Casks of Uncertainty: The NAC LWTS
The propensity for accidents in the Feather River Canyon is greatly increased by effects of one of the biggest El Ninos of the 20th Century. Data from the California Data Exchange Center, which makes available the California Department of Water Resources (CDWR) information, indicates that summary totals of water contributing to the Feather River Drainage may well be over 500% this summer when the snow melts. As early as 4/1/98, the Snow Course Measurement in the Feather River Drainage was 168% above average and a 6/1/98 measurement recorded a staggering 495% above normal. Without a doubt, mudslides will continue to occur this summer. Within the last month, there have been two major mudslides both within a 10 mile area, near Pulga, California, 25 miles from Oroville. One of the mudslides caused a Union Pacific train to derail and three cars containing wine, tequila, furniture and food tumbled into the river.
Union Pacific Railroad Company (UP) is a contributing risk factor as well, as it reels from the disastrous consequences of its merger with Southern Pacific Railroad Company. After a series of nine fatalities to its workers, significant safety problems were identified by the Federal Railroad Administration (FRA). The September 10, 1997 report, issued by FRA found that UP had a "serious breakdown in its ability to implement basic railroad operating procedures and practices essential to safe railroad operations." With unresolved gridlock and other isssues still unresolved by UP, it is questionable whether UP will commit to the additional expenditures necessary to shore up the entire Feather River alignment affected by this year's El Niño.
Western Communities Against Nuclear Transportation (WCANT) an alliance of two dozen disarmament, anti-nuclear, and tribal organizations, including: Western States Legal Foundation, Grandmothers For Peace, Chico Peace and Justice, and Concerned Citizens of Plumas County in California, Citizen Alert in Nevada, and Ohngo Gaudadeh Devia and Downwinders in Utah, concerned about the safety of transporting high-level nuclear waste in untested truck casks through the San Francisco Bay and over the Feather River Canyon rail alignment, have identified crucial questions which must be answered by the Department of Energy (DOE) and the appropriate state agencies before any shipments are made.
This report addresses the histsory of nuclear waste container testing in the U.S., it describes how the type of cask to be utilized for the FRRNFS Program, Nuclear Assurance Corporation's Legal Weight Truck cask (NAC LWT) have been tested, and the NAC LWT's actual track record, and poses questions for the DOE and state agencies to answer so that the public can can bear witness to the decisionmaking with regard to the container that will be used to transport spent fuel over the Pacific Ocean, through the San Francisco Bay, and inland over the rails to be stored temporarily at the DOE's Idaho National Engineering and Environmental Laboratory near Idaho Falls, Idaho, while awaiting a permanent disposal site.
Despite the DOE's implications, in a film entitled,"Safe Transport, The Transport of Spent Nuclear Fuel," and in press releases from the Oakland DOE office, that casks to be used for the FRRSNF program have been put through a gamut of full-scale cask tests, they have not. The cask tests which the DOE refers to in the film and press releases were conducted in the late seventies using casks that had been retired for a decade.
"Casks of Uncertainty: The NAC LWTs," confirms that these spent fuel containers have never been tested at full-scale. It also clarifies and critiques U.S. cask testing in general, and the Sandia tests in particular, describing how edited films of the Sandia and Oakridge National Laboratory tests done in 1977 and 1978 omit crucial cask vulnerabilities revealed by the actual test results.
"Casks of Uncertainty," addresses the actual twenty year track record of the NAC LWT casks, and reveals some of the basic design, manufacturing, and human error flaws to which the NAC LWT casks have been prone.
Along with "Casks of Uncertainty" WCANT is submitting to the DOE and appropriate state agencies a list of questions dealing with:
Critique of DOE's Presentation Materials
The film, Safe Transport: The Transport of Spent Nuclear Fuel," is misleading because the DOE fails to disclose when and where the tests were performed on on which casks. If they had, the public would learn that the casks in the films were not the ones the DOE plans to use for the FRRSNF program, and that the only full-scale tests in the U.S. were done twenty years ago on different casks, ones that had been retired for a decade prior to the twenty-year-old tests. In addition to the fact that the casks in the films were different from the ones to be used for the FRRSNF program, the edited versions of this films from which the video was created, fails to state the startling results of some of the tests to these obsolete casks.
Because the NRC allows small scale and computer simulations cask tests of spent fuel containers to substitute for full-scale tests, the DOE has few full scale test examples to convince the public that spent fuel transport is safe. Rather than telling the truth about the sparseness of their full-scale test data, the presentation materials (i.e., films, slides of the film, and the Feb. 16, 1998 press release) borrow from a full-scale test done in Britain on a completely different cask and from a scale model test without explaining that it is not a full-scale model, to supply confidence about the casks to be used in the FRRSNF program.
The February 16, 1998, press release from Oakland DOE states: "These types of casks have been dropped from a height of 2,000 feet onto a hard desert surface, dropped onto a steel spike, subjected to a head-on collision with a diesel locomotive traveling at 80 miles per hour, crashed into concrete walls at greater than 80 miles per hour, and subjected to a 1,475 Fahrenheit fire for 90 minutes, and tested to a water depth of 660 feet, all without breaching the casks."
What the DOE is referring to with the phrase "these types of casks" the DOE is a very broad classification for all spent fuel containers called Type B casks. Referring to tests done on other Type B containers allows the DOE to gloss over the fact that the NAC LWT has never been full-scale tested.
(1) 2000 ft drop tests: refers to full-scale tests conducted by Sandia National Lab (SNL), SAND75-0276, on two obsolete casks, the OD-1 used to ship and store from the Oak Ridge research reactor, and B of E 83 casks used in handling irradiated test capsules. It is surprising that the DOE is using this test for several reasons: one, the casks are so old they do not remotely resemble todays' casks, raising this issue of how tests on obsolete casks could yield anything other than obsolete results: and two, the conclusions of the test did not support that a 2000 ft drop test on a yielding surface would do as much damage as a shorter fall onto an unyielding surface. In the Sandia document, SAND75-0276, Air Drop Tesst of Shielded Radioactive material Containers, the author states, "Comparing these results with the observed results of the 2000 ft drop test leads the author to the conclusion that the 30 ft drop test on an unyielding surface is a more severe environment than the 2000 ft drop onto hard soil." Thus we can conclude that the fall distance is not as important as the density of the surface in which it falls.
In the video presentation, "Safe Transport: The Transport of Spent Nuclear Fuel," the DOE implies that the NAC LWT have been put through all the types of cask tests. "By computer simulation, scale model testing and full-scale prototype testing, the casks are submitted to a variety of destructive scenarios involving catastrophic collisions and stress factors." When in fact, "the NAC LWTs and all of the casks in use today have only undergone computer simulations of accident scenarios," said Marvin Resnikoff of Radioactive Waste Management and Associates, in a May 1998 interview.
DOE's use of the word, "independently tested" is misleading. One not familiar with with the field of cask design and testing would assume it refers to a process in which each manufactured cask is tested, as opposed to testing only one representative cask of the same design. Additionally, one unfamiliar with the field would not know that the DOE has ways to test without actually testing, ie, small scale and computer model test simulations, or that just computer models of tests are acceptable to the NRC.
Nuclear Regulatory Commission (NRC) Cask Requirements
For example, many truck and train accidents involve fires which burn for much longer time than 30 minutes at 1475 F. In "Nuclear Cask Testing Films Misleading and Misused," Audin stated this was due to the fact that rail and truck accidents often involve impacts that create shocks and sometimes fires, aided by vehicle fuel and/or flammable cargoes. One such severe accident in 1982 lasted for two hours with fires which burned 45 minutes over the 30 minute regulatory limit, when a gasoline tanker truck was involved in a collision with a stalled car in the Caldecott Tunnel, near Berkeley, California. Audin report cautions that "this accident involved a single tanker: today's relaxed trucking rules allow double tankers."(Oct.1991)
It is unclear how these casks could hold up immersed in water depths of 50 feet or more, since the public has not seen any computer models of immersion tests done at that depth. Many areas of the worlds oceans are deeper than the International Atomic Energy Act requirements for spent fuel containers, (200 meters.) Spent Fuel from the FRRSNF Program will travel through some of these areas. The DOE has conceded that there may be areas on the ocean floor where spent fuel would not be retrievable. The fact that the same requirements apply both to truck and rail casks is problematical. It is not uncommon for a rail fire to burn for 12 hours. Because the NRC regulations apply to both casks used for highway and rail transport, the DOE can use a truck cask designed for highway transport if that is what is available, for spent fuel shipments over the railroad. In the case of the FRRSNF Program, the NAC truck casks will carry spent fuel by boat and rail and have the potential to be involved in a maritime or rail fires which can typically burn for much longer durations.
The Sandia and Oak Ridge National Laboratories Full-Scale Tests
Notwithstanding the fact that the tests are antiquated, films of these tests have generated controversy since they were made in the late seventies. Audin's Oct. 1991 critique of these tests describes some of the problems.
Additional problems with the tests identified by Audin in the Oct. 1991 report are:
"It is the misuse of these films to represent that the casks are safe that is objectionable... The high speed collision tests represent only two of a larger number of accident scenarios that need to be analyzed to assess the safety of the spent fuel cask transportation."
For example, two of the three casks are only subjected to end-on collisions against a flat surface, in which parts of the truck or cask enclosure absorb much of the impact. The other involved a sideways strike by a train, but no actual collisions occur between the cask and a reinforced body, like a bridge abutment, an accident that where the impacts to the cask would be greater.
The fresh fuel which was used for the tests, taken from an assembly from the merchant ship SS Savannah, was clad in stainless steel, a characteristic of the Sandia tests which may have relevance to the FRRSNF Program, since some fuel which the U.S. will be taking back is clad in steel. But again, because the tests were performed on cold, unirradiated fuel rods, the cladding would pose no potential problems, as it would with irradiated spent fuel rods in which the cladding would have become brittle, and possibly cracked, from the heat and pressure of real spent fuel.
Box 5-9 of TNNG, critiques other aspects of the Sandia tests/films. One frame shows a technician checking for radiation before and after the test. But because fresh fuel was used, only one-millionth as radioactive as irradiated fuel, the author noted that there would be little likelihood of detecting much radiation from it, even if the cask had been breached, unless the geiger counter was directly over the crack.
The fact that these films were edited down from a 14 minute to a 4 minute version was and continues to be problematical. The Oct. '91 Audin report said. "The shorter versions eliminated qualifying statements contained in the longer version, and created false impressions."
The editied versions of the film did not include the cask vulnerabilities exposed by the Sandia and Oakridge full-scale tests, the Audin Oct. '91 report said. Omissions by the narrator regarding cask vulnerabilities exposed by the Sandia and Oak Ridge full-scale drop, crash and burn tests include:
Another glaring omission is the fact that the casks which the DOE used for the Sandia tests were designed to different standards than casks today. These casks were built in the sixties before the Atomic Energy Commission changed the standards. Under the old standard, casks were designed to withstand stressses, including a one-hour high temperature fire, instead of today's 30 minute requirement.
Today's casks are designed to use impact limiters, which are a kind of shock absorber on each end of the cask. The casks used in the Sandia film were designed to remain intact without these impact limiters. In addition to being tested with impact limiters, the casks in the film utilized radiative cooling fins, not present in the NAC LWTs, and the tie downs used in the crash tests were of a superior design, much stronger than chain connections typically used today, and held the cask in a position that maximized absorption of the impact by the truck, instead of the cask. (Oct. 1991 Audin)
Lindsay Audin wraps up the Oct. 91 study with the conclusion: "By such critical omissions, the Sandia presentations withhold and distort information showing that the consequences of transportation accidents could be significantly more severe than claimed by the DOE and the nuclear industry." Audin characterized such productions as "propaganda" and said "that are clearly out of place in an honest debate over nuclear transportation safety."
Omissions in DOE's Presentation Materials Regarding British Tests
As in the U.S., the purpose for the tests done by the British Central Electricity Generating Board, (CEGB) in 1984, was to confirm computer code performance and materials evaluation, on a new design of cask, but also to affect more acceptance by the public for spent fuel shipments, by demonstrating the steps which a new package went through to gain approval. During this time, it was felt that the public perception of transport safety and the nuclear option in the UK was loosing ground," David Snedeker said in his Dec. 1990 report, entitled "Nuclear Waste Transportation Package Testing: A Review of Selected Programs in the United States and Abroad."
Public acceptance was important as a considerable amount of nuclear shipments go on continuously, 500 per year, between the two dominant utility companies in the UK and the Sellafield Reprocessing facility. There, like in the US, public concerns about the shipments could not be quelled by mere models of a very narrow scope of accidents.
Prior to this crash test, two drop tests, in full public view, had caused the CEGB to redesign its spent fuel containers. The first drop test on the edge of the Magnox cask had caused the lid to unseat. During the test, for a second the lid unseated and a spray of water was released (within regulatory limit) before the lid's seal reseated itself. Because the public was allowed to watch the test and see the release, it saw the potential for contamination. To CEGB's favor, they took the information received from the tests, and used it to eliminate lid movement aspects of the Magnox and the family of casks to which it is similar, even though the "leakage" was considered minor, a decision was made to improve the design further to eliminate the lid movement that occurred during the drop and crash test.
The CEGB found it hard to convince the public of the safety of the casks through the regulatory drop sequence tests. "The 30 ft drop onto an unyielding object, which looked like ordinary steel faced concrete had no real meaning," Snedeker wrote in his report. So they decided with the upgrades to the cask lid in place, they would go for a more true-to-life accident scenario: to hit the Magnox cask with a 140 metric ton locomotive and three railroad cars going 100 miles an hour.
The casks made it through the crash and fire caused by residual diesel fuel unscathed. Pressure tests recorded the same internal pressure as before the tests, thus, CEGB determined the payload would not have suffered and the casks were assumed to be safe. Even though the British cask test went well, the DOE's use of the film did not address the flaws. According to David Snedeker's 1990 report, the demonstration, while successful from CEGB's viewpoint, did not impart as much energy to the cask as the regulatory test sequence. Calculations after the impact sequence revealed that the 100 mph, 140 metric ton train only imparted about 1/3 the energy to the cask as the cask had seen during the thirty foot drop."
To summarize, it should be noted how the UK's response to public perception and cask results differs from that of the US. While it appears that Britain used their tests results to improve design, we see no evidence of that in the US. Instead, the U.S. downplays the results by editing them out, to not have to change designs. A glimpse of why this may be can be found in the Snedeker report. "An obvious question about such a test program has to be the effect of a national utility and government policy in developing such a program," acknowledges Snedeker. "It is clear that a for profit company could ill afford the cost of the CEGB program."
The NAC LWT Casks
The first discovery of a problem involved both a design/manufactured flaw and a problem with how Nuclear Assurance Corporation responded. On March 28, 1979, technicians at Duke Power Company in South Carolina discovered that the internal shell of a NAC-1 cask was bowed out of shape. They called NAC to report the problem, but NAC "made no effort to locate the other casks of the same type or to stop any impending shipments. " (TNNG -83)
"The source of the bowing problem remains officially undetermined," TNNG said. "It is reasonable to conclude however, that there is a generic fault with either the design or the manufacturing process as it is unlikely that two different companies separated by thousands of miles and several years could make exactly the same mistake." And the NRC issued a finding, "that crash worthiness could not be assured in a cask with such a defect."
In addition to the bowing problem, several NAC casks were found to have a problem with uneven shielding, another design or manufacturing process defect. When the first bowing problem was discovered at the Duke Power plant on the NAC-D model, another discovery was made by technicians that Nuclear Assurance Corporation had welded a copper plate onto the outside of the outer shell to increase shielding. Not only were these weldings done without permission by the NRC, the NAC should have known that an accident with a high temperature fire would have alloyed the copper to the steel and seriously weaken it. Because NAC did not report these welds to increase the shielding, years of shipments had taken place with these flawed containers. The author states, "It is unfortunate that such realizations came after more than five years' use of faulty casks, involving over 300,000 miles of shipments. (NAC 1979c)(Under Box 5-8 of TNNG).
TNNG describes another example of external shielding being added to the NAC-1s by NAC. The newest model, the NAC-E, arrived at the San Onofre reactor in California emitting 11 to 40 times the legal limit of radiation at several points on the cask surface. "Unbeknownst to the San Onofre technicians, this cask had been used four months previous to ship a leaking fuel assembly from the Oyster Creek, New Jersey, plant to a research facility near Columbus, Ohio."(TNNG Box 5-8). "The cask had become so severely contaminated in the process that external lead shielding had to be added. (NRC, 1981d)
TNNG reports several mysterious episodes with the NACs:
The same NAC-1D cask that was thought to have been bowed was one year later put back into use because it was found not to be bowed, was involved in some contamination resultant of cask "weeping." Weeping is a situation in which spent fuel from wet storage contaminates the cask surface even after it has been washed and inspected. According to Audin's 1988 report entitled "A Review of Human Reliability Issues In The Transportation Of Spent Nuclear Fuel," weeping is generally more harmful to cask handlers than to the public. But a case of it combined with human error, resulted in contamination of a major interstate in Pennsylvania.
During a routine swipe test of the NAC 1D, at the Oyster Creek plant on February 1981 an inordinately high contamination level (100 times greater than the regulatory limit ) was discovered on the surface of the NAC 1D. When it was discovered, the decision was made to decontaminate the cask and apply a layer of heavy paint designed to hold contamination in place for the next phase of the cask's journey, to the Batelle Laboratories in Ohio, and an NRC inspector on duty at the time observed the process. However, according to Resnikoff(TNNG, Box 5-8), "the wrong type of paint was used." The paint began dissolving off the cask during a rain storm in Pennsylvania. When the drivers noticed the peeling paint, they continued on perhaps unaware that surface contamination was probably being spread on the highway. NAC notified the NRC five days later, but no action was taken and the amount of radiation released was never determined.
To sum up the track record of the NACs, a majority of them had to be retired because of inner liner bowing, and both the NAC-D and NAC-E, two containers in use today, held fuel which contaminated the containers and external shielding had to be used. Thus, it is reasonable for citizens to be concerned about their use to ship FRRSNF. It is also plausible to suggest that the DOE chose them not on the merits of their safety record, but rather because there are very few casks available for commercial shipments.
Critique of the NRC's Modal Study (NUREG/CR-4829)
In 1986, the NRC commissioned the Lawrence Livermore National Laboratory (LLNL) to complete some new tests on spent fuel containers to test the standards responding to criticism that the standards are only theoretical because of the lack of full-scale tests to confirm them. LLNL used a cask similar to NAC truck cask for the Modal Study, NUREG/CR-4829, "Shipping Container Response to Severe Highway and Railway Accident Conditions."
Thus, flaws exist in the premise of the Modal Study. However, these are not the only flaws to be identified. Audin's critique calls into question the integrity of the study(Dec. 1990). In his executive summary he writes, "while an improvement over many prior efforts in this area, (such as NUREG -0170), it unfortunately fails to create a realistic simulation either of a shipping cask, the severe conditions to which it could be subjected, or the potential damage to the spent fuel cargo during an accident."
It is misleading for the DOE to rest its case on the safety of the NAC's for the FRRSNF Program on the Modal Study, since the DOE only used design similarities to the NAC LWT in the Modal Study. Audin states, "Since no experimental work was to be performed, a simplified cask design was created to be used as input to various computer simulations of impact and fire"(1990). Thus, only aspects of the NAC, a lead-lined cask with steel inner and outer shells, surrounded by a water neutron shield, was used in the preliminary analysis, and in the secondary analysis, a more detailed version of the cask, using materials and dimensions nearly identical to the NAC-1 container, was used. But it is important to note that the experiments were not done on real NACs.
Audin's 1990 study finds three basic problems with the Modal Study cask analysis:
Critique of the Accident Scenarios of the Modal Study
"It is important that the reader avoid also 'signing on' to the 10CFR71 perspective while thought is given to the accident conditions that could realistically prevail," Audin goes on to say, "the degree to which LLNL does so causes the Modal Study to lose some of its credibility."
While LLNL examined a number of accident scenarios, using the 10CFR71 performance tests as a starting point, "it discounted the need to examine criticality after a collision and the immersion in water (the final 10CFR test), because its probability calculations indicated that such a scenario would occur only once in ten million years" (Audin).
Two other critiques of the Modal Study by Audin include (December 1990):
LLNL also overestimates the neutron shields ability to retain its shape when the cask is struck by an object or if the cask strikes a flat surface, in that it does not discuss a puncture and the loss of the dead air space which would allow entry of hot gases into the empty shield. Even if only a portion of the insulating space is lost when the shield collapses, that would cause very uneven heating and expansion of the lead, not necessarily near the volume allowed for such expansion.
Similar impact by a train sill or other hard object could rip away part of the outer layer of the neutron shield, once again creating a pathway for increased and uneven heat transfer. In general, LLNL found lead to yield slight warpage of the inner and outer shells but not enough to create major strain. But Audin looks at the synergistic effect of the lead melt to weaknesses in welds and penetrations which could be affected by this local lead expansion.
Outer Cask Shell
A sidewise impact with a bridge abutment or similar unyielding structure was given scant analysis, due to LLNL conclusion that such a collision is "remote." In LLNL's response to criticism from the Los Alamos National Laboratory peer review of the Modal Study, LLNL stated that "this type of impact would be similar to that calculated for an impact with a train sill, or the front of a locomotive. (L.E. Fischer, LLNL, 1/16/87)
Audin's statement is that this opinion is not acceptable (Dec. 1990). "A train sill impact involves contact over a very small area with an object having limited kinetic energy". A bridge abutment is essentially an unyielding column. Impact would yield a great deal of lead movement and bending stresses, as the cask would continue to move while the center of the cask rapidly slowed down.
As if the Modal Study was a test in a vacuum, it fails to utilize information gained from full-scale testing eight years prior with regard to the vulnerability of cask welds to impacts. LLNL's analysis of the outer cask shell assumes that, at all points on its surface, it maintains its ability to yield to strain without breaking but the study failed to acknowledge that real casks have numerous welds that may be weaker than pure steel. (Audin 1990)
To conclude the critique of the Modal Study, Audin states that "While the basic simulation of striking a flat surface may be acceptable, the puncture study is not." The I-beam test failed to adequately model a fall of the cask onto train tracks, in that the I-beam in the test height equaled the cask diameter. In real life, impact would involve contact with the top side of a much smaller I- beam, not its end, and over a considerably different surface area, changing the dynamics of the study.
Failure to fully investigate puncture and cracking of the outer shell creates the rationale for avoiding consideration of the loss of shielding, a very serious potential problem.(Audin 1990) The only mechanism for major lead movement covered by the Modal Study is slumping due to an endwise impact of the cask onto a hard surface. Since opening of the outer shell may be a realistic possibility, examination of slumping effect alone in insufficient analysis upon which to base the rest of the study.
Two other problems in the lead shielding were identified in the Audin critique of the Modal Study not accounted for by LLNL as a result of impact: lead's heat capacity and how it alloys with nickel in the steel at a certain temperature. As was discussed earlier, the loss of the neutron shield greatly accelerates heat transfer and the likelihood of rapidly reaching very high temperatures at the inner surface of the outer shell. But the Audin critique states that "Even if the exterior steel layer of the neutron shield was still intact , LLNL apparently did not examine the temperature at the point of surface contact between the lead and the steel outer shell and thus did not consider the potential for alloying at that point." Lead will alloy with the nickel in steel outer shell and sufficiently weaken it, causing it to become embrittled and cracks to form which could lead to lead loss. Because only a very thin layer of lead needs to reach 1050 F for this phenomenon to occur, LLNL should have more thoroughly examined when that point would be reached in order to properly assess its likelihood.
According to Audin, the heat capacity of lead creates another condition not covered in the Modal Study. All fire simulations in the study examined the temperatures of the lead mid-points during the fire. Since the temperature of the spent fuel will always be lower at this time due to the buffering effect of the lead, a short fire that doesn't yield a mid-point temperature of 650 F during its duration is not considered to cause fuel rod bursting or oxidation. Examination of the temperature after the fire is out could be crucial, however, to assessing the fuel rod condition since a DOE sponsored study by Pacific Northwest Laboratories (PNL-2588) examined the fuel temperature during and after a fire and found the highest point was reached hours after the fire was out, due to the delayed heat transfer into the inner shell and the insulating effect of the lead.
In the PNL study, very high temperatures resulted, sufficient to burst the rods. In Audin's final comments, he states that LLNL should have examined short fires that cause lead mid-point temperatures of less than 650 degrees to assess their delayed temperature at the spent fuel. If short fires eventually yield high internal temperatures, the likelihood for a significant release is heightened, since short fires are much more common than long duration fires (at least according to the distribution used by LLNL).
The author notes that the actual cask simulated by LLNL, the NAC-1 cask, suffered from a bowing of the inner shell. He also clarifies that 4 out of 7 were taken out of service due to this problem, and that it was not noticed until several hundred shipments had been made. Audin also speculates that had such a container been involved in a severe endwise impact, the bowing could have created a vulnerability to bending or buckling of the shell, which could damage the fuel rods, and lead to leakage in the shell. Damaged fuel rods coupled with puncture or lead melt could have serious consequences, yet this scenario was not explored in the Modal Study, yet another example of the disconnect between what is known about the actual casks from real experience and the models in the Modal Study.
Welds were also found to be overlooked in the Modal Study, even though Audin states that those involved in the installing the tubing also create vulnerabilities, as evidenced in the Oak Ridge drop tests and in earlier prototypes of British casks. In these tests without impact limiters, he states, "a cask suffered cracks in its welding along its drain lines that extended from the inner shell our to the surface of the containers." While the steel around the welds stood up, the welds did not.
Audin's Final Comments of Modal Study
Conclusion: Lingering Uncertainties
Given the lack of full-scale tests for the NAC LWTs and the NAC LWTs track record, it is difficult to have confidence in the NAC LWT's for upcoming spent fuel shipments. Additional full-scale testing is necessary to study unresolved factors still from the Modal Study:
The need for more studies to determine the safety of cask welds is needed, since cracks in welds occurred in both the Sandia and British drop tests. In addition, the Modal Study made no differentiation between the welding material and unworked steel. In general, all of the components of the Penetration Subsystems (valves, welds and tubing) were assumed to be protected by design features.
DOE has not dealt with many important factors which could potentially jeopardize the safe transport of nuclear waste. Some factors include:
Because the casks in question have not been tested at full-scale, and the full-scale tests on the obsolete casks represented by the DOE may or may not have any relevance to modern type B casks, it is important that DOE commit to the full-scale testing of these containers. While planning for the full-scale tests, the DOE should at minimum explain to the public areas in which the NAC LWT casks are similar to or different to the ones shown in the films, and answer questions as to why the NAC LWT were chosen to ship the spent fuel from overseas.
Because of the risks related to spent fuel containers, the concerns in this report speak for the need to review an alternative under the U.S. DOE's Off- Site Fuels Policy, not adequately explored in its Environmental Impact Statement for the project. The alternative to store the spent nuclear waste of other nations on-site under strict international safeguards was given scant attention because it was deemed too costly. However, in order to foster public understanding for why the U.S.must take back the spent fuel of 41 nation, the DOE must make the case -- on a country to country basis -- that it is more dangerious to leave the spent fuel where it is than to bring it here.
The burden of proof is on the DOE to demonstrate how it is less dangerous and expensive to ship high-level radioactive waste half way around the globe in untested containers, jeopardizing the earth's land and water to the possiblility of a spill, in order to store it at a temporary site in the U.S. The federal government has a duty to its citizenry to thoroughly balance the risks of nuclear proliferation against the risks of a nuclear accident caused by the transportation of spent fuel within our communities.
Questions Which Need to Be Answered
DOE CASK TESTING CLAIMS IN PRESS RELEASES AND VIDEO
Audin, Lindsay, Nuclear Waste Shipping Container Response to Severe Accident Conditions: A Brief Critique of the Modal Study, December 1990.
Audin, Lindsay, A review of Human Reliability Issues in the Transportation of Spent Nuclear Fuel, 1988.
California Public Utilities Commission, Headwaters Summer 1997, Volume 21, Issue 3
Code of Federal Regulations, 10CFR71.
Department of Energy, Proceedings of the Fifth International Symposium on Packaging and Transportation of Radioactive Materials, Full-Scale Simulations of Accidents on Spent Nuclear Fuel Systems, Las Vegas, Nevada.
Department of Energy, Safe, Transport, The Transport of Spent Nuclear Fuel video.
Department of Transportation, Emergency Response Guidebook, 1984,
Lawerence Livermore National Laboratory (commissioned by Nuclear Regulatory Commission), Container Response to Severe Highway and Railway Accident Conditions, NUREG/CR-4829, 1986.
Pacific Northwest Laboratories (DOE sponsored study), PNL-2588.
Resnikoff, Marvin and Lindsay Audin, The Next Nuclear Gamble
Sandia National Laboratory, Air Drop Test of Shielded Radioactive Material Containers, SAND75-0276.
Snedeker, David, Nuclear Waste Transportation Package Testing: A Review of Selected Programs in the United States and Abroad, 1990.
Contact Citizen Alert for more information.