Nuclear accidents: Wikis

Advertisements
  

Note: Many of our articles have direct quotes from sources you can cite, within the Wikipedia article! This article doesn't yet, but we're working on it! See more info or our list of citable articles.

Encyclopedia

(Redirected to Nuclear and radiation accidents article)

From Wikipedia, the free encyclopedia

Pathways from airborne radioactive contamination to man

Nuclear and radiation accidents may be of various types. An example of nuclear accident might be one in which a reactor core is damaged such as in the Chernobyl Disaster in 1986, while an example of a radiation accident might be some event such as a radiography accident where a worker drops the source into a river.

In the period to 2007, sixty-three major nuclear accidents have occurred at nuclear power plants. Twenty-nine of these have occurred since the Chernobyl disaster, and 71 percent of all nuclear accidents (45 out of 63) occurred in the United States, challenging the notion that severe nuclear accidents cannot happen within the United States or that they have not happened since Chernobyl.[1][2]

Radiation accidents are more common than nuclear accidents, and are often limited in scale. For instance at Soreq, a worker suffered a dose which was similar to one of the highest doses suffered by a worker on site at Chernobyl on day one.[citation needed] However, because the gamma source was never able to leave the 2-metre thick concrete enclosure, it was not able to harm many others.[citation needed]

The web page at the International Atomic Energy Agency, which deals with recent accidents is [2]. The safety significance of nuclear accidents can be assessed and conveyed using the International Atomic Energy Agency International Nuclear Event Scale.

Contents

Overview

The abandoned city of Prypiat, Ukraine, following the Chernobyl disaster. The Chernobyl nuclear power plant is in the background.

In the period to 2007, sixty-three major nuclear accidents have occurred at nuclear power plants. The worst nuclear accident is the Chernobyl disaster which occurred in 1986 in Kiev, Ukraine. That disaster "killed at least 4,056 people and damaged almost $7 billion of property".[3] Radioactive fallout from the accident concentrated near Belarus, Ukraine and Russia and at least 350,000 people were forcibly resettled away from these areas. After the accident, "traces of radioactive deposits unique to Chernobyl were found in nearly every country in the northern hemisphere".[3]

Twenty-nine major nuclear accidents have occurred since the Chernobyl disaster, and 71 percent of all nuclear accidents (45 out of 63) occurred in the United States, challenging the notion that severe nuclear accidents cannot happen within the United States or that they have not happened since Chernobyl.[4][3]

Other serious nuclear and radiation accidents include the Mayak disaster, Soviet submarine K-431 accident, Soviet submarine K-19 accident, Chalk River accidents, Windscale fire, Three Mile Island accident, Costa Rica radiotherapy accident, Zaragoza radiotherapy accident, Goiania accident, Church Rock Uranium Mill Spill and the SL-1 accident.[5][6]

Accident types

Advertisements

Loss of coolant accident

Criticality accidents

A criticality accident (also sometimes referred to as an "excursion" or "power excursion") occurs when a nuclear chain reaction is accidentally allowed to occur in fissile material, such as enriched uranium or plutonium. The Chernobyl accident is an example of a criticality accident. This accident destroyed a reactor at the plant and left a large geographic area uninhabitable. In a smaller scale accident at Sarov a technician working with highly enriched uranium was irradiated while preparing an experiment involving a sphere of fissile material. The Sarov accident is interesting because the system remained critical for many days before it could be stopped, though safely located in a shielded experimental hall [3]. This is an example of a limited scope accident where only a few people can be harmed, while no release of radioactivity into the environment occurred. A criticality accident with limited off site release of both radiation (gamma and neutron) and a very small release of radioactivity occurred at Tokaimura in 1999 during the production of enriched uranium fuel [4]. Two workers died, a third was permanently injured, and 350 citizens were exposed to radiation.

Decay heat

Decay heat accidents are where the heat generated by the radioactive decay causes harm. In a large nuclear reactor, a loss of coolant accident can damage the core: for example, at Three Mile Island a recently shutdown (SCRAMed) PWR reactor was left for a length of time without cooling water. As a result the nuclear fuel was damaged, and the core partially melted. The removal of the decay heat is a significant reactor safety concern, especially shortly after shutdown. Failure to remove decay heat may cause the reactor core temperature to rise to dangerous levels and has caused nuclear accidents. The heat removal is usually achieved through several redundant and diverse systems, and the heat is often dissipated to an 'ultimate heat sink' which has a large capacity and requires no active power, though this method is typically used after decay heat has reduced to a very small value. However, the main cause of release of radioactivity in the Three Mile Island accident was a Pilot-Operated Relief Valve on the primary loop which stuck in the open position. This caused the overflow tank into which it drained to rupture and release large amounts of radioactive cooling water into the containment building.

Transport

Transport accidents can cause a release of radioactivity resulting in contamination or shielding to be damaged resulting in direct irradiation. In Cochabamba a defective gamma radiography set was transported in a passenger bus as cargo. The gamma source was outside the shielding, and it irradiated some bus passengers.

In the United Kingdom, it was revealed in a court case that in March 2002 a radiotherapy source was transported from Leeds to Sellafield with defective shielding. The shielding had a gap on the underside. It is thought that no human has been seriously harmed by the escaping radiation.[7]

Equipment failure

Equipment failure is one possible type of accident, recently at Białystok in Poland the electronics associated with a particle accelerator used for the treatment of cancer suffered a malfunction [5]. This then led to the overexposure of at least one patient. While the initial failure was the simple failure of a semiconductor diode, it set in motion a series of events which led to a radiation injury.

A related cause of accidents is failure of control software, as in the cases involving the Therac-25 medical radiotherapy equipment: the elimination of a hardware safety interlock in a new design model exposed a previously undetected bug in the control software, which could lead to patients receiving massive overdoses under a specific set of conditions.

Human error

A sketch used by doctors to determine the amount of radiation to which each person had been exposed during the Slotin excursion.

Human error has been responsible for some accidents, such as when a person miscalculated the activity of a teletherapy source. This then led to patients being given the wrong dose of gamma rays. In the case of radiotherapy accidents, an underexposure is as much an accident as an overexposure as the patients may not get the full benefit of the prescribed treatment. Also, humans have made errors while attempting to service plants and equipment which has resulted in overdoses of radiation, such as the Nevvizh and Soreq irradiator accidents. In Japan two minor millennium bugs came to light [6]

In 1946 Canadian Manhattan Project physicist Louis Slotin performed a risky experiment known as "tickling the dragon's tail" [8] which involved two hemispheres of neutron-reflective beryllium being brought together around a plutonium core to bring it to criticality. Against operating procedures, the hemispheres were separated only by a screwdriver. The screwdriver slipped and set off a chain reaction criticality accident filling the room with harmful radiation and a flash of blue light (caused by excited, ionized air particles returning to their unexcited states). Slotin reflexively separated the hemispheres in reaction to the heat flash and blue light, preventing further irradiation of several co-workers present in the room. However Slotin absorbed a lethal dose of the radiation and died nine days afterwards.

Lost source

Lost source accidents [7][8] are ones in which a radioactive source is lost, stolen or abandoned. The source then might cause harm to humans or the environment. For example, see the event in Lilo where sources were left behind by the Soviet army. Another case occurred at Yanango where a radiography source was lost, also at Samut Prakarn a cobalt-60 teletherapy source was lost [9] and at Gilan in Iran a radiography source harmed a welder [10]. The best known example of this type of event is the Goiânia accident which occurred in Brazil.

The International Atomic Energy Agency has provided guides for scrap metal collectors on what a sealed source might look like.[11][12] The scrap metal industry is the one where lost sources are most likely to be found.[13]

Others

Some accidents defy classification. These accidents happen when the unexpected occurs with a radioactive source. For instance if a bird were to grab a radioactive source containing radium from a window sill and then fly away with it, return to its nest and then die shortly afterwards from direct irradiation then a minor radiation accident would have occurred. As the hypothetical act of placing the source on a window sill by a human permitted the bird access to the source, it is unclear how such an event should be classified, as a lost source event or a something else. Radium lost and found[14][15] describes a tale of a pig walking about with a radium source inside; this was a radium source lost from a hospital.

Also some accidents are "normal" industrial accidents which happen to involve radioactive material. For instance a runaway reaction at Tomsk (see red oil) caused radioactive material to be spread around the site.

For a list of many of the most important accidents see the International Atomic Energy Agency site [16] .

Analyses of nuclear power plant accidents

A study published in Energy Policy in May 2008 shows that, in the period to 2007, sixty-three major nuclear accidents have occurred at nuclear power plants. Twenty-nine of these have occurred since the Chernobyl disaster, and 71 percent of all nuclear accidents (45 out of 63) occurred in the United States, challenging the notion that severe nuclear accidents cannot happen within the United States or that they have not happened since Chernobyl.[9][10]

The Nuclear Regulatory Commission (NRC) now requires each nuclear power plant in the U.S. to have a probabilistic risk assessment (PRA) performed upon it. The two types of such plants in the US (as of 2007) are boiling water reactors and pressurized water reactors, and a study based on two early such PRAs was done (NUREG-1150) and released to the public. However, those early PRAs made unrealistically conservative assumptions, and the NRC is now generating a new study: SOARCA.

Nuclear Regulatory Commission

Nuclear Regulatory Commission Headquarters and Regional staff members typically participate in four full-scale and emergency response exercises each year, selected from among the list of full-scale Federal Emergency Management Agency (FEMA)-graded exercises required of nuclear facilities. Regional staff members and selected Headquarters staff also participate in post-plume, ingestion phase response exercises. On-scene participants include the NRC licensee, and State, county, and local emergency response agencies.(http://www.nrc.gov/about-nrc/emerg-preparedness/exercise-schedules/nrc-ex-schedule.html) This allows for Federal Interagency participation that will provide increased preparedness during the potential for an event at an operating nuclear reactor.

The US Nuclear Regulatory Commission (NRC) collects reports of incidents occurring at regulated facilities. The agency currently (2006) uses a 4 level taxonomy to classify reported incidents:

  • Notification of Unusual Event
  • Alert
  • Site Area Emergency
  • General Emergency

Not all reportable events constitute accidents. Incidents which threaten the normal operation or security of a facility may be reportable but not result in any release of radioactivity.

The US Department of Energy uses a similar classification system for events occurring at fuel cycle plants and facilities owned by the US government which are therefore regulated by the DOE instead of the NRC.

NRC Alerts

NRC Site Area Emergencies

NRC General Emergencies

NRC ASP Analysis Program

The NRC established the Accident Sequence Precursor (ASP) analysis program in 1979 in response to the Risk Assessment Review Group report (see NUREG/CR-0400, dated September 1978). The primary objective of the ASP Program is to systematically evaluate U.S. nuclear power plant operating experience to identify, document, and rank the operating events that were most likely to lead to inadequate core cooling and severe core damage (precursors), if additional failures had occurred. To identify potential precursors, NRC staff reviews plant events from licensee event reports (LERs), inspection reports, and special requests from NRC staff. The staff then analyzes any identified potential precursors by calculating a probability of an event leading to a core damage state.[11]

(ref NRC Commission Document SECY-05-0192 Attachment 2 NRC: Policy Issue Information)

A "significant precursor" is an event that leads to a conditional core damage probability (CCDP) or increase in core damage probability (CDP) that is greater than or equal to 1 × 10–3. In other words given that the precursor event has occurred, the probability that a subsequent failure will cause core damage is = 0.001.

As of 24-Oct-2005 the "significant" precursor events (i.e. the worst category) were (listed from highest probability of occurrence 1 to lowest probability of occurrence 0.1%):[12]

Date CDP Plant Prompt fatalities Latent fatalities Notes
1979-03-28 1.000 Three Mile Island Unit 2 0 ~1 In the only civil light water reactor core damage accident to occur in the history of nuclear power, due to a combination of poor instrumentation design and operator error, along with the extremely common problem of stuck open power operated relief valves, Unit 2 at Three Mile Island suffered a loss of coolant accident compounded by operator mistakes that resulted in core damage. A loss of proper feedwater incident occurred and the reactor tripped; a stuck open power operated relief valve incident occurred (in the classical fashion); operators failed to isolate the PORV, inhibited auxiliary coolant pumps, and inhibited HP ECC due to a misinterpretation of instrument readings. Core uncovery and subsequent core damage were the inevitable consequences; after cohesive debris bed formation and consolidation (with zircaloy-induced coolant disassociation through reactive autocatalysis), operators found the gamma alarm going off, finally correctly assessed the situation and injected all available HP and LP ECC and AFW into the core at maximum rate, preventing melt-through of the RPV, though causing a FCI and subsequent deflagration event due to autocatalysis of coolant. Though no persons were immediately killed or injured, science estimates that perhaps 1 person will die before their normal age of death due to radiation-induced cancer caused by non-condensible fission product gasses.
1975-03-22 0.200 Browns Ferry Unit 1 0 0 (ref NRC IE BULLETIN NO. - 75-04A) A fire broke out in Browns Ferry Unit 1, due to highly flammable material (urethane foam) accidentally used as firestopping that burnt the data cables between the reactor and the control room. The control room's supervisory control and data acquisition equipment was cut off from the reactor, however, heroic efforts by the plant staff made the reactor safe by manually actuating reactor depressurization valves and creatively initiating residual heat removal using a quaternary system never meant to be used for cooling: the control rod drive hydraulic pumps were aligned with the condensate storage tanks and used as a last-ditch manual coolant injection system.
1978-03-20 0.100 Rancho Seco 0 0 (ref LER 312/78-001) A loss of proper feedwater incident led to reactor trip and subsequently operators making an overestimate of secondary coolant in the steam generators, leading the steam generators to dry out, in an incident that served as a near-miss precursor to the Three Mile Island accident. HP ECC injected prior to core uncovery or damage, however.
1977-09-24 0.070 Davis-Besse 0 0 (ref NRC LER 346/77-016) During a loss of proper feedwater incident, the power operated relief valve rapidly cycled between open and closed states, eventually leading to a stuck open power operated relief valve incident (in the classical fashion), and a consequent low water level in the steam generators, in another near-miss precursor to the Three Mile Island accident. HP ECC was injected prior to SG dryout, preventing full development of the situation.
1974-05-08 0.020 Turkey Point Unit 3 0 0 (ref NRC LER 250/74-LTR) No less than three auxiliary feedwater pumps failed to start when tested while the reactor was at power, due to improper maintenance, a highly suboptimal condition that could place the reactor at risk for further failures should an unscheduled contingent evolution occur.
1985-06-09 0.010 Davis-Besse 0 0 Due to a loss of proper feedwater at power, combined with a stuck open power operated relief valve incident (in the classical fashion). This was compounded by operator errors in inhibiting auxiliary feedwater to the steam generators, leading to the reactor being placed in a dangerous state. However, operators rapidly realized that the PORV was stuck open, and correctly isolated it, eventually recovered auxiliary feedwater, and injected ECC at maximum rate.
1978-11-27 0.010 Salem Unit 1 0 0 A transformer failure led to a Main Bus 1B undervolt. As such, false signals were sent to control equipment tripping the reactor, and several AFW pumps were unavailable. ECC was incorrectly injected, leading to low coolant temperature in the reactor, and consequently further ECC injection.
1976-07-20 0.010 Millstone Unit 2 0 0 Due to a low voltage on offsite power (brownout), when a main circulator started, the reactor tripped due to the offsite voltage dipping below the high undervoltage setpoint and causing electrical bus isolation, and subsequently loss of offsite power. Further, due to the high undervoltage setpoint, whenever a load was connected to the emergency diesel generators, the emergency diesel generators tripped due to undervolt conditions on them. As such, effective station blackout was threatened, a highly suboptimal condition that could place the reactor at risk for further failures should an unscheduled contingent evolution occur.
1975-04-29 0.009 Brunswick Unit 2 0 0 A stuck open power operated relief valve incident occurred, RCIC failed inoperable, HPCI failed to run due to high water level in torus, and half of RHR failed to activate. The reactor tripped due to MSIV closure rather than the reactor operators triggering a manual trip, as they should have done.
1981-04-19 0.007 Brunswick Unit 1 0 0 RHR was seriously damaged due to the failure of a baffle while the unit was in cold shutdown, while the other half of RHR was under routine maintenance. As such, decay heat could not be removed by RHR and alternate channels had to be used.
2002-02-27 0.006 Davis-Besse 0 0 Contractor incompetence in cleaning boric acid deposits from the reactor pressure vessel head led to extreme corrosion by leaving only a thin layer of Inconel standing between the 2400 psi pressure of the primary cooling system and the 14 psi atmospheric pressure containment. The possibility of a loss of pressure control or a LOCA was greatly elevated while this fault remained undiscovered. The reactor pressure vessel head consequently had to be replaced in toto.
1991-04-03 0.006 Harris Unit 1 0 0 It was determined that HP ECC had been unavailable for at least one refueling cycle due to a series of compounded failures in ECC. This represented a suboptimal condition that could place the reactor at risk for further failures should an unscheduled contingent evolution occur.
1983-02-25 0.005 Salem Unit 1 0 0 Automatic reactor trip failed in two separate circumstances several days apart. Though manual trip worked upon operator actuation, failure to automatically trip - a highly suboptimal condition - could place the reactor at risk for further failures should an unscheduled contingent evolution - such as a low reactor period incident - develop during post-critical ascension to power from cold shutdown.
1981-01-02 0.005 Millstone Unit 2 0 0 Partial loss of offsite power due to operator error, along with an emergency diesel generator trip due to operator error, leading to a partial station blackout. Further operator errors caused the power operated relief valve to be opened at 2380 psi, leading to primary system blowdown into the containment building.
1980-02-26 0.005 Crystal River Unit 3 0 0 Thermal-hydraulic instrumentation lost power and failed due to a short to ground. The power operated relief valve opened and stuck open (in the classical fashion), operators correctly injected HP ECC, and properly did not stop because insufficient data justified inhibiting ECC and risking PZR dryout. The PZR was pumped solid, the reactor coolant tank consequently backed up and eventually the rupture disk on it ruptured, and 43,000 gallons of primary water fell into the containment sump; this created a mess.
1978-03-25 0.005 Farley Unit 1 0 0 Low water levels in a steam generator led auxiliary feedwater to be called upon to function; it did not function when called upon to function. Other ECC channels were used that worked.
1977-12-11 0.005 Davis-Besse 0 0 It was discovered that auxiliary feedwater was not available during a routine test due to blown fuses and a mechanical binding in the governor of the pumps, a suboptimal condition that could place the reactor at risk for further failures should an unscheduled contingent evolution occur.
1975-11-05 0.005 Kewaunee 0 0 Auxiliary feedwater pumps were fouled due to ion exchange resin beads that migrated from the demineralizer into the condensate storage tank, and thus failed to start. Alternate channels were used to cover for this loss of auxiliary feedwater. This represented a suboptimal condition that could place the reactor at risk for further failures should an unscheduled contingent evolution occur.
1974-04-07 0.005 Point Beach Unit 1 0 0 Auxiliary feedwater pumps had plugged filters and did not provide adequate flow during shutdown.
1994-09-17 0.003 Wolf Creek Unit 1 0 0 Operators did not follow instructions and implemented two unpermitted simultaneous evolutions, leading to water inventory transfer from the reactor coolant system to the refueling water storage tank. Operators immediately mitigated the condition, but a temperature spike of 4oC was detected prior to the termination of the evolution.
1986-06-13 0.003 Catawba Unit 1 0 0 The design basis SBLOCA occurred at the chemical and volume control system and component cooling water heat exchanger joint, due to the failure of a variable letdown orifice outlet flange due to cavitation-induced vibration. ECC injected successfully and the SBLOCA was contained within the design basis.
1978-04-13 0.003 Calvert Cliffs Unit 1 0 0 Loss of offsite power occurred, and one of the two emergency diesel generators failed to start.
1985-05-15 0.002 Hatch Unit 1 0 0
1984-09-21 0.002 Lasalle Unit 1 0 0
1981-06-24 0.002 Davis-Besse 0 0 Bus E2 undervolt occurred due to maintenance error during control rod drive logic testing, and the reactor tripped. One auxiliary feedwater pump failed to start, and a main steam safety relief valve opened and stuck open (in the classical fashion).
1979-05-02 0.002 Oyster Creek 0 0
1977-07-12 0.002 Zion Unit 2 0 0
1986-12-27 0.001 Turkey Point Unit 3 0 0 Turbine trip occurred due to loss of governor oil pressure, reactor automatic trip failed, and manual trip had to be initiated. Consequently, a stuck open power operated relief valve incident occurred (in the classical fashion).
1980-06-11 0.001 St. Lucie Unit 1 0 0
1980-04-19 0.001 Davis-Besse 0 0 Two essential electrical buses were lost, the decay heat drop line valve was shut, and air was drawn into the suction of the decay heat removal pumps, leading to the loss of decay heat removal.
1979-06-03 0.001 Hatch Unit 2 0 0
1977-08-31 0.001 Cooper 0 0
1971-01-12 0.001 Point Beach Unit 1 0 0 Containment sump power operated relief valves were discovered to be leaky, and operators operated the valves to exercise and dislodge them. The valves opened and one remained open, even when operators commanded it to close, representing a stuck open power operated relief valve incident (in the classical fashion).

See also

References

  1. ^ Benjamin K. Sovacool. The Costs of Failing Infrastructure Energybiz, September/October 2008, pp. 32-33.
  2. ^ Benjamin K. Sovacool. The costs of failure: A preliminary assessment of major energy accidents, 1907–2007, Energy Policy 36 (2008), pp. 1802-1820.
  3. ^ a b c Benjamin K. Sovacool. The costs of failure: A preliminary assessment of major energy accidents, 1907–2007, Energy Policy 36 (2008), pp. 1802-1820.
  4. ^ Benjamin K. Sovacool. The Costs of Failing Infrastructure Energybiz, September/October 2008, pp. 32-33.
  5. ^ Newtan, Samuel Upton (2007). Nuclear War 1 and Other Major Nuclear Disasters of the 20th Century, AuthorHouse.
  6. ^ The Worst Nuclear Disasters
  7. ^ BBC NEWS | England | Road container 'leaked radiation'
  8. ^ Jungk, Robert. Brighter than a Thousand Suns. 1956. p.194
  9. ^ Benjamin K. Sovacool. The Costs of Failing Infrastructure Energybiz, September/October 2008, pp. 32-33.
  10. ^ Benjamin K. Sovacool. The costs of failure: A preliminary assessment of major energy accidents, 1907–2007, Energy Policy 36 (2008), pp. 1802-1820.
  11. ^ [1] accessed 13.5.2009
  12. ^ Nrc.gov accessed 13.5.2009

External links


Advertisements






Got something to say? Make a comment.
Your name
Your email address
Message