An Analysis of Selected NOD Issues Related to Waste Analysis Plans
In 1994 EPA issued Waste Analysis at Facilities That Generate, Treat, Store, and Dispose of Hazardous Wastes, a Guidance Manual(1) . Last year Gossman Consulting, Inc. published "A Model Waste Analysis Plan for Commercial BIF Facilities". During the last year, EPA regions and states have issued a number of NODs relative to WAP's submitted as part of permit applications. These NODs suggest a lack of "real world" familiarity with lab operations at commercial TSDs on the part of the permit reviewers. Selected NOD issues are examined relative to this issue and the afore-mentioned publications.
The boiler and industrial furnace (BIF) rules became effective in 1991. Since that time a lot of attention has been given to performing required compliance testing, Certification of Compliance testing in 1992, Recertification of Compliance testing in 1995 followed by Trial Burn testing or another round of Recertification of Compliance testing. The submission of Resource Conservation and Recovery Act (RCRA) Part B permits has gone hand-in-hand with Trial Burns if the permit had not already been submitted previously. A required portion of the RCRA Part B permit submittal includes a waste analysis plan (WAP). Waste analysis plans figure heavily into the analysis of the hazardous waste processed at any BIF facility. Regulatory progress towards BIF permitting has resulted in increased scrutiny of WAP's. This paper examines United States Environmental Protection Agency (USEPA) review and requests associated with waste analysis plans. The ideal versus the practical.
THE WASTE ANALYSIS PLAN
Regulatory requirements for waste analysis plans are found in 40 CFR §264.13 as part of the Standards for Owners and Operators of Hazardous Waste Treatment, Storage, and Disposal Facilities (TSDFs). These requirements allow use of generator knowledge/information of waste in receipt and processing of that waste. A significant issue associated with a waste analysis plan, other than meeting specific regulatory requirements, is that once the WAP is finalized then the facility is automatically out of compliance at any time the WAP is not followed to the letter. This makes the process of agreeing to a finalized WAP through interaction with the appropriate regulatory agency of utmost importance.
Gossman Consulting, Inc. (or GCI personnel) have been involved in the development and or implementation of waste analysis plans since the late 1970s. Since the advent of the BIF rules GCI has been involved in modifying WAPs to meet regulatory needs and/or interacting with regulatory officials in getting those WAPs finalized(2) . The issues presented in this paper are a result of participation in a series of WAP comments and responses over the past several months for a number of BIF facilities in the final stages of permitting.
Only Regulated Activities
Many waste analysis plans were originally written to detail routine analysis at a given facility. Seemingly innocuous activities from a regulatory standpoint were included in these waste analysis plans because they were important to the facility, not a compliance issue but rather a practical matter. It has become clear that if it is not an activity directly related to regulatory compliance then it is best not to include it as part of the WAP. On the one hand that could seem odd since the activity may be an integral part of day-to-day operation of the facility, but on the other hand, it is better not to confuse WAP compliance issues.
It has also been found that many regulators ideally desire that analysis for such parameters as dioxins in waste material be included as part of the day-to-day waste analysis at a facility. This is an issue on which much could be written, however, the key issue is that dioxin analysis is a time consuming analysis and simply cannot be performed on a load receipt or even a blended tank basis. The other primary dioxin analysis issue is whether or not dioxin analysis on waste materials need be performed at all. The BIF regulations effectively discourages the burning of what the regulations term "dioxin-listed waste" in 40 CFR §266.104(a)(3) which are F020, F021, F022, F023, F026 and F027. The generator determines the appropriate waste code(s), not the TSDF. The primary method of keeping these types of waste out of the facility is by performing a simple hazardous waste code screen with the manifest of the arriving load. Discovery of even one of these codes on an arriving manifest would then result in rejection. Hazardous waste code screening, as recommended here, versus actual waste analysis, as usually expected by regulators, raises a basic conundrum of waste analysis.
Waste Code Screening vs. Waste Analysis
There is a general concept floating around out there that there should be analytical methods/activities for any and all waste that the facility receives. This is simply not true and detection of a given compound does not and cannot always confirm that a certain type of waste is present. As an example, consider that the waste code for methyl bromide, U029, is not on the facility Part A and is therefore not an approved compound. Then a routine analysis turns up detectable quantities of methyl bromide. Does this mean that the generator is trying to slip something past? Not if the Part A lists K131 or K132 which also contain methyl bromide. Therefore, much waste prohibition must be addressed through manifest screening of an arriving load. This point often generates lengthy discussion/argument with regulators.
Waste Code Screening vs. Generator Certifications
Another point of contention in waste screening involves compliance with 40 CFR §268.3(c)(3) , dilution prohibited as a substitute for treatment, specifically §268.3(c) combustion of hazardous waste codes listed in §268 Appendix XI, metal bearing wastes prohibited from dilution in a combustion unit. Some regulators have initially viewed this portion of the regulation to mean that these waste codes cannot be combusted by a BIF, when in fact, the regulation makes it quite clear that one of six listed exceptions apply and that this determination must be made at the "point of generation." Once again, generator determination is critical rather than waste analysis at the point of receipt. The most effective way to make this determination is with an appropriate generator certification.
How Often Is "As Often As Necessary?"
40 CFR §264.13(a)(3) states that analysis must be repeated as often as necessary to ensure that analysis is accurate and up to date. Analysis is to be repeated when it is determined that the waste has changed or does not match the accompanying waste manifest. Generally, facilities analyze waste in some fashion upon arrival. However, this is not an absolute regulatory requirement unless it is written into the waste analysis plan. In addition, most facilities comply with the "often as necessary" portion of §264.13 by performing an annual analysis or obtaining a generator certification. There are many interpretation issues which become resolved as part of the comment and response activities associated with finalizing a waste analysis plan.
Is A "Standard Operating Procedure" Part Of The Waste Analysis Plan?
Facilities are required to have a waste analysis plan as part of their RCRA Part B. However, most facilities have a standard operating procedures manual which is either a stand alone document or is part of a laboratory quality assurance/quality control document. The SOP is not an official part of a facility waste analysis plan. Nowhere in §264.13 is there any mention of standard operating procedures. Standard operating procedures are always available for inspection upon request and some facilities have even submitted copies of their SOPs to regulators as a file reference document. However, SOPs are not a permit requirement.
Discussion about SOPs will invariably lead into quality control (QC) issues and data quality objectives (DQOs). These are also integral parts of a well designed analytical program. However, they should not be included in the permit waste analysis plan. Again, always available for inspection and maybe even submitted as reference, but not part of the final permit WAP. SOPs and DQOs are much too likely to change over the ten year life of a permit for them to be included in the WAP.
SW-846 Versus ASTM
There are times when insistence upon usage of SW-846 methodologies becomes an issue. As noted in the preamble of the third edition of the SW-846 methods, methodologies presented in SW-846 may at times need to be "modified by the analyst"(4) . Clearly, SW-846 methods can sometimes only be used as an analytical reference and the actual SOPs will likely reflect appropriate modifications including ASTM adaptations.
Analysis Of Streams Other Than Hazardous Waste
An important issue in a finalized WAP is frequency of analysis for raw material inputs and wasted cement kiln dust (CKD). Obviously, raw material inputs come into play with BIF feedstream limits and CKD must be demonstrated to be in compliance with 40CFR §266.112. An appropriate sampling plan is the key to justifying the frequency of analysis chosen by the facility.
Ultimately, the finalized waste analysis plan that ends up in the permit determines waste analysis compliance issues. Just as importantly, flexibility to modify the standard operating procedures as better analytical techniques become available is important. This is achieved by keeping SOPs out of the WAP. It is also important that the WAP be practical and functional on a day-to-day operational basis. It is critical to work with regulators so that the final waste analysis plan is not unduly restrictive and meets basic regulatory compliance issues.
1. Waste Analysis at Facilities That Generate, Treat, Store, and Dispose of Hazardous Wastes - A Guidance Manual; U.S. Environmental Protection Agency, Research Triangle Park, NC, 1994; EPA-OSWER 9938.4-03.
2. Gossman, David; Woodford, Jim, "A Model Waste Analysis Plan for Commercial BIF Facilities", In Waste Combustion in Boilers and Industrial Furnaces conference; Air & Waste Management Association; St. Louis, MO, April, 1997, pp 124-142.