Loading...
HomeMy WebLinkAbout#199 2012 INSPECTION REPORT ROUTING SHEET To be attached to all inspection reports in-house only. Laboratory Cert. #: 199 Laboratory Name: Archer Daniels Midland Company Inspection Type: Industrial Maintenance Inspector Name(s): Todd Crawford Inspection Date: November 29, 2012 Date Report Completed: December 10, 2012 Date Forwarded to Reviewer: December 10, 2012 Reviewed by: Chet Whiting Date Review Completed: February 11, 2013 Cover Letter to use: Insp. Initial Insp. Reg. Insp. No Finding Insp. CP Corrected Unit Supervisor: Dana Satterwhite Date Received: February 13, 2013 Date Forwarded to Linda: February 18, 2013 Date Mailed: February 18, 2013 _____________________________________________________________________ On-Site Inspection Report LABORATORY NAME: Archer Daniels Midland Company NPDES PERMIT # : NC0027065 WATER QUALITY PERMIT # : WQ0004500 NC GENERAL PERMIT #: NCG060000 ADDRESS: 1730 Moore Street, S.E. Southport, NC 28461 CERTIFICATE #: 199 DATE OF INSPECTION: November 29, 2012 TYPE OF INSPECTION: Industrial Maintenance AUDITOR: Todd Crawford LOCAL PERSON(S) CONTACTED: Stewart Clarke, Scott Phillips and Jonathan Ward I. INTRODUCTION: This laboratory was inspected by a representative of the North Carolina Wastewater/Groundwater Laboratory Certification (NC WW/GW LC) program to verify its compliance with the requirements of 15A NCAC 2H .0800 for the analysis of environmental samples. II. GENERAL COMMENTS: The laboratory was clean and well organized. The facility has all the equipment necessary to perform the analyses. A system for traceability of standard and reagent preparation is in place but needs to include some additional information. Proficiency Testing (PT) samples have been analyzed for all certified parameters for the 2012 proficiency testing calendar year and the graded results were 100% acceptable. The laboratory is reminded that any time changes are made to laboratory operations; the laboratory must update the Quality Assurance (QA)/Standard Operating Procedures (SOP) document(s). Any changes made in response to the Findings, Recommendations or Comments listed in this report must be incorporated to insure the method is being performed as stated, references to methods are accurate, and the QA and/or SOP document(s) is in agreement with approved practice and regulatory requirements. In some instances, the laboratory may need to create a SOP to document how new functions or policy will be implemented. On May 18, 2012, EPA promulgated changes to the list of Clean Water Act (CWA) methods at 40 CFR Part 136.3. This action, referred to as the Methods Update Rule (MUR) approves new methods, or changes to existing methods, that affects over 100 EPA methods, Standard Methods, ASTM methods, and other test procedures in Part 136 of Title 40 of the Code of Federal Regulations (CFR). The rule also contains a number of clarifications relating to approved methods, sample preservation and holding times, and method modifications. The final rule may be found at: http://water.epa.gov/scitech/methods/cwa/update_index.cfm. Each laboratory will need to review the MUR and evaluate its effect on current laboratory practices. These changes must be made in the laboratory’s Standard Operating Procedures and in Quality Manuals, as well as any other place where the method is cited, e.g., reports, benchsheets, logs, etc. Page 2 #199 Archer Daniels Midland Company Contracted analyses are performed by Environmental Chemists, Inc. (Certification #94). The requirements associated with Findings C, D and M are new policies that have been implemented by our program since the last inspection. III. FINDINGS, REQUIREMENTS, COMMENTS AND RECOMMENDATIONS: Traceability A. Finding: The laboratory needs to increase the documentation of purchased materials and reagents. Requirement: All chemicals, reagents, standards and consumables used by the laboratory must have the following information documented: Date received, Date Opened (in use), Vendor, Lot Number, and Expiration Date (where specified). A system (e.g., traceable identifiers) must be in place that links standard/reagent preparation information to analytical batches in which the solutions are used. Documentation of solution preparation must include the analyst’s initials, date of preparation, the volume or weight of standard(s) used, the solvent and final volume of the solution. This information as well as the vendor and/or manufacturer, lot number, and expiration date must be retained for primary standards, chemicals, reagents, and materials used for a period of five years. Consumable materials such as pH buffers, lots of pre-made standards and/or media, solids and bacteria filters, etc. are included in this requirement. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy. Quality Control B. Finding: The analytical balance weights have not been verified against ASTM standard weights since 4/2/87. Requirement: ASTM Class 1 and 2 weights must be verified at least every 5 years. ASTM Class 1 weights (20 g to 25 kg) and ASTM Class 2 weights (10 g to 1 mg) are equivalent to the NBS Class S weights specified in 15A NCAC 2H .0805 (a) (7) (K). Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy. Comment: The weights were tarnished and scratched. The analyst stated that he uses metal tongs to handle the weights since the plastic tongs have been lost. Recommendation: It is recommended that the weight set be replaced due to the advanced state of deterioration and that the lab obtain plastic tongs for handling weights. Ammonia Nitrogen – Standard Methods, 4500 NH3 D - 1997 Comment: This test is only required quarterly. Samples are subcontracted and there was no regulatory data to review. If this analysis is again performed in-house, the findings associated with TKN analysis will apply. Ammonia Nitrogen – Standard Methods, 4500 NH3 D - 1997 Total Kjeldahl Nitrogen – Standard Methods, 4500 Norg B-1997 (NH3 D - 1997) C. Finding: A calibration blank and calibration verification standard (mid-range) are not analyzed initially, after every tenth sample and at the end of the sample group. Page 3 #199 Archer Daniels Midland Company Requirement: The calibration blank and calibration verification standard (mid-range) must be analyzed initially (i.e., prior to sample analysis), after every tenth sample and at the end of each sample group to check for carry over and calibration drift. If either fall outside established quality control acceptance criteria, corrective action must be taken (e.g., repeating sample determinations since the last acceptable calibration verification, repeating the initial calibration, etc.). Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy based upon Standard Methods, 20th Edition, 1020 B. (10) (c), 3020 B. (2) (b), and 4020 B. (2). Requirement: For analyses requiring a calibration curve, the concentration of method and reagent blanks must not exceed 50% of the reporting limit, unless otherwise specified by the reference method. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy. Comment: The calibration verification standard is prepared from the calibration stock standard solution. Ammonia Nitrogen – Standard Methods, 4500 NH3 D - 1997 Total Kjeldahl Nitrogen – Standard Methods, 4500 Norg B-1997 (NH3 D – 1997) NO2 + NO3 Nitrogen – Standard Methods, 4500 NO3 E - 2000 D. Finding: The laboratory is not analyzing matrix spikes. Requirement: Unless the referenced method states a greater frequency, spike 5% of samples on a monthly basis. Laboratories analyzing less than 20 samples per month must analyze at least one matrix spike each month samples are analyzed. Prepare the matrix spike from a reference source different from that used for calibration unless otherwise stated in the method. If matrix spike results are out of control, the results must be qualified or the laboratory must take corrective action to rectify the effect, use another method, or employ the method of standard additions. When the method of choice specifies matrix spike performance acceptance criteria for accuracy, and the laboratory chooses to develop statistically valid, laboratory-specific limits, the laboratory-generated limits cannot be less stringent than the criteria stated in the approved method. If the unspiked sample result is in the top 40% of the calibration range, the sample should be diluted and the matrix spike prepared using the diluted sample. The recovery of the matrix spike samples must be bracketed by the calibration range. The volume of spike solution used in matrix spike preparation must in all cases be ≤ 10% of the total matrix spike volume. It is preferable that the spike solution constitutes ≤ 1% of the total matrix spike volume so that the matrix spike can be considered a whole volume sample with no adjustment by calculation necessary. If the spike solution volume constitutes >1% of the total sample volume, the sample concentration or spike concentration must be adjusted by calculation. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy. Recommendation: It is recommended that the spike solution constitutes ≤1% of the total MS volume. Recommendation: If the duplicate sample value is near or below the reporting limit, it is recommended that a Matrix Spike Duplicate (MSD) be analyzed. The MSD fulfills the duplicate required by 15A NCAC 2H .0805 (a) (7) (C). This would result in better statistical results to monitor method and analyst precision. Page 4 #199 Archer Daniels Midland Company NO2 + NO3 Nitrogen – Standard Methods, 4500 NO3 E - 2000 E. Finding: A second source standard is not analyzed. Requirement: When a standard curve is manually prepared (as opposed to a factory-set calibration), it is required to analyze one known standard in addition to calibration standards each day samples are analyzed to document accuracy. This standard must be prepared from materials obtained from a source independent from the one used for preparing the calibration standards (often referred to as a second source standard or external reference standard). A second source standard may be: • a quality control standard obtained from a vendor, • a standard prepared from primary standards obtained from a second vendor, or • a standard made from primary standards from the same vendor but from a different lot number (i.e., an independent lot) as those used to make the calibration standards. Second source standards must be evaluated using one of the following: vendor supplied criteria, method-defined acceptance criteria, in-house calculated acceptance limits that are statistically-derived from historical data based on three standard deviations from the mean in the detectable range or other statistically viable evaluation criterion. If the results fall outside of acceptance limits, the analysis is out of control. The analysis must be terminated and the problem corrected prior to sample analysis. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy based upon 15A NCAC 2H .0805 (a) (7) (B), (a) (7) (F), and Standard Methods, 20th Edition, 1020 B. (2) and (5). F. Finding: The method blank analysis is not documented. Requirement: Each laboratory shall develop and maintain a document outlining the analytical quality control practices used for the parameters included in their certification. Supporting records shall be maintained as evidence that these practices are being effectively carried out. Ref: 15A NCAC 2H .0805 (a) (7). Requirement: For analyses requiring a calibration curve, the concentration of method and reagent blanks must not exceed 50% of the reporting limit, unless otherwise specified by the reference method. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy. G. Finding: The calibration curve does not always bracket the concentration of the samples. Requirement: For analytical procedures requiring analysis of a series of standards, the concentrations of those standards must bracket the concentration of the samples analyzed. One of the standards must have a concentration equal to the laboratory’s lower reporting concentration for the parameter involved. Ref: 15A NCAC 02H .0805 (a) (7) (I). Comment: Samples analyzed on 10/24/12 and 11/14/12 were documented to have concentrations that exceeded the concentration of the highest calibration standard of 1.0 mg/L and corrective action was not taken. Comment: The proper corrective action in this case would have been to either dilute the sample and reanalyze or extend the high end of the curve. Page 5 #199 Archer Daniels Midland Company Conductivity – Standard Methods, 2510 B - 1997 H. Finding: The Automatic Temperature Compensator (ATC) was not verified initially and every 12 months thereafter. Requirement: The ATC must be verified annually (i.e., every twelve months) and the process documented. The ATC must be verified by analyzing a standard at 25 °C (the temperature that conductivity values are compensated to) and a temperature(s) that brackets the temperature ranges of the samples to be analyzed. This may require the analysis of a third temperature reading that is > 25 °C. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy. I. Finding: The true values of the calibration standard and the calibration check standards are not correctly labeled on the bottles or on the Conductivity Standards Preparation sheet. Requirement: Each laboratory shall develop and maintain a document outlining the analytical quality control practices used for the parameters included in their certification. Supporting records shall be maintained as evidence that these practices are being effectively carried out. Ref: 15A NCAC 2H .0805 (a) (7). Residue, Total Suspended – Standard Methods, 2540 D - 1997 Comment: Although all observed data was correctly calculated and reported, the benchsheet incorrectly states the minimum required weight gain as 10 mg. The minimum weight gain allowed by the approved method is 2.5 mg. Choose sample volume to yield between 2.5 and 200 mg dried residue. This establishes a minimum reporting value of 2.5 mg/L when 1000 mL of sample is analyzed. Please submit a revised benchsheet. J. Finding: Filters are not weighed to constant weight prior to sample analysis, nor is a dry filter blank analyzed with each set of samples. Requirement: If pre-prepared filters are not used, the method requires that filters must be weighed to a constant weight after washing. Repeat cycle of drying, cooling, desiccating, and weighing until a constant weight is obtained or until weight change is less than 4% of the previous weighing or 0.5 mg, whichever is less. In lieu of this process, it is acceptable to analyze a single daily dry filter blank to fulfill the method requirement of drying all filters to a constant weight prior to analysis. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy based upon Standard Methods 20th Edition 2540 D. (3) (a). K. Finding: The samples are not weighed to constant weight, nor is an annual multiple weighing study to verify the adequacy of the drying time, performed. Requirement: Constant weights must be documented. The approved method requires the following: “Repeat the cycle of drying, cooling, desiccating, and weighing until a constant weight is obtained or until the weight change is less than 4% of the previous weight or 0.5 mg, whichever is less.” In lieu of this, an annual study documenting the time required to dry representative samples to a constant weight may be performed. Verify minimum daily drying time is greater than or equal to the time used for the initial verification study drying cycle. Drying cycles must be a minimum 1 hour for verification. Ref: North Carolina Wastewater/Groundwater Laboratory Certification Policy based on Standard Methods 20th Edition, 2540 D. (3) (c), 2540 B. (3) (b), and 2540 C. (3) (d). Page 6 #199 Archer Daniels Midland Company Comment: North Carolina allows for an annual drying study in lieu of the requirement above to repeat the drying cycle for every sample. A random full set of samples should be used for the drying study. The repeated drying time in the oven should be at least 1 hour long. The time used for the annual drying study is the minimum time that samples are to be dried until a new drying study is performed. COD – Standard Method, 5220 D – 1997 Comment: Only one calibration curve verification is required every 12 months. Each analyst is not required to perform a calibration curve verification when samples are being analyzed against a factory- set calibration curve. Comment: A new calibration curve verification would be required prior to the 12 month requirement if a new lot of tubes is put into use, the slope changes by more than 10% at the mid-point or a new stock standard is prepared. Comment: Two mid-range standards from separate sources are not required when samples are being analyzed against a factory-set calibration curve. Any standard would be considered to be from a second source when analyzed against a factory-set curve. Therefore, only one mid-range standard is required initially (i.e., prior to sample analysis), after every tenth sample and at the end of each sample group to check calibration drift. Fecal Coliform – Standard Method, 9222 C E – 2000 (MPN) L. Finding: Duplicate analyses are not analyzed. Requirement: Duplicate analyses – Perform duplicate analyses on 10% of samples and on at least one sample per test run. A test run is defined as an uninterrupted series of analyses. If the laboratory conducts less than 10 tests/week, make duplicate analyses on at least on sample each week. Ref: Standard Methods, 9020 B. (8) (4). BOD – Standard Method, 5210 B – 2001 M. Finding: Extra nutrient, mineral, and buffer solutions are not added to the Biochemical Oxygen Demand (BOD) bottles containing more than 67% (i.e., > 201 mL) sample. Requirement: When a bottle contains more than 67% of the sample after dilution, nutrients may be limited in the diluted sample and, subsequently, reduce biological activity. In such samples, add the nutrient, mineral, and buffer solutions (3a through e) directly to individual BOD bottles at a rate of 1 mL/L (0.30 mL/300-mL bottle) or use commercially-prepared solutions designed to dose the appropriate bottle size. Ref: Standard Methods, 5210 B-2001, (5) (c) (2). Comment: Dilution water prepared as normal with nutrient, mineral, and buffer solutions would still be used in the BOD bottles. For the bottles containing more than 67% sample, extra nutrient, mineral, and buffer solutions would be added in addition to the dilution water. Recommendation: Rather than pipetting 0.30 mL of each reagent into the BOD bottle, it is recommended that commercially prepared nutrient buffer pillows for 300 mL bottles be used. Dissolved Oxygen – Standard Methods, 4500 O G – 2001 N. Finding: Sample collection time is not documented. Page 7 #199 Archer Daniels Midland Company Requirement: Sample identification must be associated with the date and time of sample collection and analysis. The time elapsed between sampling and analysis must be documented to determine if hold times are met. Ref: North Carolina Wastewater/Groundwater Laboratory Certification memorandum dated June 20, 2007, “Required Documentation for Sampling Preservation and Hold Time”. Comment: The DO sample was being collected at the same time as the pH sample but the documentation did not indicate that. pH – Standard Methods, 4500 H+ B - 2000 Recommendation: It is recommended that the time of the meter calibration be documented. This will soon be required by North Carolina Wastewater/Groundwater Laboratory Certification Policy. O. Finding: The calibration check buffer is not documented to the tenths place to show that it is within the ± 0.1 Standard Units acceptance criterion. Requirement: If the meter response shows a difference greater than 0.1 pH unit from the expected value, look for trouble with the electrodes or the potentiometer. Ref: Standard Methods, 4500 H+ B.-2000 (4) (a). IV. PAPER TRAIL INVESTIGATION: No paper trail performed. The Wilmington Regional Office’s Linda Willis performed a paper trail investigation in June of 2012. V. CONCLUSIONS: Correcting the above-cited findings and implementing the recommendations will help this lab to produce quality data and meet certification requirements. The inspector would like to thank the staff for its assistance during the inspection and data review process. Please respond to all findings. Report prepared by: Todd Crawford Date: December 10, 2012 Report reviewed by: Chet Whiting Date: February 11, 2013