|Staff Evaluation of the Reactor Vessel and Internals Removal Project
I. INTRODUCTION AND BACKGROUND A. Introduction
On January 15, 1997, Portland General Electric (PGE) submitted a proposed change to the Decommissioning Plan for the Trojan Nuclear Plant. The proposed change would permit removal and disposal of the reactor vessel and its internals as a single piece. If permitted, the reactor vessel and internals would be shipped by barge up the Columbia River and disposed of at the U.S. Ecology low level waste site at the Hanford reservation in Washington. PGE proposes to complete the project in 1999.
PGE has entitled this project Reactor Vessel And Internals Removal (RVAIR). The RVAIR project requires approval from the Oregon Energy Facility Siting Council (EFSC or "the Council"), the U.S. Nuclear Regulatory Commission (NRC), the U.S. Department of Transportation (USDOT) and the Washington Department of Health (WDOH).
This report evaluates RVAIR for compliance with EFSC requirements as set forth in Oregon Administrative Rules, Chapter 345 Division 26, abbreviated as OAR 345-026. Specifically, OAR 345-026-0370 provides criteria for review of Trojan´s Decommissioning Plan.
PGE permanently ceased operating the Trojan Nuclear Plant in January 1993. Since then, PGE has moved all of the spent nuclear fuel to the spent fuel pool and amended their NRC license to reflect the shutdown status of the plant. In October 1993, PGE issued its Defueled Safety Analysis Report (DSAR), which analyzes possible radiological accidents at Trojan and provides the basis for all safety evaluations concerning plant activities and modifications.
In January 1995, PGE submitted its Decommissioning Plan. After a one year review and opportunity for contested case, the Council approved the Decommissioning Plan on March 11, 1996. The NRC approved the Decommissioning Plan in April 1996.
As originally approved, the Decommissioning Plan called for removal of the reactor vessel internals by segmentation and removal of the vessel itself either by segmentation or intact. Most of these components would have then been shipped by truck to U.S. Ecology. PGE states that segmentation and overland shipping would have required at least 45 shipments, assuming the vessel was removed intact. If the vessel were segmented, an additional 10 shipments would be needed.
The reactor vessel "internals" consist of structural supports for the fuel, guide structures for control rods and instruments, core support plates and baffles, and other internal components that were located close to the fuel during reactor operations. The internals are highly radioactive due to the presence of activation products. The term "activation" means that the steel components have become radioactive as a result of neutron irradiation during reactor operations. The radioactivity is in the steel itself and is not in a form that can be dispersed. Portions of the internals are activated to the point where, if segmented and disposed of separately, they would not meet NRC criteria at Title 10 Part 61 of the Code of Federal Regulations (10 CFR 61) for burial at a federally licensed low level waste site such as U.S. Ecology. This waste, called Greater Than Class C or GTCC, would have to be stored at Trojan in casks similar to those proposed for dry spent fuel storage until a federally licensed facility can take them.
The RVAIR plan describes several reasons why PGE wishes to ship the vessel and internals as one piece. These reasons include significant reduction in low level waste volume, reduction in occupational radiation exposure, reduction in public exposure during transportation, and reduction in cost. OOE agrees that a single shipment by barge has apparent advantages over a series of at least 45 shipments by truck. Also, RVAIR removes the need for interim storage of GTCC waste at Trojan.
II. DESCRIPTION OF THE RVAIR PROJECT
PGE described the major steps of the RVAIR project in Section 2 of the "Reactor Vessel and Internals Removal Plan". The following summary is excerpted from PGE´s description.
A. Reactor Vessel Description
The reactor vessel is cylindrical, with a welded hemispherical bottom head and a removable hemispherical upper head. The vessel contains internal structures (internals) that provided support for the reactor core and reactor coolant flowpaths during plant operations. There are four inlet nozzles (27.2" in diameter) and four outlet nozzles (28.8" in diameter) that provided the flowpath for reactor coolant. Figure 1-3, reprinted from PGE´s Safety Analysis Report for the Reactor Vessel Package, provides a schematic of the reactor vessel. (Figures available in hard copy only.)
The reactor vessel is nominally 42´6" long, 16´ in diameter in the shell region, and 21´10" at the nozzles. The reactor vessel is fabricated from carbon steel with minimum 0.156" thick weld-deposited austenitic stainless steel cladding on the inside wall. The dry weight of the reactor vessel and internals is nominally 634 tons. The vessel thickness varies from 5 3/8" at the lower head to 10 1/2" in the nozzle area. The upper head has penetrations for instruments and control rods, and the lower head has penetrations for instruments.
B. Major Project Activities
Some of the activities needed for the project are already complete. These are activities that would be performed regardless of how the vessel is removed. Therefore, PGE concluded that it was safe and prudent to perform these steps in advance. These steps include removal of concrete structures that could interfere with vessel removal, removal of piping and other components attached to the vessel, enlarging an opening in the containment wall to allow vessel removal, and installation of a metal door to close containment when the vessel is being lifted. Some of these steps had already been partially taken during Large Component Removal in 1995, such as creating the opening in the containment wall.
However, PGE has enlarged the opening to accommodate the larger size of the reactor vessel.
If EFSC, WDOH and NRC approve the project, PGE will begin the activities needed for vessel removal. PGE will install and tension the studs that hold the upper head to the main part of the vessel. The vessel will be drained and injected with low density cellular concrete (LDCC). The purpose of the LDCC is to fix any movable contamination inside the vessel to preclude its escaping or shifting. The LDCC also provides radiation shielding. PGE also injected LDCC into the steam generators and pressurizer during the 1995 LCR project.
PGE will install shielding onto the outside of the vessel as required to meet 10 CFR 71 radiation level limits for radioactive waste transportation. Those limits are:
i. 200 mR/hour or less at the surface
ii. 200 mR/hour at any point on the vertical planes projected from the outer edges of the vehicle, on the upper surface of the load, and on the lower external surface of the load.
iii. 10 mR/hour or less at any point two meters from the vertical planes projected from the outer edges of the vehicle.
iv. 2 mR/hour in normally occupied spaces.
After the reactor vessel has been filled with LDCC and the attachments severed, it will be lifted and downended onto a rail system just above the 93´ level of containment (approximately 48´ above grade level). At this time PGE will install the main shielding sections. After final package preparations are completed, the reactor vessel package will be moved through the containment opening and lowered onto cradle assemblies attached to a specially designed transport vehicle (transporter). PGE will install impact limiters onto the reactor vessel package. The impact limiters will be constructed of a specially engineered impact absorbing plastic and are designed to absorb the shock in the unlikely event of a transportation accident. Figure 1-2, reprinted from PGE´s Safety Analysis Report for the Reactor Vessel Package, shows the reactor vessel package with the impact limiters installed, as it will appear on the transporter. (Figures available in hard copy only.)
The reactor vessel package will be secured to the transporter and inspected by ODOT. The loaded transporter will then carry the reactor vessel package to the barge slip on the Trojan site, using a road entirely on PGE property. The loaded transporter will be secured to the barge. Following a barge inspection by the U.S. Coast Guard, the barge will carry the reactor vessel package to the Port of Benton, Washington, approximately 270 miles up the Columbia river. Figure 1-1, reprinted from PGE´s Safety Analysis Report for the Reactor Vessel Package, shows the vessel package as it will appear on the barge. (Figures available in hard copy only.)
The barge will be pushed by a primary tug. A second tug will accompany the primary tug as backup. Both tugs will be equipped with dual communication methods and with navigational equipment approved by the U.S. Coast Guard. Communication will be established between the tug and a base station prior to transport. During transport the tug will contact the base station at least once every 4 hours.
PGE radiation protection personnel and a transportation coordinator will escort the shipment. The maximum speed will be 10 knots. At times, slower speeds may be required because of local conditions, such as traffic.
Once it arrives at the Port of Benton, the loaded transporter will be moved approximately 30 miles overland to the U.S. Ecology site for burial.
III. COMPARISON WITH LARGE COMPONENT REMOVAL
In 1995, PGE removed all four steam generators and the pressurizer and transported them by barge to the U.S. Ecology disposal site. This project was known as Large Component Removal (LCR). PGE and OOE have both made use of the experience gained from LCR.
There are many obvious similarities between LCR and RVAIR. The two projects have many major steps in common, such as :
· lifting a heavy component in containment and placing it a horizontal position,
· ensuring that all openings and penetrations are covered and welded closed,
· filling the component with low density cellular concrete,
· coating it with a suitable material to fix any external contamination,
· adding external shielding to reduce occupational and transportation dose,
· transporting the package out of containment onto a specially designed transporter, and
· barging up the Columbia
Because of the size and weight of these components, PGE uses a contractor to provide specialized lifting equipment and expertise. For RVAIR, PGE has hired the same contractor who did the LCR lifts. For the transportation phase of the project, PGE has also hired the same contractor who provided tugboat services for LCR. Most of the PGE engineers involved with RVAIR have prior experience from LCR.
There are also significant differences. These differences include:
· Larger Component Size: The steam generator removal required lifts of approximately 500 tons, while the reactor vessel will weigh over 1000 tons. The greater size of the reactor vessel presents different engineering challenges from those faced during the LCR project. For LCR, PGE was able to use the polar crane already installed in containment. For RVAIR, PGE´s lifting contractor will supply special lifting equipment for use inside containment. Also, the barge must be larger to accommodate the greater weight of the reactor vessel.
· Greater Radiation Source: The radiation dose from the unshielded reactor vessel would be greater than the dose from the steam generators, so more shielding is needed.
· Potential for Vessel Embrittlement: Since the reactor vessel was subject to high neutron flux during operations, the NRC required PGE to perform structural analyses demonstrating that the vessel would not crack during handling due to neutron Embrittlement.
· Order of Steps: In LCR, PGE injected the components with LDCC in containment and completed the package preparation outdoors. For RVAIR, PGE will fill with LDCC, attach the shielding, coat the package, and do almost all of the package preparation in containment.
Two new factors will make RVAIR easier to accomplish. In 1995, PGE had to maintain strict security around containment during LCR because the containment building was considered part of the protected area. PGE has now removed this concern by isolating the containment building from all areas where security is necessary and by welding shut all possible means of accessing the protected area from containment.
A second factor is the impact of the project on decommissioning and spent fuel storage. In 1995, PGE had to ensure that steam generator removal would not impact the spent fuel pool or spent fuel cooling systems. However, in 1998, PGE installed a new stand-alone spent fuel cooling system, which is fully self-contained and does not have piping or other equipment in any locations that could be affected by the reactor vessel shipment. Moreover, PGE has eliminated all equipment access between containment and the fuel building, removing the possibility that heavy equipment associated with RVAIR could drop into the spent fuel pool. Therefore, there is no credible way that RVAIR could impact safe storage of spent fuel.
Lessons Learned from LCR
PGE made use of the lessons learned during LCR by documenting them in an internal report entitled "Trojan Nuclear Plant Large Component Removal Project". This report describes PGE´s experiences during LCR and includes suggestions should PGE perform similar projects in the future. This type of report is not a regulatory requirement; however, we reviewed the suggestions made by PGE and verified that PGE took many of those suggestions into account while planning RVAIR.
OOE Observation of LCR
The OOE resident inspector was present at Trojan for the entire LCR project. Much of OOE´s expectations for RVAIR are based on our observations from 1995.
Our overall observation was that the LCR project proceeded smoothly and without major incidents that could have impacted safety. PGE encountered problems with equipment and tools early in the project, particularly with the containment crane. These caused some delays and schedule problems but in no way impacted safety. PGE made good use of wooden mockups for training purposes, and held briefings with all workers before each major milestone in the project. Workers had ample opportunity to suggest better ways of accomplishing their tasks throughout the LCR project, which resulted in a steady decline in personnel radiation exposure as the project progressed.
There was also a high level of involvement by radiation protection personnel. PGE removed four steam generators and the pressurizer, for a total of five shipments. With each successive shipment there was a noticeable decline in radiation exposure as PGE staff became more familiar with the tasks. PGE´s radiation exposure estimates for LCR proved to be very conservative. In their application to the Council, PGE estimated that they could complete the LCR project and incur not more than 138 person-rem. The actual personnel exposure for the project was approximately 65 person-rem. Based on this experience, OOE believes PGE can complete the RVAIR project within their predicted radiation exposure, and will likely incur less dose than predicted.
NRC Inspections of LCR
The NRC had regional inspectors present at Trojan for most major steps in the LCR project. OOE asked PGE to identify any NRC inspection reports that described NRC observations of the LCR project. In all, NRC inspectors commented on various aspects of LCR in 10 separate reports issued between November 1994 and January 1996.
Generally, NRC inspectors found that PGE´s LCR activities were satisfactory and complied with NRC regulations. The NRC reviewed PGE´s safety evaluation prior to LCR and found it satisfactory. Several of the inspection reports include favorable comments regarding PGE´s job planning, ALARA practices, and general radiological protection practices. Early in the project, an NRC inspector noted a problem in the procedures and guidance for radioactive waste transportation, which was corrected by PGE. NRC inspectors also had favorable comments on the program for rigging safety, especially the training program for crane operators and riggers. The NRC also inspected the steam generator packages in their as-shipped configuration and found that they met the requirements of PGE´s Certificate of Compliance under 10 CFR 71. Also, NRC inspectors observed that the Quality Assurance Department was actively involved in the project.
At the conclusion of the LCR project, the NRC concluded that "The radiological safety program and the quality assurance program implemented during this project contributed to the safe completion of this effort."
LCR Observation Summary
Despite the larger size of the reactor vessel package, OOE believes the successful completion of the LCR project, the relative lack of incidents, and the low radiation exposure incurred provide strong indications that PGE can complete the RVAIR project safely and without incident.
IV. REGULATORY REQUIREMENTS
The RVAIR project requires approvals from four separate agencies: NRC, USDOT, Washington DOH, and Oregon EFSC.
A. NRC Approval
PGE proposes to ship the reactor vessel as its own shipping container. To do this, PGE must show that the reactor vessel, as packaged, meets NRC requirements in the Code of Federal Regulations (10 CFR 71) for transportation of radioactive waste. On March 31, 1997 PGE submitted its Trojan Reactor Vessel Package Safety Analysis Report (PGE-1076) to the NRC.
The requirements of 10 CFR 71 address conventional radioactive waste shipments. They were not written to address one-piece shipment of a reactor vessel for compliance. This caused a delay in the NRC review. However, NRC review of the transportation package is nearly complete. On July 30, 1998, NRC representatives met with PGE to resolve technical issues in PGE´s application for NRC approval. In a separate meeting on the evening of July 30, 1998, the NRC, OOE and USDOT held a joint public meeting in Kelso, Washington to describe the RVAIR project to the public and gather any public concerns. No concerns were raised at that meeting.
NRC requirements at 10 CFR 71 address all aspects of transportation safety for radioactive waste. They specify the allowed radiation dose rate from the transportation package, require safety analyses for postulated transportation events, and require the licensee to demonstrate that the transportation package will not result in significant radiation dose or contamination to the public. The NRC application process also requires quality assurance measures to ensure that the transportation package will perform as described in the licensee´s safety analysis.
A typical application under 10 CFR 71 must include analyses for a 1 foot drop and a 30 foot drop. In its application to the NRC, PGE requested approval with alternate drops, including an 11 foot horizontal drop and 11 foot oblique pivot drops with impact limiters. PGE based this request on the favorable transportation route and the extensive transportation controls proposed for this shipment. PGE provided a probabilistic analysis showing that the risk of a transportation accident was extremely low (less than one in a million). PGE´s application states that the proposed shipment qualifies for exemption under 10 CFR 71.8, because the favorable route, extensive controls and low risk provide a level of safety "...equivalent to the shipments which would be necessary if segments were shipped in smaller packages which meet the environmental and test conditions of Sections 71.71(c)(7) and 71.73(c)(1)."
At its meeting of June 24, 1994, EFSC decided that the NRC requirements were sufficiently comprehensive and detailed that additional state requirements for transportation were not needed.
Therefore, this report does not include a detailed review of PGE-1076, nor does EFSC make findings of compliance with 10 CFR 71. EFSC approval of PGE´s Decommissioning Plan change would not imply approval of the transportation package. OOE recommends that EFSC approval of RVAIR be contingent on NRC approval under 10 CFR 71.
B. USDOT Approval
In addition to NRC approval, the project must be approved by USDOT in accordance with 49 CFR 173. Because the reactor vessel package will not meet the definition of a Type B package provided in 49 CFR 173.403, PGE has requested an exemption. The exemption request is supported by the same reasoning and analyses provided to the NRC. The USDOT requirements parallel the NRC requirements, and the USDOT will rely on the radioactive material transportation package being approved by the NRC.
C. State of Washington Approval for Burial
The RVAIR project is practical only if the reactor vessel package can legally be disposed of at the low level waste site operated by U.S. Ecology in Hanford. Specifically, the reactor vessel package must meet waste disposal requirements of 10 CFR 61.
NRC regulations at 10 CFR 61.55 divide low level radioactive waste into classes A, B and C. These classifications are based on the concentration of radionuclides, particularly long- lived radionuclides. Class A waste has the lowest concentration, and Class C has the highest. Class C waste must meet rigorous requirements on waste form to ensure stability, and requires additional measures at the disposal facility to protect against inadvertent intrusion. Waste that exceeds the requirements for Class C is considered Greater Than Class C (GTCC) and is generally considered unsuitable for near surface land disposal.
As described in the Decommissioning Plan, the reactor internals are activated due to neutron radiation during plant operation. More than 99% of the radioactivity in the reactor vessel is in the form of activated metal. Since the metal itself is radioactive, it could not readily disperse in a transportation accident. Portions of the reactor internals are activated to the point where, if disposed of separately, they would be GTCC waste as defined in 10 CFR 61. By volume, the GTCC waste is a small fraction of the total reactor vessel. PGE estimates that the total reactor vessel package, if disposed of in one piece, would be 8341 ft3 . If segmented and stored separately, there would be 13,236 ft3 of Class A waste, 231ft3 of Class B waste, 4513 ft3 of Class C waste, and 340 ft3 of GTCC waste.
If removed from the reactor vessel and handled separately, these components would not be eligible for disposal at the U.S. Ecology site, and would remain at the Trojan site until a federally licensed disposal site became available. However, in 1995, the NRC issued a "Technical Position on Concentration Averaging and Encapsulation" (Branch Technical Position), which describes alternative provisions for waste classification. The Branch Technical Position states that activated metals might be acceptable for near surface disposal under special circumstances. For example, a very large package containing only a small amount of highly activated metal might have an average radionuclide concentration that meets the definition of Class C. The Branch Technical Position also states that 10 CFR 61 requirements for near surface disposal might be met for a "...large activated component filled with a structurally stable medium (e.g. cement), or enclosed in a massive robust container capable of meeting structural stability requirements."
Using the guidance in the Branch Technical Position, PGE submitted an analysis to show that the vessel and internals, if handled as a single piece, meet the performance objectives of 10 CFR 61 for disposal at the U.S. Ecology site.
The State of Washington is an "Agreement State" under NRC regulations. This means the NRC and Washington have signed an agreement delegating authority and responsibility for licensing the low level waste site to the WDOH. Therefore, PGE´s analysis, showing that the vessel is suitable for disposal at the U.S. Ecology site, must receive approval by WDOH.
The 10 CFR 61 performance objectives address long-term safety of radioactive waste storage. PGE summarized their understanding of the 10 CFR 61 objectives and their methodology for demonstrating that the vessel is safe for disposal in the following excerpt from their analysis:
"The performance objectives addressed by 10 CFR 61 for land disposal of LLRW are:
1. Long-term protection of the public health and safety (and the environment);
2. Protection of an inadvertent intruder;
3. Protection of workers and the public during operation of a LLRW disposal facility; and
4. Long-term stability of the disposal site after closure
This report evaluates projected exposures during disposal operations and for a group of hypothetical radiation release scenarios involving the disposal of the vessel at the LLRW Facility including the groundwater pathway. These exposure scenarios encompass possible events that may occur during site operations and following vessel disposal and LLRW Facility closure, including the hypothetical breakdown of active institutional controls. The hypothetical scenarios are considered appropriate for conservatively estimating the potential impacts from disposal of the vessel intact versus removing and sectioning the internals. The hypothetical radiation release scenarios were evaluated qualitatively in a bounding case analysis. Those found to warrant detailed quantitative analysis were included in the pathways analysis and dose assessments.
The hypothetical doses via the groundwater pathway from disposal of the vessel are calculated up to 10,000 years into the future. The vessel will be modeled separately from the other waste previously disposed at the LLRW facility. The incremental dose effect from the disposal of the vessel will be compared to the dose projections for the entire disposed waste volume anticipated at LLRW Facility closure.
The groundwater pathways analyses were performed consistent with the recommended approaches to performance assessments described in the Draft NUREG-1573. In instances were the site-specific analyses modified the approach described in the Draft NUREG-1573, a description of the modification is included. The inadvertent exposure pathways where individuals were allowed to remove the closure cap and waste from the disposal trench were evaluated following the assumption presented in the Draft 10 CFR Part 61 Generic Environmental Impact Statement (EIS) and the Final EIS."
Because the suitability of the package for disposal is under WDOH jurisdiction, OOE did not review the PGE pathway analysis in detail. However, we did examine the pathway analysis to see what scenarios PGE considered, what assumptions they used, and what results they arrived at.
The pathway analysis includes eleven different scenarios, including an agricultural scenario in which a family resides near the LLRW Facility and consumes crops, milk and groundwater potentially impacted by site operations; a construction scenario where the buried reactor vessel is breached by a driller attempting to install a recovery well; and a residential a scenario where an individual builds a house with a basement built into the closed disposal trench cap. PGE performed detailed quantitative analyses on these three scenarios.
Additionally, PGE performed qualitative analyses on scenarios involving natural phenomena such as flooding, natural erosion, and biotic transport. All of the doses PGE calculated were a small fraction of the WDOH regulatory limits. The NRC required PGE to assume that the reactor vessel corrodes after 500 years, exposing the internals. This is a conservative assumption considering the arid climate in the Hanford area and the fact that the reactor vessel is scheduled to be disposed of at a depth exceeding five meters.
In summary, OOE considers the 10 CFR 61 performance objectives appropriate, and we consider the requirement for a 10,000 year groundwater analysis to be very conservative. WDOH will not approve the package for disposal until they have completed their detailed review of the pathway analysis and confirmed PGE´s conclusions. EFSC approval of the RVAIR project and the associated Decommissioning Plan change should be contingent on WDOH approval of the vessel package for disposal.
D. EFSC Approval of Decommissioning Plan Change
PGE submitted its Decommissioning Plan for EFSC approval in January 1995. EFSC issued its order approving the Decommissioning Plan on March 14, 1996. The Decommissioning Plan as approved by EFSC, described segmentation of the internals, vessel removal either by segmentation or whole, and shipping the segments overland. Without RVAIR, 340 ft3 of activated metal from the reactor internals would be GTCC waste and would be stored at Trojan in casks similar to those used for spent fuel under the original plan.
EFSC rule OAR 345-026-0370(4) sets forth criteria for determining whether a change to the Decommissioning Plan requires EFSC approval before implementation. PGE has determined, and OOE concurs, that the RVAIR project is significant because it involves a change in the provisions made for hazardous or radioactive waste material removal.
PGE´s request for EFSC approval includes an analysis demonstrating that the RVAIR project would not violate EFSC criteria for a decommissioning plan described in OAR 345-026-0370.
In reviewing the RVAIR Plan, OOE considered the Plan´s impact on the Decommissioning Plan, on PGE´s compliance with the decommissioning criteria in OAR 345-026-0370, and on compliance with EFSC´s March 14, 1996 order approving the Decommissioning Plan. OOE´s conclusions follow in Sections VI and VII of this report.
If EFSC approves the RVAIR Plan, EFSC will amend its 1996 order approving the Decommissioning Plan. The amendment may either approve the plan as written or subject to additional conditions.
V. EVALUATION OF PROJECT SAFETY
A. Heavy Lift Safety
Because PGE does not have cranes capable of lifting the reactor vessel package, PGE has hired a special lifting contractor, Bigge Crane and Rigging Co., to provide engineering and heavy lift services. Bigge is the same contractor who performed the lifts for LCR.
Prior Lifting Experience
OOE asked what prior experience Bigge had with lifts of this size. In addition to LCR, Bigge has removed steam generators at other operating nuclear plants, including Seabrook in New Hampshire and Salem in New Jersey. Bigge has installed new steam generators weighing 550 tons at the South Texas nuclear plant and at St. Lucie in Florida. In 1975, they lifted a 900 ton reactor vessel (new and clean) for Hope Creek in New Jersey. Bigge states that in a 1989 contract for Pacific Gas and Electric, they lowered loads weighing 3,600 tons. Based on this information, OOE believes PGE´s lifting contractor has the necessary experience.
The lift will not use the installed containment building crane, as was done for LCR. Instead, Bigge will use hydraulic jacks. They will not lift the vessel in one motion, but will use a ratcheting scheme where the vessel is raised a small amount, locked into place, and then raised again. This will be repeated until the vessel has reached the required height. The hydraulic jacks are modular, so that if more lifting capacity is needed, Bigge will use more jacks.
OOE reviewed plans and drawings for the jacking platform and attachments, and interviewed the PGE engineers responsible for the lift. We also reviewed the drawings showing how the lifting device is attached to the vessel. The points of attachment are the studs that hold the upper head onto the reactor vessel. These are the same studs which held the reactor head in place during power operations. During operations, the pressure inside the vessel was 2250 lbs.per square inch, and each stud was tensioned to 1,490,000 lbs. Therefore, PGE believes, and OOE concurs, that these studs will adequately support the weight of the vessel.
Special forgings are used as lifting lugs for the vessel package. These forgings were constructed expressly for Trojan. The special forgings used for this lift were designed and constructed in accordance with ANSI N14.6, which requires safety factors of three times the yield strength and five times the ultimate tensile strength of the materials
PGE will test lifting components in accordance with ANSI 14.6. The forgings that attach the reactor head to the jacking platform will be tested to 150% of design load. The jacks, supporting I-beams, and rigging will be tested in the as-lifted configuration to 110% of the design load prior to the RVAIR project.
Quality Assurance Measures
The Decommissioning Plan states that radioactive waste disposal processes falls under PGE´s Quality Assurance (QA) Program. In our review of the Decommissioning Plan, OOE examined the QA Program and concluded that it was acceptable and effectively implemented. In our January 1996 report to EFSC, OOE stated that:
"...the Quality Assurance and Quality Control programs appear effective in assessing plant performance through a comprehensive program of reviews, audits, surveillance, and involvement in plant staff meetings."
OOE has extensively monitored PGE´s QA involvement in the independent spent fuel storage installation project (ISFSI). OOE accompanied PGE QA inspectors on a vendor audit and has observed QA activities continuously throughout our review of the dry spent fuel storage project. We concluded that PGE has an aggressive vendor audit program and will not hesitate to require corrective actions whenever vendor activities, services or products do not meet PGE´s QA program.
In connection with the RVAIR project, we discussed PGE´s audit of Bigge´s QA measures with the lead auditor, and we reviewed several audit reports. PGE´s QA oversight of the RVAIR project and vendor supplied equipment or services was performed at a level equivalent to what we observed firsthand in our ISFSI review.
Because most of Bigge´s business is outside the nuclear industry, some of the components were purchased "commercial grade". This means that they were originally manufactured outside of a nuclear QA program, and then proved through testing, analysis or documentation to function properly. The testing varies with the task. An example of a commercial grade item is the shielding that PGE will place on the reactor vessel.
For load bearing components such as slings and tie-downs, PGE will rely more on testing than on documentation to make sure the components have adequate strength. During the lift, the procedure will be under the QA Program and will have QA "hold points" - steps at which work stops while a PGE quality control inspector verifies that all steps are performed adequately and according to approved plans or procedures. Those procedures are not yet ready for review, but they will be OOE inspection items if RVAIR is approved.
Conclusion - Lift Safety
Our review of the lift safety focused on four factors: the experience of the lifting contractor including LCR, the designed safety factors for load bearing components, the testing prior to lift, and the level of QA involvement. Based on these factors PGE concluded, and OOE concurs, that the likelihood of a reactor vessel drop or other accident is remote.
B. Reactor Vessel Drop
1. PGE 10 CFR 50.59 Safety Evaluation
OOE asked if PGE had evaluated the consequences of a drop or other accident involving the reactor vessel during RVAIR. PGE reported that they performed a safety evaluation in accordance with the requirements of 10 CFR 50.59, which includes, by attachment, an analysis of a hypothetical reactor vessel drop.
NRC regulations at 10 CFR 50.59 permit PGE to make changes in the Trojan facility as described in the Defueled Safety Analysis Report (DSAR), or make changes to the procedures described in the DSAR, or conduct tests or experiments not described in the DSAR, without prior approval of the NRC, as long as the proposed change, test or experiment does not constitute a change in the Technical Specifications or an unreviewed safety question. An unreviewed safety question occurs:
(1) if the probability of occurrence or the consequences of an accident or malfunction previously evaluated in the DSAR may be increased; or
(2) if a possibility for a new type of accident or malfunction not previously evaluated in the DSAR may be created; or
(3) if the margin of safety as defined in the Technical Specifications is reduced.
PGE´s 50.59 evaluation states that the RVAIR project does constitute a change to the DSAR, but not to the procedures described in the DSAR, and that it does not involve a test or experiment not described in the DSAR. The evaluation states further that the project constitutes changes to both the Decommissioning Plan and the Security Plan, but not to the Technical Specifications, QA Program, Permanently Defueled Emergency Plan, Fire Protection Program, or Radiological Environmental and Effluent Monitoring Program. Because RVAIR was determined to constitute a change to the DSAR, PGE proceeded with a detailed safety evaluation.
As stated, the bases for a 10 CFR 50.59 safety evaluation are the DSAR and Technical Specifications. Both documents focus primarily on protecting the spent fuel. Because PGE has physically isolated the containment building from the fuel building and has installed an independent spent fuel cooling system, PGE concluded, and OOE concurs, that the RVAIR project will not impact any Technical Specification or the margin of safety in the basis for any Technical Specification.
The DSAR describes three general classifications of accidents:
(1) Radioactive release from a subsystem or component.
(2) Fuel handling accident.
(3) Loss of spent fuel decay heat removal capability.
Specifically, the postulated DSAR accidents for an offsite radiological release include a radioactive gas waste system leak or failure and a fuel handling accident. The RVAIR project was not planned when the DSAR was written, a reactor vessel drop was not postulated, and consequently, there is no consideration in the DSAR for a release to the environment from the reactor coolant system. The DSAR states, "The Reactor Coolant System (RCS) and secondary system are no longer operated at elevated temperature or pressure. Therefore, releases to the environment from these systems were not considered." The other accident scenarios evaluated in the DSAR include those involving the spent fuel pool. However, as previously stated, PGE´s 50.59 safety evaluation also considered the impact of RVAIR on other documents, including the Decommissioning Plan, which does evaluate offsite radiological releases from material handling events.
PGE´s 50.59 safety evaluation, therefore, makes the following conclusions:
(1) The probability of occurrence or the consequences of an accident or malfunction previously evaluated in the DSAR is not increased.
(2) The possibility for a new type of accident or malfunction not previously evaluated in the DSAR is not created.
(3) The margin of safety as defined in the Technical Specifications is not reduced.
OOE concurs with conclusions (1) and (3), as described in the 50.59 safety evaluation, and also with conclusion (2), based on the additional consideration of the accident analyses described in the Decommissioning Plan. Therefore, the RVAIR project does not constitute an unreviewed safety question.
The consequences of a reactor vessel drop are considered in the 50.59 safety evaluation. The evaluation states that "A drop of the reactor vessel is the worst case accident/failure associated with the RVAIR activities." It states further that "No postulated radioactive releases from RVAIR activities will exceed dose limits at the Exclusion Area Boundary." This conclusion is presented in Sections 5.2 and 18.104.22.168 of the RVAIR Plan, which is attached to the safety evaluation, and described below.
PGE assumed that, regardless of height, the reactor vessel would break open and release loose radioactive contamination. PGE then compared the consequences of this release to those previously analyzed. The Decommissioning Plan references a calculation of the largest release that could occur without exceeding Environmental Protection Agency (EPA) protective action guidelines for intermediate term releases, as agreed to with the Oregon Health Division. Those guidelines require that the dose to any persons at the site boundary be less than 0.5 REM Total Effective Dose Equivalent. PGE calculated that the EPA guidelines would not be exceeded for any release of less than 2.07 Curies (Ci). Both the NRC and OOE reviewed and accepted this calculation. PGE then showed that the release from a hypothetical drop of the reactor vessel would release less than 2.07 Ci.
PGE´s analysis of the postulated reactor vessel drop is based on two key items: the amount of loose contamination in the reactor vessel and the amount of that contamination which would become airborne. PGE calculated that the vessel contains 155 Ci of surface contamination, based on samples taken from inside the steam generators and portions of the reactor coolant system piping in preparation for Large Component Removal. PGE then assumed that 10% of this contamination would be released in a drop and subsequent breech of the reactor vessel. 1% of this amount, or 0.1% of the total surface contamination, would then become airborne. This assumption comes from NRC Reg. Guide 1.25, which PGE has used previously in dose calculations for the DSAR, Large Component Removal project, and the Decommissioning Plan. Using this guidance, PGE calculated that 0.155 Ci (0.1% of 155 Ci) would be released offsite. This is well below the 2.07 Ci calculated in the Decommissioning Plan accident analysis. OOE reviewed these assumptions and calculations during our review of the Decommissioning Plan and considers them acceptable. Additionally, there is ample margin between the 0.155 Ci released in this event and the value of 2.07 Ci.
Because this analysis relies on the assumption from NRC Regulatory Guide 1.25 that only 0.1% of contamination becomes airborne, OOE reviewed Regulatory Guide 1.25 to ensure it was appropriate for this type of accident. We found that Regulatory Guide 1.25 was originally written to consider fuel handling accidents, and we had no basis to determine if a dropped reactor vessel would result in a higher or lower percentage of contaminants released. However, the NRC subsequently released guidance in NUREG0170, which considers transportation accidents. This guide, while not quoted in PGE´s safety evaluation, recommends an assumption of an identical fraction of 0.1% for airborne contamination release. Based on this guidance, OOE concurs that PGE´s assumption of 0.1% is acceptable.
Based on this analysis, PGE concluded, and OOE concurs, that in the unlikely event of a reactor vessel drop, the offsite consequences would not exceed the radiation releases previously analyzed in the Decommissioning Plan.
2. Best Estimate Analysis of Reactor Vessel Drop
In its 10 CFR 50.59 evaluation, PGE analyzed a hypothetical "worst case" accident in which they assumed the reactor vessel breaks on impact. We asked if PGE had done a more realistic analysis of the reactor vessel drop. PGE reported that they completed analyses of three drop scenarios, one inside and two outside the containment in different orientations.
The calculation basis for these analyses is provided in EPRI NP7551, Structural Design of Concrete Storage Pads for SpentFuel Casks, April 1993. This document is primarily intended for the design of storage pads. It presents a method of evaluating pad design relative to the loading experienced by a dropped or tipped over dry storage cask. It can also be used, however, for the design of casks. The methodology determines the energy absorption of a concrete slab during drop accidents. This can be used, in turn, to determine the actual force experienced by a dropped cask. A similar energy balance approach was applied by PGE to the dropping of the reactor vessel on a concrete surface inside containment and a rock surface outside containment.
The results of these analyses was that in the unlikely event of a dropped reactor vessel either inside or outside containment, the vessel experiences only elastic deformation. In other words, the reactor vessel does not break in any of the drop scenarios.
For a drop outside containment, PGE´s analysis shows that there would be no significant release of contamination, since the vessel remains intact. For a drop inside containment, there would be a release of radioactive material, which would be contained in the containment building. As the postulated drop occurs, the reactor nozzles strip away as much as 3 feet of the biological shield wall. This concrete wall contains activated products from reactor operations. From PGE´s Site Characterization Report (submitted with the Decommissioning Plan of 1995) and from interviews with members of PGE´s Health Physics staff, we learned that the amount of activation products available for release is between 300 and 500 Curies. The release would be in the form of concrete rubble, ranging in size from large chunks to dust. To estimate how much of this material would become airborne, PGE relied on studies performed by U.S. DOE and described in U.S. DOE Handbook 3010-94. This handbook contains empirical correlations based on experimental results on concrete performance when subjected to impacts and shocks. Using these correlations, PGE calculated that approximately .02% of the concrete would become airborne as dust. Taking no credit for the containment building or for filtering by the containment building ventilation system, this would be equivalent to an airborne release of 0.1 Ci , which is well below the limiting 2.07 Ci calculated in the Decommissioning Plan accident analysis. Further, PGE has committed to closing the containment opening during this lift, so that the airborne release would be confined to the containment building.
OOE reviewed the basis for and assumptions used in the calculations and found them to be reasonable and conservative. We also did a spotcheck of the calculational methods and found them to be satisfactory. Therefore, OOE concurs with PGE´s analysis that the reactor vessel does
not break in any of the reactor vessel drop scenarios, and the radiological consequences for a drop inside containment are below the prescribed limits.
C. Transportation Safety
As stated above, the NRC has primary authority over radioactive waste transportation safety, including extensive regulations at 10 CFR 71. In our review of the RVAIR plan, OOE used PGE´s 10 CFR 71 application as a source of information. EFSC approval of the RVAIR plan does not imply approval of the shipping package in accordance with 10 CFR 71. That approval must come from the NRC. However, OOE did review PGE´s 10 CFR 71 application to verify that PGE has taken appropriate precautions. Some of the precautions described in PGE´s 10 CFR 71 application include:
· PGE verification that there is no adverse weather forecast along the route
· Use of a backup tug
· Arrangements with the U.S. Army Corps of Engineers to provide priority passage and exclusive use through the locks.
· A commitment to limit the barge speed to 10 knots.
· A PGE radiation protection representative will accompany the shipment, equipped with radiation detection equipment
· U.S.Coast Guard involvement and support.
OOE has emergency preparedness responsibility for any incidents involving transportation of radioactive waste from Trojan. In its rule approving the LCR project, EFSC required a Transportation Safety Plan. Specifically, OAR 345-027-0370(9)(r) states that:
"Portland General Electric shall submit to the department a comprehensive transportation safety plan with prior coordination between State and Federal agencies with emergency responsibilities prior to component shipments."
In anticipation of a similar requirement for RVAIR, PGE submitted a Transportation Safety Plan to OOE in March 1998. OOE´s radioactive waste transportation coordinator reviewed that plan and commented. OOE recommended that PGE develop a fact sheet modeled after the fact sheet used for U.S. Navy shipments of submarine reactor compartments, and provide timely notification of news media as significant milestones are reached. OOE also requested that PGE include further details of emergency public information coordination and a notification list of public information contacts. PGE agreed to revise their plan to include OOE´s suggestions.
OOE recommends that EFSC approval of the RVAIR project be contingent on OOE´s final approval of the Transportation Safety Plan prior to shipment.
D. Radiological Protection
The RVAIR project will be subject to the PGE radiation protection program. OOE previously reviewed the radiation protection program in connection with the Decommissioning Plan and found it acceptable. In OOE´s January 22, 1996 review of the Decommissioning Plan, we state:
"The department concludes that PGE´s radiation protection and ALARA programs contain effective programmatic controls. They are adequate to maintain public, occupational and environmental radiation exposures within established limits and according to the ALARA principle."
The same occupational radiation protection program will apply whether PGE removes the vessel in one piece or by segmentation. Therefore, the RVAIR project is not expected to have any adverse impact on occupational exposure, when compared with the option to segment the vessel.
PGE states that RVAIR will reduce occupational exposure when compared with the segmentation option. As originally approved, the Decommissioning Plan estimated that PGE would incur approximately 85 person-rem for vessel removal by segmentation. In the RVAIR Plan, PGE estimates occupational doses of 67 person-rem. As stated earlier in this report, PGE´s occupational exposure estimates for LCR proved very conservative, indicating that occupational exposure estimates for RVAIR may also be conservative.
The RVAIR Plan also states that the radiation exposure to the general public from a single reactor vessel shipment would be 0.111 rem, compared with a general public exposure estimate of 1.539 rem for the multiple shipments required to remove the vessel by segmentation. OOE concurs.
VI.COMPLIANCE WITH OAR 345-026-0370
The RVAIR Plan, if approved, would be a change to the Decommissioning Plan, which EFSC approved in March 1996. Any change to the Decommissioning Plan must continue to meet the EFSC decommissioning acceptance criteria set forth in OAR 345-026-0370, and particularly sections (2) and (3).
In the RVAIR plan, PGE analyzed the impact of RVAIR on PGE´s continued compliance with the criteria of OAR 345-026-0370(2) and (3).
OAR 345-026-0370(2)(a) and (b): These subsections contain the EFSC criteria for unrestricted release of the site after decommissioning is complete. PGE states that RVAIR will not affect compliance with these criteria because PGE will still be required to meet the same criteria (or any alternate criteria subsequently adopted by EFSC) for unrestricted release. We concur.
OAR 345-026-0370(2)(c): This subsection requires provisions to remove all radioactive waste on a schedule acceptable to the Council. The RVAIR project is consistent with this requirement because it calls for the removal of a significant volume of radioactive waste on an accelerated schedule, and because it facilitates removal of GTCC waste, which otherwise would remain at Trojan. In approving the plan, EFSC would also be finding the schedule acceptable.
OAR 345-026-0370(2)(d) & (e): These subsections require programs for effluent monitoring and control and for radiological environmental monitoring. PGE´s current effluent monitoring and radiological environmental monitoring program is the Offsite Dose Calculation Manual, which OOE approved on December 12, 1994. PGE has revised the ODCM since its approval in December 1994. OOE examined the current revision to ensure it remains acceptable and does not reduce the program´s effectiveness, and concluded that it was acceptable. PGE states that the RVAIR project does not require any further changes to the ODCM.
The RVAIR project is not expected to result in any radioactive releases beyond what is already considered in the Decommissioning Plan. During heavy lifts, the containment building will be closed and any airborne contamination would be contained. The reactor vessel will already be shielded and coated before it is removed from containment. All activities such as cutting and welding associated with RVAIR are similar to activities already used in decommissioning. Therefore OOE concurs with PGE´s statement that the RVAIR project does not affect compliance with the requirement for effluent monitoring and control or environmental radiological monitoring.
OAR 345-026-0370(2)(f): This subsection requires provisions for removal of hazardous waste.
PGE´s programs for control and removal of hazardous waste were described in the Decommissioning Plan and accepted by EFSC in its March 1996 approval of the Plan. Any wastes generated during RVAIR would be similar to wastes generated previously during LCR or during routine decommissioning activities. For example, the coating PGE will use to fix any external contamination will be similar to the coating used during LCR. Therefore, OOE concurs with PGE´s statement that the RVAIR project will not adversely impact PGE´s compliance with EFSC requirements for hazardous waste removal.
OAR 345-026-0370(g): This subsection required PGE, in its Decommissioning Plan, to choose
a decommissioning alternative and analyze that alternative in comparison to DECON and SAFSTOR, as those terms are defined by the U.S. Nuclear Regulatory Commission. In its Decommissioning Plan, PGE selected the DECON alternative and showed that the impacts of its plan are within the impacts described in the NRC´s Generic Environmental Impact Statement on decommissioning. The RVAIR project is consistent with the DECON alternative. It calls for removal of the reactor vessel and internals, but it changes the method of removal and transportation. OOE agrees with PGE´s statement that, compared with the segmentation option, RVAIR would reduce the potential for occupational and public exposure. We expect that WDOH will agree with PGE´s conclusions on the impact of burial. We conclude that RVAIR does not have an adverse impact on PGE´s compliance with this rule.
OAR 345-026-0370(3): This section addresses the adequacy of the decommissioning fund. PGE states that RVAIR will reduce decommissioning costs, compared with the segmentation option. PGE estimates that RVAIR will cost $23.8 million, compared with an estimate of $38.4 million for separate disposal. OOE agrees with PGE´s statement that RVAIR will not have an adverse effect on the continued adequacy of the decommissioning fund.
VII. IMPACT ON DECOMMISSIONING PLAN
The RVAIR Plan states that RVAIR activities will not change the results of the evaluations presented in the Decommissioning Plan.
In the RVAIR Plan, PGE evaluated the events considered in the Decommissioning Plan. These events
include materials handling events, loss of support systems, fires, explosions, and external events such as earthquake or flood. The RVAIR plan states that none of these postulated events would result in an increased radioactive release, when compared with the currently approved segmentation option. OOE concurs.
A reactor vessel drop was the only event that could cause a radioactive release different from those analyzed in the Decommissioning Plan. As discussed in Section V.B.1 of this report, PGE analyzed the consequences of a reactor vessel drop and concluded that the activity released would be well below the 2.07 Ci needed to exceed the EPA protective action guidelines.
Therefore, OOE agrees with PGE´s conclusion that the RVAIR project would not result in increased risk to the public when compared with the approved Decommissioning Plan.
OOE recommends that the Council approve PGE´s Decommissioning Plan change and a mend its Decommissioning Order of March 14, 1996, to allow implementation of the RVAIR project. We further recommend the following conditions:
1. Representations in PGE´s January 30, 1997 "Reactor Vessel and Internal Removal Plan" , as updated by July 9, 1998 letter JDW-014-98TF from PGE to OOE, shall be considered commitments on PGE´s part.
2. Representations in PGE´s "Application for 10 CFR 71 Certificate of Compliance" shall be considered binding commitments by PGE.
3. PGE shall not take action to preclude reactor vessel internals removal by segmentation and overland shipment until the NRC has issued final approval of the reactor vessel package for shipping under 10 CFR 71 and the Washington Department of Health has issued final approval of the reactor vessel package for burial under WAC 246-250.
4. PGE shall not proceed with shipment of the reactor vessel package until OOE has issued final approval of the Transportation Safety Plan, PGE-1077.
5. Prior to heavy lifts inside or outside containment, PGE shall test special attachment forgings to 150% of design load and shall test the jacking platform and rigging in the as lifted configuration to 110% of design load as described in Section V.A of this report.