John P. Schmuck
Morrison Knudsen Corporation

In conjunction with remediation of an historic disposal trench at the Rocky Flats Environmental Technology Site, an innovative approach was used to manage the excavated drum carcasses. The solvent contaminated drum carcasses were subjected to thermal desorption and a Determination of Equivalent Technology (DET) was obtained from EPA Headquarters. The DET allowed the drum carcasses to be disposed as low level waste rather than mixed waste, a change that yielded significant cost savings.

As a result of the exercise, two conclusions were reached. First, the hazardous debris treatment rules offer opportunity for substantial flexibility and possible cost savings. The self-implementing regulations should be used wherever appropriate. In contrast, wherever approval from EPA Headquarters is required, it is necessary to balance the cost savings against the time and energy required to obtain approval.

As a second conclusion, it was discovered that taking more sample "increments", compositing the increments and then performing a single analysis provides several benefits over collecting multiple samples and performing multiple analyses. The benefits include cost, equivalent accuracy and the ability to compare an average value, as opposed to an 80% upper confidence limit to the regulatory threshold.

During late 1995 and 1996 Rocky Mountain Remediation Services (RMRS) performed two Environmental Restoration cleanup actions at the Rocky Flats Environmental Technology Site (RFETS). The first action involved the excavation, treatment and offsite disposal of VOC contaminated soil and debris from Ryan's Pit. Ryan's Pit was used from 1966 to 1970 for disposal of a variety of wastes and debris including waste solvents and solvent still bottoms. As a result, portions of the soil and debris were either RCRA characteristic or contained RCRA F001 listed hazardous waste.

The second action also involved the excavation, treatment and disposal of VOC contaminated soil and debris, from trenches designated as T3 and T4. T3 and T4 were used from 1964 to 1967 for disposal of sanitary sewage sludge contaminated with low levels of plutonium and uranium, along with crushed drums and unknown quantities of solvent contaminated oils. Similar to Ryan's Pit, portions of the T3 and T4 soil and debris were either RCRA characteristic or contained RCRA F001 listed hazardous waste.

The actions were conducted under decision documents prepared pursuant to the Rocky Flats Cleanup Agreement (RFCA) . RFCA acts as an interagency agreement/consent order that integrates EPA CERCLA authorities and Colorado RCRA Corrective Action authorities at RFETS. The RFCA decision documents provided joint authority to excavate, stage, treat, and disposition these RCRA remediation wastes.

Low temperature thermal desorption was the technology selected to treat the excavated VOC-contaminated soil and debris. A mobile thermal desorption unit was brought onsite and the soil and debris were treated together in batches.

Treated soil that met the risk-based performance goals for hazardous constituents and dose-based performance goals for radionuclides was returned to the excavation. Treated soil that did not meet the dose-based performance goals for radionuclides but did meet the Land Disposal Restrictions (LDRs) was disposed offsite as LDR-compliant mixed waste.

Most of the debris consisted of drum carcasses. The balance of this paper discusses the strategy used to dispose the treated drum carcasses.

The alternative treatment standards for hazardous debris found in Table 1 of 40 CFR §268.45 were evaluated as a means of treating the hazardous debris, to allow land disposal. The evaluation revealed that the debris could be treated so that it is no longer subject to management as a RCRA hazardous waste. In that case, the debris could be disposed as low-level waste. The cost for disposal of low-level waste is one-third the cost for disposal of mixed waste.

Successful use of the alternative treatment standards for hazardous debris appeared to have other benefits. Because the alternative treatment standards for debris are technology based, the standards are intended (by EPA) to be less sampling intensive. (See 57 FR 37238, middle column, bottom). In addition, if the debris is treated to be non-hazardous, the Nevada Test Site (NTS) has higher radioisotope limits.

The debris treatment rules were promulgated on August 18, 1992. The new rule allowed the generator:

• to treat to the existing LDRs for the underlying hazardous wastes;
• to treat using specified technologies; or
• to obtain a contained-in determination from the EPA regional administrator.

Treating to existing LDRs was not new. That option requires that the generator comply with all of the existing LDRs. (See 40 CFR §268.45(a)).

Treatment of hazardous wastes using specified technologies was not wholly new, but had four facets unique to debris. First, many of the specified technologies were unique to debris. Some of the specified technologies are self-implementing while others require agency approval in the form of an "Equivalent Technology" approval. (See 40 CFR §268.45, Table 1 and §268.42(b)). The approval process is also called a Determination of Equivalent Technology (DET). "DET" will be used for the remainder of this paper.

Second, if the generator decided to apply a specified technology to characteristic hazardous debris, the generator was not required to address underlying hazardous constituents. Instead EPA introduced the concept of "contaminants subject to treatment" and limited the contaminants subject to treatment to either the "EP constituents for which the debris exhibits the TC toxicity characteristics" or the constituents for which treatment standards are established under 40 CFR §268.40. (See 40 CFR §268.45(b)(1) and (b)(2) respectively). This is in contrast to non-debris characteristic wastes that must attain the universal treatment standards for underlying hazardous constituents. (See 40 CFR §268.40(e)).

Third, if the specified technology is an extraction or destruction technology, and the debris does not exhibit a hazardous characteristic, the debris is no longer hazardous waste. (See 40 CFR §268.45(c)). This is the first codified example of delisting by rule. As noted earlier, some of the specified technologies are also self implementing and do not require intervention or approval by the regulatory authorities.

Fourth, application of the specified technology is intended to eliminate or minimize the need to perform sampling. Because the debris standards are technology specific, analysis of the debris is generally not necessary unless knowledge is not available to determine the debris does not exhibit a hazardous waste characteristic. (See 57 FR 37238, middle column, bottom).

The most unique attribute of the debris rule was EPA's decision to regulate debris as a solid waste which "contains" RCRA hazardous waste. In contrast to the mixture rule or derived-from rule for process waste, the debris rule in the only place that EPA has codified the "contained-in policy". (See 57 FR 37225, right column, bottom and 40 CFR 261.3(f)(2)).

The authority for a risk-based contained-in determination for debris is available and in many instances has been delegated to individual states. If the EPA or an authorized state determines, with or without treatment, that the debris no longer contains hazardous waste, then the debris is no longer subject to the RCRA hazardous waste requirements. In effect, this is a "delisting" if the debris contained listed hazardous waste.

The definition of debris is broad. Based upon a visual determination, as long as greater than 50% of the waste, by volume, has a particle size over 60 millimeters it meets the definition of debris. Neither process wastes or intact drums capable of holding greater than 75% of their contents are within the definition. (See 40 CFR §268.2(g) and 57 FR 37222).

Debris treatment can be performed in permitted tank, containers, containment buildings. In addition unpermitted 90-day generator debris treatment is allowable. (See 57 FR 37242).

RMRS recognized that the thermal desorption being use to treat the VOC contaminated soil and debris was an extraction technique listed in Table 1 of the debris rule. As a result, application of the alternate debris treatment standards to the treated drum carcasses could render the debris non-hazardous, require less sampling and substantially lower the disposal cost. The Waste Acceptance Criteria at the NTS is more flexible and requires less sampling than Envirocare. The cost per cubic foot for disposal of low level waste is approximately one-third the cost for mixed waste. In addition, NTS accepts higher levels of radioactivity.

At the outset, it was necessary to identify which regulator held the authority to make the determination. This was not an easy task, and it became evident that "DETs" were not being widely used by the regulated community. After a number of calls to state agencies, EPA Region VIII, the RCRA Hotline and EPA Headquarters, RMRS determined that the DET authority was held exclusively by EPA Headquarters and had not been delegated to the either the EPA regions or states.

RMRS was somewhat discouraged at the prospect of obtaining an approval from a distant agency with no familiarity with the project. For that reason RMRS began exploring the possibility that a more local authority might consider making a contained-in determination with potentially the same result.

The inquiries were very interesting. There exists tension between the EPA Regions and EPA Headquarters and between the States and EPA Headquarters on the issue of program delegation. Both the States and EPA Regions were disgruntled and felt that authorities could be more readily applied by local staff that are familiar with the project conditions. In fact, after conferring with his associates, one regulator said that his state would not consider granting a contained-in determination in lieu of a DET. He said that the only way to convince EPA Headquarters to relinquish the debris DET authority was to punish them with numerous requests for approvals.

Obtaining the DET from EPA Headquarters proved challenging. No guidance exists and no procedure or schedule is provided in the rule. Each DET is evaluated and approved on a case by case basis. The importance of good relations with the local regulatory authority cannot be underestimated. EPA Region VIII support was invaluable in conveying information to EPA Headquarters and in convincing EPA Headquarters that the project was conducted in an effective and compliant manner.

Initially, the petition consisted of a letter that briefly described the project and summarized the sample results from the treated soil. Deep in the Federal Register preamble to the final debris rule is direction to use the treated soil as a surrogate for debris sampling. (See 57 FR 37231, middle column, center). Using treated soil as a surrogate avoids the problems associated with sampling debris - issues like heterogeneity and the physical problems encountered when sampling. EPA reasoned that samples demonstrating successful soil treatment (e.g. treatment of soil to meet LDRs) can be used as a surrogate to demonstrate successful debris treatment. Consistent with SW846, the 80% upper confidence limit (UCL) of the treated soil data was compared to the regulatory threshold for each volatile contaminant subject to treatment. (See SW846, Part III Sampling, Chapter 9, Section 9.1.1.1, page NINE-6).

To obtain the DET, RMRS ultimately provided EPA Headquarters, Office of Solid Waste, with substantial project documentation. For example, the rule requires that any soil be separated from the debris "by reasonable physical means". (See 40 CFR §268.45, Table 1, footnote 9). In the context of thermal desorption this requirement is somewhat irrelevant because the treated soil becomes a fine powder and does not adhere to the debris to any significant degree. For individuals (including EPA Region VIII) familiar with the project, satisfaction of this element of the rule was clear. Convincing EPA Headquarters was another matter. In the end, significant project documentation (decision documents, completion reports, sampling and analysis plans etc.) had to be provided to build EPA Headquarters' confidence.

EPA Headquarters issued the DET with several conditions. DOE was required to make several certifications, including that the debris did not exhibit a hazardous characteristic. In addition, the authority to close out the DET was delegated to EPA Region VIII. Following discussion with EPA Region VIII, it was determined that knowledge of process was sufficient and that sampling for ignitability, corrosivity or reactivity was not warranted. Sampling was performed to demonstrate that the drum carcasses did not exhibit a Toxicity Characteristic.

When planning the sampling, an interesting attribute of the RCRA waste analysis requirements was uncovered. A question arose as to the number of samples required, from a statistical perspective, to obtain an accurate result. Recent attendance at a seminar on sampling analysis had highlighted the fact that it is the number of physical increments that comprise the sample that determine the accuracy of the result.

For comparison, suppose one sampling plan directs collection of ten increments which are then properly composited and a single analysis is performed. Alternatively, a second sampling plan directs collection of ten individual samples from the same waste stream. Under the second plan the samples are not composited and instead ten analyses are performed.

In theory, if the ten values generated by the second plan are averaged, the result should be the same as the single result generated by the first plan. In effect, proper compositing in the first plan creates a result that reflects the "average" of the waste stream. Sample handling aside, since ten increments were taken under both plans, both results should have very similar accuracy.

The difference between the two plans becomes more evident when the analytical results of each are handled consistent with SW846. As noted earlier, SW846 requires use of an 80% UCL when comparing a suite of data to a regulatory threshold. Under the first plan, when you composite the ten increments and perform a single analysis you are comparing an average value to the regulatory threshold. In contrast, under the second plan when you take ten samples and analyze them separately you are penalized and required to compare a less favorable 80% UCL to the regulatory threshold. In addition, the second plan will result in higher analytical costs.

Ultimately, the treated non-hazardous drum carcasses were disposed at NTS. NTS seemed willing to accept EPA Headquarters' authority without question.

Several conclusions can be drawn. Without question, the self implementing extraction or destruction technologies can be applied to delist debris containing listed hazardous wastes. Examples of off the shelf technologies include spalling, high pressure steam or water washing, and water washing and spraying.

If the treatment is performed within 90 days of generation in a compliant tank or container, a permit is not required. In many instances, decontamination pads may be viewed as "tanks". Care should be taken to document each element (ie. less than 5% staining per square inch) necessary to demonstrate compliance. If a CERCLA action, broad debris treatment authority can be gained by including the concept in the decision document ARARs.

When sampling the debris, compositing a healthy number of increments into a single sample for analysis can have three distinct benefits. First the analytical costs will be lower. Second you will be comparing an average to the regulatory threshold rather than an 80% UCL. Third, the accuracy of the result will be roughly equivalent to the accuracy achieved if the same number of analyses are performed. Very often, the number of samples is governed by the cost of analysis. If compositing is performed, a larger number of increments can yield better accuracy with little increase in cost.

The contained-in authority found at 40 CFR 261.3(f)(2) has enormous potential if local regulators participate to effectively execute projects. For example, if it is difficult to meet the less than 5% staining per square inch requirement, petitioning your local regulators for a contained-in determination may yield a sensible, protective result.

Finally, any decision to petition for a DET from EPA Headquarters should be considered carefully. Unless the amount of waste is significant, the benefits from reduced disposal costs may be quickly overwhelmed by the cost of the transaction. Supportive local regulators will also be important to a successful outcome.

What will happen if you try and take a state 40 CFR §261.3(f)(2) contained-in determination across state lines?

Is there a higher standard or particular legal significance to the language found at 40 CFR §261.3(f)(1) that states, in reference to extraction and destruction technologies:

BACK