected tape or disk or hard copy. If the laboratory changes its computer hardware or software, it should make provisions for transferring old data to the new system so that it remains retrievable within the time frames specified above. Data which is expected to become part of a legal action may need to be maintained for a longer period of time. Check with your legal counsel. 8.3 Sampling Records: Data should be recorded in ink with any changes lined through such that the original entry is visible. Data may also be kept electronically. Changes need to be initialed and dated. The following information should be readily available 8.3.1 Date, location(including name of utility and PWSS ID #), site within the system, time of sampling, name, organization and phone number of the sampler, and analyses required 8.3.2 Identification of the sample as to whether it is a routine distribution system sample, check sample, raw or finished water sample, repeat or confirmation sample or other special purpose sample 8.3.3 Date of receipt of the sample 8.3.4 Sample volume/weight, container type, preservation and holding time and condition on receipt 8.3.5 pH and disinfectant residual at time of sampling (if required)(from plant records); 8.3.6 Transportation and delivery of the sample(person/carrier, conditions) Analytical Records Data should be recorded in ink with any changes lined through such that original entry is ole. Changes need to be initialed and dated The following information should be readily available 8.4.1 Laboratory and persons responsible for performing analysis, 8.4.2 Analytical techniques/methods used 8.4.3 Date and time of analysis 8.4.4 Results of sample and quality control analyses; 8.4.5 Calibration and standards information 8.4.6 Analyst and technician Initial Demonstration of Capability documentation should be kept on file as well as results of proficiency testing 8.5 Reconstruction of Data: Adequate information should be available to allow the auditor to reconstruct the final results for compliance samples and PT samples ⅣV-9
IV- 9 8. Records and Data Reporting 8.1 Legal Defensibility: Compliance monitoring data should be made legally defensible by keeping thorough and accurate records. The QA plan and/or SOPs need to (EPA Order 5360.1) describe the policies and procedures used by the facility for record integrity, retention and storage. If samples are expected to become part of a legal action, chain of custody procedures should be used (See Appendix A). 8.2 Maintenance of Records: Public Water Systems are required to maintain records of chemical analyses of compliance samples for 10 years (40 CFR 141.33) and lead and copper for 12 years (40 CFR 141.91). The laboratory should maintain easily accessible records for five years or until the next certification data audit is complete, whichever is longer. Changes in ownership, mergers, or closures of laboratories do not eliminate these requirements. The client water system should be notified before disposing of records so they may request copies if needed. This includes all raw data, calculations, and quality control data. These data files may be either hard copy, microfiche or electronic. Electronic data should always be backed up by protected tape or disk or hard copy. If the laboratory changes its computer hardware or software, it should make provisions for transferring old data to the new system so that it remains retrievable within the time frames specified above. Data which is expected to become part of a legal action may need to be maintained for a longer period of time. Check with your legal counsel. 8.3 Sampling Records: Data should be recorded in ink with any changes lined through such that the original entry is visible. Data may also be kept electronically. Changes need to be initialed and dated. The following information should be readily available: 8.3.1 Date, location (including name of utility and PWSS ID #), site within the system, time of sampling, name, organization and phone number of the sampler, and analyses required; 8.3.2 Identification of the sample as to whether it is a routine distribution system sample, check sample, raw or finished water sample, repeat or confirmation sample or other special purpose sample; 8.3.3 Date of receipt of the sample; 8.3.4 Sample volume/weight, container type, preservation and holding time and condition on receipt; 8.3.5 pH and disinfectant residual at time of sampling (if required) (from plant records); 8.3.6 Transportation and delivery of the sample (person/carrier, conditions). 8.4 Analytical Records Data should be recorded in ink with any changes lined through such that original entry is visible. Changes need to be initialed and dated The following information should be readily available: 8.4.1 Laboratory and persons responsible for performing analysis; 8.4.2 Analytical techniques/methods used; 8.4.3 Date and time of analysis; 8.4.4 Results of sample and quality control analyses; 8.4.5 Calibration and standards information. 8.4.6 Analyst and technician Initial Demonstration of Capability documentation should be kept on file as well as results of proficiency testing. 8.5 Reconstruction of Data: Adequate information should be available to allow the auditor to reconstruct the final results for compliance samples and PT samples
8.6 Computer Programs: Computer programs should be verified initially and periodically by manual calculations and the calculations should be available for inspection. Access to computer programs and electronic data need to be limited to appropriate personnel 9. Action in Response to Noncompliant Laboratory results When a laboratory is responsible, either by contract or State policy, to report sample results which would indicate a system is out of compliance, the laboratory must (141.23(a)(4)(i)) promptly notify the proper authority so that the authority can request the water utility to resample from the same sampling location(s)immediately. See Chapter Ill. ⅣV-10
IV- 10 8.6 Computer Programs: Computer programs should be verified initially and periodically by manual calculations and the calculations should be available for inspection. Access to computer programs and electronic data need to be limited to appropriate personnel. 9. Action in Response to Noncompliant Laboratory Results When a laboratory is responsible, either by contract or State policy, to report sample results which would indicate a system is out of compliance, the laboratory must (141.23(a)(4)(i)) promptly notify the proper authority so that the authority can request the water utility to resample from the same sampling location(s) immediately. See Chapter III
Table Iv-1 Glassware Cleaning Procedures-consult the method for complete details; do not heat volumetr Method Washing Drying 502.2/504/504.1/524.2 Detergent wash, rinse with tap 105° C for l hour and distilled water 505 Detergent wash, rinse with tap 400C for 1 hour or rinse with and reagent water acetone 506 Rinse immediately with last 400 C for l hour or rinse with solvent used, wash with hot water acetone and detergent, rinse with tap and reagent water 507/508 Rinse immediately with last 400C for I hour or rinse with solvent used, wash with hot water acetone and detergent, rinse with tap and Detergent wash, rinse with tap 400°Cfor2 hours and reagent water or solvent rinse 508A No specifications, suggest the no specification, suggest the same as 515.1/515.2 same as515.1/515.2 515.1/515.2 Rinse immediately with last 400C for 1 hour or rinse with olvent used, wash with hot water acetone and detergent, rinse with dilute acid, tap and reagent water 5153 Rinse immediately with last 400C for I hour or rinse with solvent used, wash with hot water acetone and detergent, reagent water 515.4 Wash with tap water and In place of solvent rinse, muffle detergent, rinse with tap and at400°Cfor2 hours reagent water. A solvent rinse Heat volumetrics may be necessary at120°C 524.2 Not described in method Not described in method 525.2 Detergent wash, rinse with tap air dry or muffle(no specs and distilled water or solvent (suggest 400%C for I hour) 531.1/6610 Rinse immediately with last 450 C for I hour or rinse with solvent used, wash with hot water acetone and detergent, rinse with tap and reagent water ⅣV-11
IV- 11 Table IV-1 Glassware Cleaning Procedures - consult the method for complete details; do not over-heat volumetric glassware Method Washing Drying 502.2/504/504.1/524.2 Detergent wash, rinse with tap and distilled water 105°C for 1 hour 505 Detergent wash, rinse with tap and reagent water 400°C for 1 hour or rinse with acetone 506 Rinse immediately with last solvent used, wash with hot water and detergent, rinse with tap and reagent water 400°C for 1 hour or rinse with acetone 507/508 Rinse immediately with last solvent used, wash with hot water and detergent, rinse with tap and reagent water 400°C for 1 hour or rinse with acetone 508.1 Detergent wash, rinse with tap and reagent water or solvent rinse 400°C for 2 hours 508A No specifications, suggest the same as 515.1/515.2 no specification, suggest the same as 515.1/515.2 515.1/515.2 Rinse immediately with last solvent used, wash with hot water and detergent, rinse with dilute acid, tap and reagent water 400°C for 1 hour or rinse with acetone 515.3 Rinse immediately with last solvent used, wash with hot water and detergent, rinse with tap and reagent water 400°C for 1 hour or rinse with acetone 515.4 Wash with tap water and detergent, rinse with tap and reagent water. A solvent rinse may be necessary. In place of solvent rinse, muffle at 400°C for 2 hours. Heat volumetrics at 120°C. 524.2 Not described in method Not described in method 525.2 Detergent wash, rinse with tap and distilled water or solvent rinse air dry or muffle(no specs) (suggest 400°C for 1 hour) 531.1/6610 Rinse immediately with last solvent used, wash with hot water and detergent, rinse with tap and reagent water 450°C for 1 hour or rinse with acetone