Posted by James Wesley Rawles on September 6, 2010
It is with some trepidation that I write this article, since what I write will be controversial and will alarm some members of the public as well as your readers. Some of my colleagues have urged me not to bring this subject into the open or to even discuss it in public. However, I think the topic is important and needs to be brought to the attention of the public. The issue is the effect that an electromagnetic pulse (EMP) attack, or for that matter, even a great geomagnetic storm created by a coronal mass ejection (CME) from the sun, would have on a nuclear power plant. Personally, I believe that an EMP attack from a rogue nation such as Iran or North Korea or even a terrorist organization is perhaps the most serious security threat that we face as a nation today.
As many readers may know, nuclear power provides about 20% of the electricity generated in the United States. It is an important component of our energy mix, reduces our dependence on foreign oil, and although some people are concerned about what to eventually do with nuclear waste, nuclear power plants emit no greenhouse gases and are generally quite benign. [I would rather live next to a nuclear power plant than say a chemical plant. How may people recall the incident in Bhopal, India? Over 6,000 people died or were maimed in that tragedy. No member of the public has ever been killed (or even injured) by a commercial nuclear power plant in this country.]
Many readers (if they are old enough) will remember both the Three Mile Island incident (where a Babcock and Wilcox reactor actually partially melted its core) as well as the Chernobyl accident, where an explosion damaged the core of a Soviet-era RMBK graphite-moderated reactor and spread radioactive fission products over a large portion of Europe. We in the nuclear power industry have been saying for years that a Chernobyl-type accident could never happen in the USA. All of the commercial reactors in the USA have concrete and steel containment structures that would prevent (or at least greatly reduce) any release of radioactive fission products to the public. The reactors at Chernobyl had no such containment structure, and the explosion literally blew the roof off of the reactor building.
As a nuclear engineer who has worked in the industry for nearly 30 years, I have agreed with this premise – that all of the US commercial reactors are very safe. Under normal circumstances, I still believe this. However, as I have been studying the effects of EMP for the last several years, my concerns have grown.
I have recently been in contact with a member of the intelligence community who is highly knowledgeable in the area of EMP. I have communicated my fears regarding the effects that an EMP attack might have on nuclear power plants, and this person has confirmed (through independent sources) that my concerns are well founded. I have also gotten concurrence from eight other engineers of various disciplines at my power plant (such as transient analyses, simulator, reactor engineering, a Shift Technical Adviser and nuclear analyses) that the scenario that I describe here is accurate.
Nuclear power plants are not isolated electrically. They are tied into the power grid and are also dependent upon it. There is a postulated accident for nuclear power stations called “Station Blackout,” where all off-site power is lost. Every nuclear power plant must prove to the NRC that they have the ability to withstand this event without core damage. Every US nuclear power plant has emergency diesel generators just for this purpose. These are designed to start automatically in the event of the loss of off-site power. This kind of event has actually happened before in the USA, and the systems responded as designed, and off-site power was restored within a reasonable period of time.
However, in the event of an EMP attack, the grid will come down, and it may not come up for many months, if not years. It is likely that a substantial number of transformers that are used to link power plants (and this applies to all power plants – coal, gas, oil and nuclear) to the grid will be “fried.” There will be no way to obtain off-site power to restart the nuclear power plants. Most station blackout events are assumed to be concluded (i.e., “over”) within 24 hours. No one that I know of has seriously analyzed the effects of prolonged station blackouts.
Assuming that the emergency diesel generators will start after an EMP event (and this is up for debate), most power plants only have enough diesel fuel on site to keep them running for about one week (though some may have up to 30 days of fuel). If they don't start, or if the controls systems do not operate, then everything that I describe here will still come to pass, only much more rapidly. The power from the diesel generators is needed to operate the pumps that circulate the water in the reactor (called the “primary side”) and that also feed the steam generators with water (part of the “secondary side”). If power to the reactor coolant pumps in the primary side is lost, the reactor will likely begin what is known as “natural circulation.” However, in order to remove heat from the reactor core, water still needs to be continuously pumped through the steam generators so that the heated water in the secondary side can be cooled either via cooling towers, spray ponds or some other ultimate heat sink. If these secondary side (feed water) pumps will not operate, then the steam generators will dry out and then the cooling effect for the core is lost. (A steam generator is just a very large heat exchanger. Think of the steam generator as the “radiator” in your car. If your water pump goes out, water will not be able to flow through the radiator, and your car will overheat.) The result is that the reactor core will heat up, pressure will build to the point that the reactor coolant system (RCS) will not be able to withstand the pressure. Special spring-loaded valves will automatically lift and vent steam to the containment building to reduce the pressure in the primary system. Loss of pressure control will occur eventually, the coolant inventory in the RCS will drop to the point that the core becomes uncovered. Charging pumps normally would pump additional water into the primary system, but without power, these will not be available. Essentially, this event is similar to what is known as a Loss of Cooling Accident (LOCA). Again, all power plants are designed to “survive” this type of accident with minimal fuel damage. However, that assumption is based on having power available to operate the safety systems, including the High Pressure and Low Pressure Safety Injection (HPSI and LPSI) pumps to pump additional water into the primary system. There are other emergency systems, such as Safety Injection Tanks (SIT), which are passive and will inject water into the core when the pressure is reduced enough such that the SIT tank pressure is greater than the RCS pressure and then the check valves will open automatically. [It should be pointed out here that there are also steam-driven auxiliary pumps that will still function for a while to run the auxiliary feed water system to feed additional water into the steam generators (until there is no water left in the secondary system to turn into steam).]
The HPSI and LPSI pumps are designed to ensure that the core remains covered (as much as possible) by injecting water into the core so that the core can still be cooled. If these pumps are not working due to lack of electrical power, then no additional water is being injected into the core. When the water level in the reactor drops below the top level of the fuel, the core will begin to melt. This is what happened at Three Mile Island. However, the containment structure prevented large releases of radioactive fission products to the public.
You might ask, “well, if the containment structure can contain the melted reactor core, is there a real danger to the public?” The answer is, “yes,” but not from where you think. The reactor core may well be the focus of most people, but the real concern is somewhere else.
What many people don't know about nuclear power plants is that when spent fuel is off-loaded from the reactor core, the fuel is then placed into what is essentially a large, very deep swimming pool called the “spent fuel pool.” Fuel that has been removed from an operating reactor core is still very hot (both in the sense of temperature and radiation level). In fact, if you were to stand within even 50 feet of a spent fuel assembly with no shielding, you would receive a lethal dose of radiation in just seconds. The water in the spent fuel pool, in addition to cooling the fuel assemblies, acts as a biological shield. In fact, water is an excellent shielding material. You can stand at the top of the spent fuel pool in virtually any nuclear power plant in the US and receive virtually no dose of radiation, so long as the fuel assemblies are covered by about 25 feet of water.
The building that houses the spent fuel pools at nuclear power plants in this country is usually a simple building, with concrete sides and floors but usually with nothing but a thin, corrugated steel roof. This is the root of the problem. Just like the fuel in the reactor, the fuel assemblies in the spent fuel in pool must also be cooled. These pools have their own independent, multiply redundant systems for cooling, separate from the systems that cool the reactor core. However, these pool cooling systems can be cross-tied with the reactor cooling systems in an emergency. The water in the spent fuel pool must be continuously circulated through heat exchangers (again, like your car radiator) to reject heat. Loss of off-site power will also cause a loss of spent fuel cooling. Normally, the temperature in these spent fuel pools is somewhere around 100 to 110 degrees F or so (similar to a typical suburban “hot tub”). When the spent fuel cooling system pumps stop operating, the fuel assemblies in the spent fuel pool will immediately begin to heat up. These fuel assemblies will continue to heat the water in the spent fuel pool until it boils. The best case scenario of “time to boil” for these spent fuel pools is perhaps 90 hours. The worst case, such as just after a core offload, would be much shorter, perhaps as little as four hours or even less. At that point, once the fuel assemblies in the spent fuel pool become uncovered because the water has boiled off, the effects mirror what would happen in the reactor core. The spent fuel assemblies will heat up until the fuel cladding starts to melt. As bits of the melting fuel fall into what is left of the water in the pool, the process will just accelerate as the heat source is now more concentrated since it has fallen back into the water and the water may flash to steam and this may cause the pressure in the building to increase, and radioactive steam, carrying radioactive particles, will now begin to exit the building through the non-sealed penetrations, portals or doors in the building.
Of course, there are usually multiple sources of water than can be called upon to re-fill the spent fuel pool before the water all boils off. But virtually all of these systems are dependent upon working, electrically operated pumps to move this water. If control systems have failed due to the EMP and there is no power to operate the pumps (either to add additional water or to pump water through the heat exchangers), then the fuel will ultimately become uncovered. Exposing the hot zirconium fuel cladding to air and steam causes an exothermic reaction, and the cladding will actually catch fire at about 1,000 degrees C. Even the NRC concedes that this type of fire cannot be extinguished, and could rage for days (Source: Bulletin of the Atomic Scientists, Vol. 58, No. 1, Jan./Feb. 2002).
The bottom-line is that if the spent fuel cooling pumps cannot be operated or the system cannot be cross-tied with the reactor shutdown cooling system, then the fuel assemblies in the spent fuel pool will melt, catch fire, and radioactive fission products will be released into the atmosphere and much of the countryside downwind of the nuclear power plant will be contaminated for many years. Thus, an EMP attack has the potential to cause a Chernobyl type accident at every nuclear power plant in the country!
There are a lot of “ifs” to this scenario. IF there is an EMP attack or solar event. IF the emergency diesel generators will function (or not) and IF the spent fuel pooling system can get power from the diesels or be cross-tied to the shutdown cooling system. Perhaps the emergency diesel generators will still function, but what happens when they run out of fuel? In the event of an EMP attack, can tanker trucks with diesel fuel get to all of the nuclear power plants in the US in time to re-fuel them before they stop running? Will tanker trucks even be running themselves?
I think it also bears noting that the volume of fuel in the spent fuel pools is many times greater than that in the reactor cores. Most nuclear power plants have 10 to 20 years or more of spent fuel stored in their spent fuel pools. Therefore, the consequences of a spent fuel pool melting down and subsequently spewing radioactive fission products into the air is potentially worse than if just the reactor core were to melt and its fission products releases into the air. Assuming all of the spent fuel in the pool melts, catches fire and the radioactive isotopes are released into the atmosphere, lethal dose rates may be accumulated even 5 to 10 miles from the plant site (>500 REM), with dose approaching 50 REM even out as far as 50 miles. Since Cesium-137 would be the largest released isotope in terms of curies (which the body preferentially uptakes over potassium), it will be about 300 years before the area might be habitable again. This is because Cesium-137 has a half-life of about 30 years, and the “rule of thumb” is that you need to wait ten half-lives before the isotope has decayed away to a negligible level. (Results for dose were calculated for a typical pressurized water reactor (PWR) spent fuel pool using the RASCAL radiation dose code from Oak Ridge National Laboratory assuming 100% release over two days, winter conditions, calm winds at 4 mph.)
I urge anyone living within 50 miles downwind of a nuclear power plant to be prepared to bug out in the event of an EMP attack. You will likely have a few days to pack and leave, but no more than a few. If the reactor near you has just refueled, and the emergency diesels do not start, you may have less than one day (since the heat load in the spent fuel pool immediately after a refueling is much greater than normal, and boiling will occur much faster). Many people have already expressed here the importance of having a G.O.O.D. bag and a plan to leave their current location if required. However, many people may need to evacuate on foot or by bicycle if the EMP attack renders their vehicles useless. I think this puts added emphasis on having a G.O.O.D. vehicle that is not reliant on computers or complex electronics.
For those of you who commute long distances to work I would also suggest that you have and maintain a G.O.O.D. mini-bag. (Nutnfancy on YouTube has produced an excellent series of videos on this – he has called it an “Urban Survival Kit” or “USK”). If your primary commute vehicle fails due to an EMP (or if your train or bus fails to function) while you are at work, then you may have a long walk home. It is wise to have pre-positioned (if you are able), a bag or backpack which contains items that may help you to get home more comfortably and safely.
I will cover what is in my mini-bag that I have pre-staged in the event that an EMP happens while I am at work at my power plant. (I would need to walk more than 30 miles to get home) in another letter. But I certainly hope that I never have to use it! - B.Z.
JWR Adds: At a minimum, in addition for G.O.O.D. and "get me home" kits, I recommend stocking up on potassium iodate pills, for thyroid protection, in the event of a nuclear accident. These are available from several SurvivalBlog advertisers. In some locales, they are made available free of charge to down-wind residents.