Sunday, February 12, 2012

The Mark I Containment: Recent Insights from Japan

Last March 11, the Tohoku-Chihou-Taiheiyo-Oki Earthquake and Tsunami struck Japan's east coast. In its wake, three nuclear reactors at the Fukushima Daiichi Nuclear Power Station could not be shut down safely. The station's six reactors are boiling water reactors. Three of the most impacted units, Units 1 - 3, were online at the time of the earthquake; the other units were shutdown for inspection.

Aerial view of Fukushima Daiichi Nuclear Power Station, Oct. 3, 2008. The reactor buildings of Units 4 to 1 are in the foreground, and of Units 5 and 6 in the background (courtesy
The nuclear fuel cores in the operating reactors, however, melted down, generating hydrogen from zirconium fuel rod cladding/steam reactions and the radiolysis of water. The hydrogen accumulated in the buildings, leading to explosions and partially destroying the reactor buildings. The explosions released massive amounts of highly-toxic radioactive material into air, ground and ocean.

Since then, the power station's operator Tokyo Electric and Power Co. (TEPCO) has striven to diminish the releases of radioactivity, made progress with clean-up and stabilization of the reactor buildings and achieved to reduce the cooling water temperature in the reactor pressure vessels to less than 100 °C. TEPCO estimates that decommissioning the reactors will take three or four decades.

Roughly 80,000 residents living within a 20-kilometer exclusion zone declared by the government eleven months ago still cannot return home permanently (see Chris Meyers' report with the title "A year on, only brief home visits for Japan nuclear evacuees" published online by Reuters Feb. 13, 2012). The long-term impact on public health and the environment are difficult to estimate, and the costs of the disaster are unfathomable.

Aerial view of Fukushima Daiichi Nuclear Power Station on March 24, 2011, after hydrogen explosions devastated the upper floors of the reactor buildings of Units 1 (background), 3 and 4 (foreground) in the wake of the March 11 earthquake and tsunami. Unit 4 was offline for inspection. TEPCO believes that hydrogen seeped into its building from Unit 3 via standby gas treatment system piping. The building of Unit 2 lost a blowout panel on the eastern side and was spared (courtesy

The other day I came across TEPCO's interim report with the title "Fukushima Nuclear Accident Analysis Report (Interim Report)" released Dec. 2, 2011, providing the most detailed narrative of the event sequence resulting in the fuel meltdowns and the explosions. The report unequivocally demonstrates that vital functions of the Mark I primary containment vessel (PCV) failed.

Schematic drawing of a boiling water reactor 3 (BWR-3) similar to Fukushima Daiichi Nuclear Power Station Unit 1, prominently featuring the Mark I primary containment vessel (PCV) comprised of the pear-shaped drywell (D/W) and the ring-shaped water-filled suppression chamber (S/C) also known as wetwell (courtesy Letsbereal's comment #461 on Prison Planet Forum).

The primary containment vessel (PCV) houses the reactor pressure vessel (RPV) in which the heat produced by a sustained nuclear chain reaction generates the steam for driving power generating turbines. The primary containments are supposed to absorb a potentially detrimental transient increase in pressure released from the reactor pressure vessel that is produced by the enormous decay heat emanating from the fuel after a sudden disruption of the nuclear chain reaction in an emergency shutdown known as SCRAM.

Schema of a Mark I primary containment system, showing the reactor pressure vessel (RPV) inside the containment composed of the pear-shaped drywell (D/W) connected to the ring-shaped, water-filled suppression chamber (S/C). In addition, the schema illustrates the paths for so-called hardened venting as a last resort to relieve pressure from the primary containment, including the motor-operated (MO) and compressed air-operated (AO) valves involved (courtesy TEPCO).

At Fukushima, seismic sensors triggered the reactor SCRAMS within seconds after the arrival of the first quake jolts. The Mark I containments consist of a pear-shaped drywell (D/W) which surrounds the reactor pressure vessel (RPV). The drywell is connected with large radial pipes arranged like spokes on a hub in a wheel to a ring-shaped suppression chamber (S/C) roughly half-filled with water. Pressure building up in the reactor pressure vessel can be relieved with safety relieve valves (SRVs) through the drywell into the suppression chamber's water pool, thus removing condensible gases and lowering the pressure in the primary containment (Lahey and Moody, 1993).

Concerns were raised in the U.S. already in the early 1970s that the Mark I primary containment system might not be capable of entirely absorbing the extreme pressures that could build up under adverse circumstances (see the famous memo of Stephen H. Hanauer, DRTA, U.S. Atomic Energy Commission with the title "Pressure Suppression Containments" dated Sep. 20, 1972). A series of improvements followed. For example, containments could fill with incondensable gases produced as decay products and by radiolysis or, in the worst case, when the fuel rod cladding melted and steam reacted with the cladding material as would happen at Fukushima Daiichi Nuclear Power Station. As a precaution, Mark I containments were retrofitted with hardened vents, permitting the operators to release pressure from the primary containments directly into the atmosphere, should the suppression chamber fail to lower excessive pressure. The Japanese reactors were outfitted with such vent lines exiting from the drywell as well as from the suppression chamber (see schema above).

Below I excerpt passages verbatim from TEPCO's December interim report on the accident that pertain to the company's attempts of controlling reactor pressure at Fukushima Daiichi Nuclear Power Station in the days after the earthquake. Annotations and edits are bracketed in curly braces. The vigilant reader will note that the pressure relief provisions noted above could not reduce the pressure in the reactor pressure vessels adequately.

{pp. 59:}
8.1 Response Status at Fukushima Daiichi Unit 1
  • After the tsunami, monitoring of reactor water level could no longer be conducted, and at 21:19 on March 11, temporary batteries were connected, enabling reactor water level to be monitored. Furthermore, the valve for starting up the IC {isolation condenser} was operated at around 18:00 and 21:00. At 23:00, in front of the air lock on the north side of the first floor of the turbine building, 1.2mSv/h was measured, and at the air lock on the south side, 0.5mSv/h was measured.
  • The D/W pressure was verified using power from a small generator, and there was the possibility that it might exceed 600 kPa[abs]. At 0:06 on March 12, the site superintendent (director, of ERC {emergency response center} at the power station) gave instructions to proceed with preparations for venting the PCV. At 0:49, because there was a possibility that the PCV pressure may exceed the maximum operating pressure (maximum operating pressure of 528 kPa[abs] (427 kPa[gage])), the site superintendent deemed that the condition fell under an event corresponding to Article 15 of the Nuclear Disaster Act (abnormal rise in PCV pressure).
  • On March 12 at around 1:30, the Prime Minister, the METI {Ministry of Economy, Trade and Industry}, as well as the NISA were notified of the implement{at}ion of the PCV venting for Units 1 and 2, and it was accepted.
  • On March 12 at 5:46, alternative cooling (freshwater) was started using a fire engine pump.
  • On March 12 at 9:04, venting the PCV for depressurizing of the D/W was started; however, inside of the reactor building was already a high radiation dose environment. At around 9:15, the motor-operated valve (MO {motor-operated} valve) on the venting line of the PCV was operated manually in accordance with the procedure manual so that it was 25% open. Moreover, workers headed into the field in order to manually open the air-operated value (AO {air-operated} valve), which is on the venting line from the S/C. However the radiation dose was high, and the operation could not be carried out. Consequently, a temporary air compressor was set up for operating the air-operated valve, and the PCV venting was carried out.
  • On March 12 at 14:30, on confirming that the D/W pressure dropped, it was deemed that venting of the PCV was successful.
  • On March 12 at around 14:54, the site superintendent ordered the injection of seawater into the reactor.
  • Subsequently, on March 12 at 15:36, an explosion, which was thought to be attributable to hydrogen gas, occurred in the upper structure of the reactor building, and the roof and outer walls of the refueling floor (top floor) were damaged. This explosion damaged the hose for seawater injection, and workers were evacuated from the field and confirmation of their safety was carried out. The restoration and preparation work was suspended until the field condit{i}ons could be verified. During these processes, radioactive materials were released into the environment; therefore, the radiation dose in the area surrounding the site increased.
  • On March 12 at 19:04, a FP {fire pump} line was used, and the seawater injection was commenced.
{pp. 68:}
(3) Response Status for PCV Venting at Fukushima Daiichi Unit 2 [Attachment 8-5]
  • On March 14 at 22:50, because the D/W pressure exceeded the maximum operating pressure of 427 kPa[gage], the site superintendent determined that an event corresponding to Article 15 of the Nuclear Disaster Act (abnormal rise in PCV pressure) had occurred.
  • While the D/W pressure tended to increase, the pressure in the S/C was stable at 300 to 400 kPa[abs]; however, the pressure between D/W and S/C would not equalize. The S/C pressure was lower than the pressure to operate the rupture disk while the D/W pressure was increasing; therefore, on March 14 at around 23:35, a decision was made on a course to conduct PCV venting by opening the air-operated valve (bypass valve《(4)》) on the vent line from the D/W.
  • On March 15 at around 0:02, operators opened the air-operated valve (bypass valve《(4)》) on the venting line from the D/W; however, a few minutes later, it was confirmed to be closed. The D/W pressure did not decrease from 750 kPa[abs] but remained high, and no effect from the venting was shown.
  • At between 6:00 and 6:10, a large explosive sound occurred. At almost the same time, <>the pressure of the S/C showed 0 MPa[abs] (Described in “9. Plant Hydrogen Explosion Assessment,” and the explosive sound is believed to have resulted from the explosion at Unit 4).
  • Meanwhile, the D/W pressure maintained at 730 kPa[abs] as of 7:20.
  • The D/W pressure as of 11:25, which was when the next measurement was made, had decreased to 155 kPa [abs], and it is thought that during this time, the gas in the PCV was released into the atmosphere in some way, and the monitoring car reading near the main gate drastically increased.
{pp. 76:}
(3) Response Status Pertaining to Venting of PCV at Fukushima Daiichi Unit 3 [Attachment 8-6]
  • On March 12 at 17:30, the site superintendent ordered the beginning of preparations for the PCV venting. (Review of the procedures and the necessary valve locations were confirmed along with other details.)
  • On March 13 at 4:50, in order to open the air-operated valve on the vent line from the S/C, the portable generator being used for temporary lighting in the MCR, was used as a power source for the solenoid valve, and it was forcibly energized.
  • On March 13 at 5:15, the site superintendent ordered to complete the vent line up except for the rupture disk.
  • When operators went to the torus room (where the S/C is installed) to confirm the valve opening condition, it was fully closed. Accordingly, beginning at 5:23 on March 13, the compressed air cylinder was replaced, and the vent valve was then able to be opened.
  • On March 13 at 5:50, a press conference was commenced regarding the implementation of PCV venting, and at 7:35, TEPCO reported to the government agencies and other such institutions the assessment results of radiation exposure to the area surrounding the power station when the PCV venting was to be implemented.
  • At around 8:35, the MO valve on the vent line from the S/C was manually opened to 15 %.《(1)》Standard procedures call for the vent to be opened to 25%; however, this was lowered in order to prevent excessive decrease in PCV pressure drastically.
  • At 8:41, alignment of the vent lineup, excluding the rupture disk, was completed. However, PCV pressure was too low to rupture the rupture disk. (427 kPa[gage]) Therefore, the system would not vent (waiting for rupturing the rupture disk), and the vent system alignment was kept open《(2)》and PCV pressure was monitored.
  • At 9:24, PCV pressure drop was verified; therefore, at approximately 9:20, it was determined that the S/C had been vented. [Attachment 8-7]
  • On March 13 at 11:17, due to decreasing pressure of the compressed air cylinder, the aforementioned air-operated valve《(2)》was closed. Therefore, the air cylinder was replaced and the valve opened again at 12:30.
  • After that, the valve needed to be maintained to be an open; however, operators could not keep the valve open due to the difficulty of the high room temperature at the torus room.
  • On March 13 at around 17:52, workers headed to the field to set up a temporary compressor at the truck bay of the turbine building and connected it to the instrument air system. At around 21:10, as the D/W pressure decreased, it was deemed that the air-operated valve《(2)》on the vent line from the S/C was opened.
  • On March 14, beginning at around 2:00, the D/W pressure was uptrending; therefore, at 5:20, another air-operated valve《(3)》(bypass valve), which was also on the vent line from the S/C, was opened and it was confirmed to be open at 6:10.
  • {From page 70:}
  • Subsequently, on March 14 at 11:01, a hydrogen explosion occurred in the reactor building, and everything above the refueling floor and the south and north outside walls of one floor below of the refueling floor was damaged. During this event, radioactive materials were released into the environment, and the radiation dose around the power station increased.
Taking TEPCO's findings highlighted above together, the three stricken reactors' suppression chambers seemingly failed to fulfill their intended function. The pressure released from the reactor pressure vessels into the primary containments did not sufficiently equilibrate with the suppression chambers and was not absorbed in their pools of water. The retrofitted hardened venting provisions did not help reduce primary containment pressure adequately either.

As a consequence, the pressure in the reactor pressure vessels could not be lowered enough to permit the operator to inject the amounts of water needed to keep the fuel cores covered and prevent the fuel from melting. Whether the suppression chambers failed because of inherent design shortcomings or because of explicit seismic impact remains an open question.

As attested in the U.S. Nuclear Regulatory Commission (NRC) meeting discussed below, the design of the Mark I primary containment system required that comparably small-volumed containments absorb extraordinary pressures. The Fukushima reactor disaster may bear witness that the system could not meet this challenge.

In response to the reactor disaster in Japan, Beyond Nuclear requested a public hearing on the safety of Mark I containments from the NRC in its petition with the title "10 CFR 2.206 Petition to immediately suspend the operating licenses of GE BWR Mark I units pending full NRC review with independent expert and public participation from affected emergency planning zone communities" submitted Apr. 11, 2011, and the NRC convened a series of public meetings at the agency's headquarters in Rockville, Maryland. The roughly 90-minute October 7 2011 meeting with the title "Beyond Nuclear 10 CFR 2.206 Petition Public Meeting" can be viewed in the video below.

About 39 minutes into the meeting, retired nuclear engineer and former General Electric program project manager for Mark I containment system safety Dale Bridenbaugh phoned in his opinion as a co-petitioner. General Electric developed the Mark I containment system. Mr. Bridenbaugh submitted an impressive plea for the urgent review of the containment system in its current design, which he believes has not improved to meet the safety goals set in his time with General Electric four decades ago. I transcribed his submission to the best of my ability. My annotations are bracketed in curly braces:
“Thanks for this opportunity. As I said this is Dale Bridenbaugh. I want to reiterate I am appearing as an individual. I am not representing anyone other than myself.
I am here today to present my opinion on the ongoing operation of the 23 Mark I containment boiling water reactors in the United States. My opinion is based on some 40 years of experience in the commercial nuclear power industry, approximately 20 years as an engineer and manager for General Electric Company's nuclear business, and another 20 as a private consultant in nuclear plant studies performed for more than 20 state agencies and several foreign countries.
The first generation of large boiling water reactors was built by General Electric in spherical dry containment vessels designed to contain the energy that could be released in the event of a break in the primary system. Dresden I, where I first worked, is a 200-megaWatt electric plant and was housed in a 190-foot {58-meter} diameter sphere which had a volume of nearly 3.6 million cubic feet {101,941 m3}. This same concept was followed with the next few plants. But as the design ratings were increased, the dry containment became problematic due to size and cost. With the higher energy content of the larger systems, dry containments were found not to be economically viable.
The Mark I pressure suppression system concept was developed and the resulting containment size was reduced by nearly a factor of 10. Even for unit ratings some five times greater {for comparison Fukushima Daiichi Nuclear Power Station Unit 1 produced 460 MW and Units 2 and 3 each 784 MW}. The result was a cheaper containment at the cost of difficulty in conducting required maintenance and inspections, and with less resistance to severe accident consequences.
In early 1975, the NRC issued letters to all licensed Mark I plant owners asking for assurance that the Mark I plants did in fact meet required licensing criteria, and that would include that the containments would provide essentially a leak-tight response to design-basis accidents.
This generic request was in part the result of information shared by GE with the NRC concerning testing of the Mark III containment concept and was backed up by some early failures at the first Mark I plants in operation.
The 16 Mark I utilities contacted GE for assistance in answering the NRC request. GE proposed that a safety re-evaluation program be initiated to determine the nature and extent of the problem, and I was tapped to be the project manager of the program. This program, called the Mark I owners program, began its work in the spring of 1975 and continued well into the 1980s. The intended function of the program was to develop accepted definitions of the unquantified design-basis accident loads and to develop appropriate modifications.
Throughout the program a great deal of uncertainty was encountered in quantifying the loads and the containment response to those loads. The early effort devolved into an exercise in defending continued operation of the plants through arguments of the low probability of the possible event. The owners group program finally resulted in NRC-approved Mark I hydrodynamic load definitions, and subsequent fixes were implemented to overcome the design deficiencies.
The fact remains, however, that the Mark I plants continued operations for as much as 12 or 13 years outside of the requirements under which they were originally licensed. The ongoing period of operation under uncertain safety conditions played a large part in my decision to resign in 1976 from the program and from GE.
The recent experience at the Fukushima Mark I units calls into question, again, whether those fixes, assuming they were properly implemented in Japan, are adequate to meet license requirements so as to safeguard the health and the safety of the public. Even the so-called hardened vent modification in the early nineties seems to have been inadequate at Fukushima to prevent hydrogen explosions and containment damage. It will be at least several years, if ever, before the full extent of the Fukushima accident sequences are known and understood. There are indications that some of the failures at Fukushima are not limited to the combined earthquake and tsunami effects, but may have been initiated by the seismic pulses alone. That remains to be seen.
It is unreasonable for all of the U.S. citizens who could be affected by a major accident at a U.S. Mark I plant to be held at risk for yet another period of years, when it is uncertain similar consequences could happen here. In my opinion, it is absolutely essential that commitments be made that plant-specific analyses be performed as soon as possible to consider the broad range of challenges the Fukushima accident presents to the 23 Mark I units in the U.S.
Further, date-certain limits should be issued for all currently licensed Mark I units so as to assure that unlimited periods of operation not be allowed to continue outside of appropriate licensing conditions. Thank you very much for your time.”

General Electric assures that the design of the Mark I containment system remains safe (see GE report with the title "Setting the Record Straight on Mark I Containment History" published online Mar. 18, 2011). Contrary to General Electric's assurances, TEPCO's experience offered in the company's interim report of last December seems to support the concerns that Mr. Bridenbaugh raised at the October 9 NRC hearing on Beyond Nuclear's petition.

I am indebted to the contributors of without whose input I could not have written this essay. I thank Dale Bridenbaugh for sharing his expertise and extensive experience with General Electric's Mark I containment system.

  • CNN's Matt Smith discussed GE's Mark I containment systems yesterday in his article with the title "U.S. nuclear plants similar to Fukushima spark concerns". The post contains informative illustrations and sparked roughly 1000 comments within 24 hours. In the article, GE is cited blaming the failures at Fukushima squarely on the devastating effects of the tsunami and reiterating that the company's Mark I containment system is safe. CNN is preparing to air a two-installment broadcast on the issue today and tomorrow evening 8 ET/PT on CNN Presents. I would not be surprised, if lawyers were already assembling briefs for possible litigation, law suites would be filed over damages, and imbroglios between the defendants would ensue similar to those we witness today in the aftermath of the 2010 Deepwater Horizon Accident in the Gulf of Mexico (see my post with the title "Energy & The Mind" published Apr. 27, 2010 (02/18/2011).


Watch some great animations on the performance of the Mark I containment system at Fukushima Dai-ichi Unit 1 and potential pitfalls.

No comments: