|Gossman Consulting, Inc|
Practical Quality Assurance/Quality Control in
the Commercial Thermal Treatment Facility Laboratory
The commercial thermal treatment facility laboratory is a radically different environment from that found in the typical commercial environmental lab. Many of the QA/QC procedures designed for normal environmental lab performance evaluations by EPA and others, require significant modifications to be usable in the fast turnaround, one sample at a time environment of the commercial TSD laboratory. An examination of practical QA/QC items and scenarios for implementation are discussed and contrasted with EPA guidance and ASTM standards.
There are different types of laboratories that one may consider when looking at analytical information gathering. Some laboratories are research directed, some are environmentally directed, some are process quality control directed. To an untrained observer, all types of labs obtain analytical information on a sample and therefore they may assume that the quality assurance and quality control (QA/QC) procedures should be the same. The generalization about analytical information gathering by this observer is essentially true, however the assumption about QA/QC does not take into account the very real differences in the purpose of that information. The difference in the purpose of the analytical information affects the environment of the laboratory, the analytical procedures and the QA/QC that is used.
An examination of the QA/QC procedures used in a commercial thermal treatment facility laboratory under the boiler and industrial furnace (BIF) regulations must begin with an understanding of the purpose of the lab. To some, this type of lab may be an environmentally directed lab; however, this is not completely true. This type of lab is actually a combination of the process quality control and the environmentally directed labs. It is heavily influenced by the need to be a process quality control lab given that the samples that it analyzes are of material which will be used in an industrial process. The environmentally directed aspects of the lab come from the regulations which pertain to most of the materials being analyzed by the lab. It is important therefore, to review the purpose, environment and procedures of the lab. Understanding differences in these areas support different QA/QC procedures for the treatment facility laboratory which allow the QA/QC to be both practical while meeting the criteria of a sound QA/QC program. A comparison of the treatment facility lab purpose with the purpose of a commercial environmental lab will aid in understanding the necessary differences in procedures.
The general purpose of a regular commercial environmental laboratory is to provide very specific quantitation information on trace amounts of environmental contaminants in a wide range of matrixes such as water, organic mixtures, and soils. Analytical information on quantity and type of contaminants are obtained using specific procedures such as those of EPA SW-846. These procedures are specific to the matrix type and the contaminant. The ultimate purpose of the information is to characterize the material as some category of hazardous or non-hazardous material as defined by the amount of contaminant as compared to the environmental regulation. Thus QA/QC procedures are specific to the trace levels and specific characterizations normally required by the regulations. Given that the only salable result of this type of lab operation is analytical information about a sample, many samples must be analyzed. The lab, in essence, must analyze a very large number of samples to achieve economies of scale.
The commercial BIF lab typically uses the analytical information to prepare a usable product that can be substituted for other materials in an industrial process. The analytical information is not the product as in a commercial environmental lab but primarily a tool to produce a product that meets a certain quality. The analysis is actually part of the quality control on the material not the end result. As such, the analysis is not generally concerned with trace levels but rather with the major components or characteristics which may affect the process. The commercial BIF lab will base some of their analysis and QA/QC procedures on SW-846 but because of the difference in purpose is not required to follow them as is the commercial environmental lab. The United States Environmental Protection Agency (USEPA) has, in essence, recognized these facts and the limits of SW-846 in the Preamble to the 3rd Edition of SW-846 in the Federal Register, February 8, 1990, which states:
"This notice, or the subsequent final rule, should not be construed to require the use of SW-846, Third Edition methods except where specifically prescribed by the regulation."
"Except for those situations where the RCRA regulations specify use of a particular method, it is appropriate for the chemist to use judgement, tempered by experience, in selecting an appropriate set of methods from SW-846 or the scientific literature for preparing and analyzing a given sample.
"Implicit in the proceeding argument is the fact the SW-846 was designated largely for use in showing that a waste does not contain certain hazardous constituents or characteristics. In that regard, many SW-846 sample preparation methods are designed around trace analysis rather than the percent level determinations often required for concentrated wastes. These methods, however, might be suitable for percent level determination analysis when appropriately modified by the analyst."
Some low level analysis is normally required to exclude some substances, but primarily the analytical quantitation levels are much higher than those required by a commercial environmental lab. In most cases the BIF lab is also not specifically trying to categorize the material as hazardous or non-hazardous from a regulatory perspective, as does a commercial environmental lab. Rather, the materials are normally expected to be hazardous. However, the analysis does seek to confirm some of the hazardous nature of the materials to ensure safe handling by process personnel. Thus QA/QC procedures are specific to analysis at normally well above trace levels and to process requirements of the material. As a result of the differences from SW-846, data quality objectives (DQO's) will likely be different from those of the commercial environmental laboratory. The DQO's will also be different because end use of the data is primarily process quality control.
The physical make-up of the commercial environmental laboratory is normally much different than the BIF lab because of a much larger number of samples to be run daily and the QA/QC procedures required for the specific regulatory analysis. Because of the large number of samples, sample matrixes and the accompanying number of QA/QC procedures, a larger number of duplicate analytical instruments and sample preparation equipment are necessary. Often the lab must be organized into specialized areas for different types of analysis where large numbers of samples are run along with relatively large numbers of QA/QC samples. This allows the lab to meet economies of scale as well as the specific regulatory requirements. The samples are normally only analyzed for a limited set of specific criteria rather than a broad base of parameters. While time is money, these labs are not normally under short turnaround pressure. Thus samples can be held, within regulatory limits, until several similar samples requiring the same analysis are received and all can be run through prep and analysis together to again meet economies of scale. The information derived from the analysis is normally not required to be used immediately. Thus data processing procedures, including QA/QC which are more time consuming, can be more easily accomodated.
By comparison a BIF lab is concerned with analysis of far fewer samples on a daily basis but analyses samples generally for many more parameters. Primarily, these parameters are not for determining levels of contaminants but for physical parameters which are not environmentally directed but process decision directed. Or they are directed for process limitations on material quality. In some cases, the limits are mandated by regulations or a permit but not the same regulations governing hazardous vs. non-hazardous determinations as in the commercial environmental lab. Because a primary purpose is not for the numbers of sample analysis but for process quality control information of material, the lab make-up is based on different needs than the commercial environmental lab. A quick turnaround analysis for process decision making, and single or small numbers of samples needing analysis at one time are important. Also, a broad range of analytical information for the various process parameters require the BIF lab, analytical methods and QA/QC procedures to be developed to meet that need. Unlike the commercial environmental lab, most of sample prep equipment and different instrumentation are located in the same general area. Analytical QA/QC procedures are designed around the less frequent individual runs to allow for a more practical quick turnaround result. In addition, the results do not need trace level detection limits and the accompanying trace level QA/QC as do the commercial environmental labs. Another key component for the BIF lab is that materials are normally analyzed at several points in the process and, in most cases, for exactly the same parameters so there are built-in duplication QA/QC procedures. Further, a prospective sample normally has accompanying documentation as to it source and components of interest prior to being analyzed. This is unlike a commercial environmental lab where source of the material is not a primary consideration and sample information is normally limited to actual sampling label information and chain of custody. This documented source information is taken into account by the BIF lab analyst in a quality evaluation of both the material and the resulting analytical information. A final significant difference is that materials which are sampled are normally pre-selected for the specific uses at the BIF facility and are normally similar in type and potential for that use. This selection process normally limits some of the variability in sample types that is often seen at a commercial environmental lab. In a BIF laboratory, environment and purpose together support QA/QC procedures which are practical as well as providing for quality information.
This section provides an examination of some of the practical QA/QC items and scenarios in use at a typical BIF facility. The examination will contrast those procedures with EPA guidance and ASTM methods.
Quality Assurance/Quality Control
The quality assurance and quality control program is designed to ensure that data produced from the sampling and analysis of the waste materials will accurately represent the parameters for which the waste is being tested. The program is based on meeting defined objectives through the use of standard operating practices and following the guidence of standard quality control procedures.
The quality assurance program objectives are to ensure that all safety, regulatory, and process requirements for the waste analysis are met. This is accomplished through the design of the facility, the training of all employees, the use of appropriate methods for sampling, analyzing, and processing materials entering and leaving the facility.
The quality control program is comprised of standard operating procedures (SOP's) for all activities that impact waste analysis and processing. These SOP's include procedures for waste sampling and analysis, chain of custody, routine quality control procedures, audit procedures, and record keeping. Chain of custody is comprised of proper logging, storage and disposal of samples. Routine quality control procedures ensure that daily analyses are of consistent high quality. The activities to ensure this include instrument optimizations, standardizations, performance check runs, trouble shooting analysis problems, and maintenance. Audits are performed both internally and externally. Internal audits include periodic duplicate sampling and analysis. External audits may include round- robin exchange of samples with another lab and the analysis of standard reference materials prepared specifically for the audit. Record keeping is comprised of periodic evaluations of all sampling, analysis, process, maintenance, and quality control records. The evaluations are to ensure that the established procedures are followed and ensure that the quality assurance objectives are being met.
1. Quality Control Procedures:
The laboratory uses standard Quality Control procedures as part of an overall Quality Control Assurance Program. These Quality Control procedures specify that the performance of the operations are to be challenged and evaluated using control samples with known characteristics and actual samples as received. Each analytical procedure uses the following QC checks, where applicable:
(a.) Calibration and Reagent Standardization - Each time an instrument is calibrated or a reagent is standardized, a record must be kept of the results. The quality control assurance program objectives specify the procedure and frequency required to maintain accuracy.
(b.) Known Standards - If a method is not calibrated frequently, a known standard will be analyzed on a regular basis.
(c.) Blanks - Where applicable, methods blanks will be run each analysis day and the results recorded.
(d.) Duplicates - A duplicate sample will be run daily or weekly as indicated.
(e.) Blind Standards - Where available, a blind standard will be analyzed every three months. A blind standard will contain a known compound at a concentration which is unknown to the analyst.
(f.) Others - When applicable spikes, internal standards, laboratory control standards, etc. may also be utilized.
2. Quality Control Checks Summary:
Quality Control Frequency varies from every sample, to daily, weekly, monthly and quarterly depending on the parameter and method. To a large extent, these are very similar from method to method, as shown in a brief summary on Table 1, from some of the more frequently used methods.
3. Data Quality Objectives:
Data quality objectives are qualitative and quantitative statements which specify the quality of the data required to support decisions related to the various options of waste acceptance and treatment at the facility. DQO's are determined based on the end use(s) of the data to be collected. They are established prior to data collection during the formulation of standard operation procedures and are not treated as a separate consideration. Rather, the DQO development process is integrated with the project planning process that functions within the quality control assurance program previously outlined. The results are incorporated into the sampling and analysis plans and written SOP's.
The DQO development process results in a well-thought-out sampling and analysis plan which details the chosen sampling and analysis options and statements of the confidence in decisions made using this data. Confidence statements are possible through the application of statistical techniques to the data.
DQO's are established to ensure that the data collected is sufficient and of adequate quality for its intended uses. Data collected and analyzed in conformance with the DQO process described below can be used in assessing the uncertainty associated with a given decision. The term "uncertainty" is used in a broad sense to describe the likelihood of all types of errors associated with a particular decision.
Comparing some of the BIF lab's analytical parameters and QA/QC to those of the referenced methods illustrates some of the differences. Table 2 gives a condensed view of results of that comparison. However, examining some the reasons behind the differences can lead to a more complete understanding.
For the organic composition and polychlorinated biphenyl (PCB) parameters, some of QA/QC differences such as fewer matrix spikes and calibration checks are performed because the quantitation levels for the BIF lab's purposes is normally 0.1% for organics and 50 ppm for PCBs. These levels are at a minimum 10 to 100 times higher than the quantitation levels for the methods. Some addition QA/QC procedures, such as blind standards, are added to the BIF program and thus further the support for the procedure's accuracy.
The halogen and sulfur parameter similar to the organic determinations has much higher levels of reporting than the methods produce. Normally, 0.1% is the minimum reporting level for these elements and that is again 10 to 100 times high than the method will achieve. Thus, a need to account for a matrix effect is negligible. Also, it can be noted that the reference method calls for a calibration check with every ten runs which can be close to or even more than a Treatment Storage and Disposal Facility (TSDF) lab may run in one day.
Several of the other parameters such as viscosity, density, ash and heat content are test methods to obtain information important primarily for processing decision making. The differences found are not as significant because the reference methods allow considerable flexibility.
It is clear that the requirements to provide a quality product for use in the BIF facility direct the type of analytical procedures and therefore the QA/QC procedures used in this type of TSDF lab. To ensure that product quality, meets both environmental regulation and process requirements for timeliness and accuracy, practical procedures for QA/QC have been developed. These procedures provide the necessary assurance that materials are being properly handled, prepared and consumed to make use of their inherent value in an industrial process.
1. ASTM Standards as Published by the American Society for Testing Materials, 1916 Race Street, Philadelphia, PA various standards and dates.
2. EPA SW-846 Third Edition; U.S. Environmental Protection Agency, Office of Solid Waste and Emergency Response, Washington, D.C.
3. Gossman D., "A Review of the Usefulness of Various ASTM and SW-846 Methods Which May Be Used by the Thermal Treatment Industry", Presented at the Air & Waste Management Association, Boiler and Industrial Furnace Conference, March 1995, Kansas City, MO.
4. Gossman D., "Quality Control of Hazardous Waste Fuel", presented
at the 30th Institute of Electrical and Electronics
Engineers Cement Industry Technical Conference, May 1988, Quebec
Table 1. Typical Frequency for Quality Control Checks at a TSDF
|Heat Content (BTU/lb.)||none||none||yearly||weekly||weekly|
|Halogens & Sulfur||quarterly||daily||daily||weekly||weekly|
|Specific Gravity (Density)||none||none||none||none||weekly|
Table 2. Comparison of QC Frequency for Process Quality Control Versus Reference Methods
Process Quality Control
|quarterly||daily||yearly||weekly||daily||SW-846 8260, 8270B||Blind Standards-N/A;
Matrix spike-as needed
|PCBs||quarterly||daily||bi-weekly||daily||daily||SW-846 8081, ASTM D4059||Blind Standards- N/A;
Calibration- as needed;
Matrix spike- as needed
|Heat Content (BTU/lb.)||none||none||yearly||weekly||weekly||ASTM D 5468||Standards- monthly;
Analytical balance accuracy- weekly
N/A; Duplicates- each sample;
Matrix spikes- every 20 samples;
Calibration check- 10 runs
|Ash||none||none||none||none||weekly||ASTM D5468||Analytical balance accuracy- weekly|
|Viscosity||none||none||none||weekly||weekly||ASTM D2196||Calibration check- as needed|
|none||none||none||none||weekly||ASTM D5057||Standards- as needed;
Duplicates- as needed
|Radioactivity||none||daily||none||weekly||daily||ASTM D5928||Standard check- daily;
Calibration (factory)- yearly
|Metals:||quarterly||daily||weekly||daily||daily||ASTM D4326||Blind Standards- N/A|
|Compatibility||none||none||none||none||weekly||ASTM D5058||Calibration (thermometers)- as needed|
|Reactivity||none||weekly||none||weekly||weekly||ASTM D5058||Calibration (thermometers)- as needed|
SW-846 9045, 9040
|Check standards- as needed:
Duplicates- as needed