Quality Assurance

B. M. Gillespie and B. P. Gleckler

Quality assurance (QA) and quality control (QC) practices encompass all aspects of Hanford Site environmental monitoring and surveillance programs. Samples are analyzed according to documented standard analytical procedures. Analytical data quality is verified by a continuing program of internal laboratory QC, participation in interlaboratory cross-checks, replicate sampling and analysis, submittal of blind standard samples and blanks, and splitting samples with other laboratories.

QA/QC for ground-water environmental surveillance also includes procedures and protocols for 1) documenting instrument calibrations, 2) conducting activities in the field and laboratory, 3) maintaining wells to ensure representative samples are collected, and 4) using dedicated sampling pumps to avoid cross-contamination.

This section discusses specific measures taken to ensure quality in project management, sample collection, and analytical results.

Environmental Surveillance

Comprehensive QA programs, including various QC practices, are maintained to ensure the quality of data collected through the surveillance programs. QA plans are maintained for all surveillance activities, defining the appropriate controls and documentation required to meet the guidance of the American Society of Mechanical Engineers (ASME) NQA-1 QA program document (U.S. nuclear industry's standard, ASME 1989) and DOE Orders.

Project Management Quality Assurance

Site surveillance and related programs, such as processing of TLDs and performing dose calculations, are subject to an overall QA program. This program implements the requirements of Richland Operations Office Order DOE 5700.6C, "Quality Assurance," and is based on ASME NQA-1, Quality Assurance Program Requirements for Nuclear Facilities (ASME 1989). The program is defined in a QA manual (PNL 1992c), which provides guidance for implementation by addressing the following 18 QA elements. These 18 elements are:

  1. Organization
  2. Quality Assurance Program
  3. Design Control
  4. Procurement Document Control
  5. Instructions, Procedures, and Drawings
  6. Document Control
  7. Control of Purchased Items and Services
  8. Identification and Control of Items
  9. Control of Processes
  10. Inspection
  11. Test Control
  12. Control of Measuring and Test Equipment
  13. Handling, Storage, and Shipping
  14. Inspection, Test, and Operating Status
  15. Control of Nonconforming Items
  16. Corrective Action
  17. Quality Assurance Records
  18. Audits.
The environmental surveillance projects have current QA plans that describe the specific QA elements that apply to each project. These plans are approved by a QA organization that conducts surveillances and audits to verify compliance with the plans. Work performed through contracts, such as sample analysis, must meet the same QA requirements. Potential equipment and services suppliers are audited before contracts are awarded for services or the purchase of materials are approved, which could have a significant impact on quality within the project.

Sample Collection Quality Assurance/Quality Control

Environmental surveillance samples were collected by staff trained to conduct sampling according to approved and documented procedures (PNL 1992a). Continuity of all sampling location identities is maintained through careful documentation. Field duplicates are collected for specific media, and results are addressed in the individual media sections of 5.0, "Environmental Surveillance Information."

Samples for ground-water monitoring are collected by trained staff according to approved and documented procedures (PNL 1993). Chain-of-custody procedures are followed (EPA 1986a) that provide for the use of evidence tape in sealing sample bottles to maintain the integrity of the samples during shipping. Full trip blanks and field duplicates were obtained during field operations. Summaries of the 1994 results are provided in Tables 7.0.1 and 7.0.2.

Analytical Results Quality Assurance/Quality Control

Routine hazardous and nonhazardous chemical analyses for environmental and ground-water surveillance water samples are performed by DataChem Laboratories, Inc., Salt Lake City, Utah. The laboratory participates in the EPA Water Pollution and Water Supply Performance Evaluation Studies. DataChem Laboratories maintains an internal QC program that meets the requirements of EPA SW-846 (EPA 1986a), which is audited and reviewed. PNL submits additional QC double-blind spiked samples for analysis.

Routine radiochemical analyses for environmental and ground-water surveillance samples are performed by International Technology Corporation's (IT) Richland Laboratory. The laboratory participates in the DOE's Quality Assessment Program and the EPA's Laboratory Intercomparison Studies. An additional QC blind spiked sample program for each project is conducted. IT's Richland Laboratory also maintains an internal QC program, which is audited and reviewed. Additional information on these QC efforts is provided in the following subsections.

U.S. Department of Energy and U.S. Environmental Protection Agency Comparison Studies

DataChem Laboratories participated in the EPA Water Pollution and Water Supply Performance Evaluation Studies. Standard water samples were distributed blind to participating laboratories. These samples contained specific organic and inorganic analytes with concentrations unknown to the analyzing laboratories. After analysis, the results were submitted to EPA for comparison to known values and other participating laboratory concentrations. Summaries of the results during the year are provided in Table 7.0.3. Approximately 97% of the results during the year were within the typically used "3-sigma control limits" (3 standard errors of the mean).

The DOE Quality Assessment program and EPA's Intercomparison Studies Program provide standard samples of various environmental media (water, air filters, soil, and vegetation) containing specific amounts of one or more radionuclides that are unknown by the participating laboratory. After sample analysis, the results were forwarded to DOE or EPA for comparison with known values and results from other laboratories. Both EPA and DOE have established criteria for evaluating the accuracy of results (Jarvis and Siu 1981, Sanderson 1985). Summaries of the 1994 results for the programs are provided in Tables 7.0.4 and 7.0.5. Approximately 83% of the results during the year were within the typically used "3-sigma control limits" (3 standard errors of the mean).

Pacific Northwest Laboratory Evaluations

In addition to DOE and EPA interlaboratory QC programs, a QC program is maintained by PNL to evaluate analytical contractor precision and accuracy and to conduct special intercomparisons. This program includes the use of blind spiked samples and replicate samples. Blind spiked QC samples and blanks were prepared and submitted to check the accuracy and precision of analyses at DataChem Laboratories and IT's Richland Laboratory. In 1994, blind spiked samples were submitted for air filters, vegetation, soil, water, and ground water. Overall, 81% of the DataChem Laboratories blind spiked determinations were within control limits and 93% of IT's Richland Laboratory blind spiked determinations were within control limits (Table 7.0.6 and 7.0.7). This indicates, overall, acceptable results.

PNL also participates in a Quality Assurance Task Force, a program conducted by the DOH. Two Hanford Site ground-water samples were analyzed in triplicate, and a National Institute of Standards and Technology (NIST) reference soil sample was analyzed in duplicate or triplicate by up to six other participating laboratories in the Pacific Northwest (Table 7.0.8). Well 199-N-56 was analyzed for cobalt-60 and tritium concentrations. PNL data compared favorably with those of other participating laboratories. Well 399-1-16A was analyzed for tritium, uranium isotopics, and total uranium concentrations. PNL concentrations compared favorably within the statistical standard deviations of the mean of the other participating laboratories. The NIST reference soil sample was analyzed for cesium-137 and strontium-90 concentrations. PNL's results fell within the of the mean of the concentrations of the other participating laboratories and were acceptable.

Laboratory Internal Quality Assurance Programs

The DataChem Laboratories and IT's Richland Laboratory are required to maintain an internal QC program. Periodically, the laboratories are internally audited for compliance to the QC programs. At the DataChem Laboratories, the QC program meets the QC criteria of EPA SW-846 (EPA 1986a). This program also requires the laboratory to maintain a system for reviewing and analyzing the results of the QC samples to detect problems that may arise from contamination, inadequate calibrations, calculation errors, or improper procedure performance. Method Detection Level determinations are performed semiannually.

IT's Richland Laboratory internal QC program involves routine calibrations of counting instruments, yield determinations of radiochemical procedures, frequent radiation check sources and background counts, replicate and spiked sample analyses, matrix and reagent blanks, and maintenance of control charts to indicate analytical deficiencies. Available calibration standards traceable to the NIST were used for radiochemical calibrations. Minimum detectable concentration verification is conducted (when requested) for radionuclide-media combination analyses. Equation 37 from Chapter 6 in EPA 520/1-80-012 (EPA 1980a) is used in the minimum detectable concentration calculations, which involves the use of factors such as the average counting efficiencies and background for detection instruments, length of time for background and sample counts, sample volumes, radiochemical yields, and a predesignated uncertainty multiplier.

In 1994, one inspection of the DataChem Laboratories and one inspection of IT's Richland Laboratory were conducted. These inspections documented conformance with contractual requirements of the analytical facility and provided the framework for identifying and resolving potential performance problems. Responses to audit and inspection findings were documented by written communication, and corrective actions were verified by follow-up audits and inspections.

Internal laboratory QC program data are summarized by the laboratories in quarterly reports. The results of the QC sample summary reports and the observations noted by each laboratory indicated an acceptably functioning internal QC program.

Media Audits and Comparisons

Additional audits and comparisons are conducted on several specific types of samples. The DOH routinely cosampled various environmental media and measured external radiation levels at multiple locations during 1994. Media that were cosampled with the DOH included: 25 ground-water wells; two Columbia River sites; one riverbank spring; one on-site pond; two offsite water systems; three Columbia River sediment sites; five air monitoring stations; 15 TLD sites; offsite onions, potatoes, clams, and milfoil. Also cosampled were upwind and downwind samples of eggs, alfalfa, asparagus, carrots, peaches, apples, pheasant, chicken, and wine. These data will be available in the DOH report, Environmental Radiation 1994 Annual Report, 33rd Edition, when published. The National Food and Drug Administration also co-sampled vegetables, fruit and wheat. The data are presented in Table 7.0.9.

Quality Control for environmental TLDs includes the audit exposure of three environmental TLDs per quarter to known values of radiation (between 17 and 28 mR). A summary of 1994 results is shown in Table 7.0.10. On average, the TLD measurements are biased only 0.04% higher than the known values.

Table 7.0.1

Table 7.0.2

Table 7.0.3

Table 7.0.4

Table 7.0.5

Table 7.0.6

Table 7.0.7

Table 7.0.8

More