10.1.1 General

There are four aspects concerning the public acceptance of nuclear energy: severe accident risk, proliferation, vulnerability to sabotage, and nuclear waste disposal. Severe accident risk in nuclear power stations is treated in this chapter. The reason for this choice is that, until now, only reactor accidents have had serious consequences.

In 2013, approximately 16% of the electricity production worldwide and one third in the European Union was obtained from nuclear fission [1]. In 2010, 62% of the nuclear reactors in power stations worldwide were pressurized water reactors (PWRs) and 19% were boiling water reactors (BWRs) [2]. Both reactor types belong to the category of light water reactors (LWRs). These reactor types will be described in some detail in Section 10.2 as serious accidents occurred with these reactor types. Still, they are often considered to be options for the future. Nuclear power stations were introduced in the 1950s. Until now, three major accidents occurred in large nuclear power stations. First, in 1979, there was the TMI‐2 accident at Harrisburg, Pennsylvania, USA. It concerned a PWR. The consequences were modest. The accident is discussed in Section 10.3. Second, in 1986, an accident occurred at Chernobyl in Russia. The consequences of the accident at Chernobyl were very serious. The reactor type was different from either a PWR or a BWR, and the accident is not discussed in this book. Generally, that reactor type is not considered an option for the future as its safety characteristics are not optimum. Third, in 2011, there was an accident at Fukushima in Japan. It concerned three BWRs. The consequences were serious. What happened with the BWR of Unit 1 of the nuclear power station Fukushima Daiichi is described in Section 10.4.

Both at Harrisburg and Fukushima, primary process protection was by means of active and procedural safety measures. Two reactor types having passive safety as primary process protection are discussed in Section 10.5. It concerns the pebble bed reactor (PBR) and the prismatic block reactor. The reason for discussing these two reactor types and not, e.g. the molten salt reactor, is that large power stations have functioned with these two reactor types, albeit with mixed results. THTR‐300 in Germany, having a capacity of 300 MWe(lectrical), operated for 4 years with a PBR. Fort St. Vrain Unit No. 1 in the United States, having a capacity of 330 MWe, functioned for 10 years with a prismatic block reactor.

Finally, these two reactor types are compared in Section 10.6. It is recommended to further explore the possibilities of the prismatic block reactor.

10.1.2 Physics

Matter consists of atoms. A specific type of atom is characteristic for a chemical element. In many materials, atoms are combined to form molecules. The properties of a molecule are characteristic for that material. It has been for a long time considered that atoms cannot be split up and thus are the smallest particles of matter.

Mendeleev arranged the chemical elements in the Periodic Table of Elements. At present, 118 elements are incorporated into this Periodic System. Hydrogen is found in the upper‐left corner of that scheme. The hydrogen atom is the smallest atom. The famous Danish scientist Niels Bohr proposed a model for the hydrogen atom, which is basically still used today. It consists of a positively charged nucleus, called a proton, and a small, negatively charged, electron. The electron rotates around the nucleus, and the force with which these two particles attract each other is counterbalanced by the centripetal force (see Figure 10.1).

The atoms of deuterium and tritium are related to the hydrogen atom. The nucleus of a deuterium atom contains, in addition to a proton, an uncharged, neutral particle having the same mass as a proton. It is called a neutron. The nucleus of a tritium atom contains, in addition to a proton, two neutrons. All three atoms have one electron rotating around its nucleus. Hydrogen, deuterium, and tritium are called isotopes. Isotopes have the same chemical properties because their electronic configurations are equal. However, their atomic masses are different, and that fact provides possibilities to separate them physically. Other elements also have isotopes. In this context, it is useful to distinguish between light water and heavy water. Normal water is light water. The hydrogen atoms of light water have a nucleus that consists of one proton. One electron rotates around the nucleus. The deuterium atoms of heavy water have a nucleus that consists of one proton and one neutron. One electron rotates around the latter nucleus. In the molecules of both light water and heavy water, two hydrogen or two deuterium atoms are combined with one oxygen atom.

Relatively light atomic nuclei contain approximately equal numbers of protons and neutrons. For example, the nucleus of a nitrogen atom contains seven protons and seven neutrons. It is customary to provide the chemical symbol of an element with two numbers: the mass number and the atomic number. The mass number equals the sum of the number of protons and neutrons in an atomic nucleus, whereas the atomic number equals the number of protons, e.g. ¹⁴7N for nitrogen.

The number of electrons equals the number of protons in a neutral atom. On moving from relatively light atoms to heavier atoms, the number of neutrons in the nuclei of those atoms becomes larger than the number of protons. If, in an atomic nucleus, the number of neutrons exceeds the number of protons significantly, the distance between the protons increases and the nucleus loses stability. The element bismuth (Bi) has the heaviest atomic nucleus that is still stable. The number of protons in its nucleus is 83, whereas the number of neutrons in its nucleus is 126. Thus, the mass number is the sum of these two numbers, i.e. 209, and the atomic number is 83.

The majority of nuclear power stations use uranium (U) as fuel. In nuclear reactors, uranium is present in the oxide, UO₂, or in the carbide. Uranium occurs in an ore having the chemical formula U₃O₈, meaning a combination of three uranium atoms and eight oxygen atoms. The atomic nucleus of U contains 92 protons and its atomic number is 92. Naturally occurring uranium contains three isotopes of U and their atomic masses are 234, 235, and 238. It is customary to indicate them as U‐234, U‐235, and U‐238. Naturally occurring uranium contains a very small and negligible amount of U‐234. It consists of 99.3% by weight of U‐238 and 0.7% by weight of U‐235. The nuclei of both isotopes are unstable and exhibit radioactivity; they decay and emit radiation. Decay means that other nuclei are formed out of these uranium nuclei. The sum of the masses of the nuclei formed is less than the mass of the original nucleus. The difference is emitted as radiation energy according to the famous formula E = mc ². E is the energy in J, m is the mass difference in kg, and c is the velocity of light in m s⁻¹. The radiation energy can be converted into heat and that is what occurs in nuclear power stations.

U‐238 is much stabler than U‐235. Technically, it is possible to split U‐235 to produce energy. It is difficult to split U‐238. In order to be effective as fuel in nuclear power stations, the percentage by weight U‐235 of the sum of the masses of U‐235 and U‐238 must, for the majority of nuclear reactors, increase to at least 3–5. That increase is called enrichment and can be achieved by physical processes. The process used widely today is enrichment by means of ultracentrifuges. The internationally accepted maximum degree of enrichment in use today is 20% by weight. The reason is the necessity to avoid proliferation.

Thorium can also be used as a nuclear fuel. Thorium is a naturally occurring element having an atomic number 90 and a mass number 232, hence often indicated as Th‐232. It can be used as nuclear fuel in combination with U‐235. Worldwide, there is much more thorium available than uranium. In nuclear reactors, thorium is present as the oxide, ThO₂, or as the carbide.

10.2.1 Introduction

The common characteristics of these two reactors are dealt with in this section. They both belong to the category of LWRs. Light water is normal water, and the difference between light water and heavy water has been mentioned in Section 10.1.2. Both a PWR and a BWR is a vessel having a thick steel wall and containing water, either as liquid (PWR) or as liquid and vapor (BWR), and fuel in metal tubes; see Figure 10.2 that depicts a BWR. The most used fuel is uranium dioxide (UO₂). The physical form of the dioxide is a sintered pellet having a diameter of approximately 1 cm and a height of approximately 1 cm. The pellets are stacked one on top of the other in fuel tubes. Typically, the uranium in the oxide consists of 5% by weight of U‐235 and 95% by weight of U‐238. The metal tubes are close to each other to reach the critical mass and to enable nuclear fission. U‐235 decays, neutrons are emitted, and heat is produced. The neutrons emitted cause further fission and heat production.

Moderation of the neutrons is an important function of water in the reactor. Moderation is the deceleration of the emitted neutrons from approximately 14 000 km s⁻¹ to approximately 2.2 km s⁻¹ [3]. Only these relatively slowly moving neutrons can accomplish further fission of U‐235. Another key function of water is the absorption of heat to produce mechanical and subsequently electrical energy. Steam is raised in the reactor shown in Figure 10.2 and passed on to a turbine. The turbine rotates and drives an electricity generator. The condensed steam is returned to the nuclear reactor.

On moving the control rods in a vertical direction between the fuel tubes, the fission process can be controlled. The metal or alloy of the control rods absorbs neutrons. The nuclear fission is halted by moving the control rods fully between the fuel tubes. However, the decaying radioactive materials left from the fission continue to heat the reactor's coolant water. These decaying radioactive materials are called actinides. Immediately after the fission is halted, the heat in MW raised in the reactor is approximately 7% of the heat in MW raised before the fission was halted. The percentages are approximately 2, 0.5, and 0.1 after, respectively, an hour, a day, and a year [4].

When a nuclear reactor produces at, e.g. 100% capacity, 93% of the heat development is due to the fission of U‐235. The remaining 7% of the heat development is due to the decay of actinides. The heat development due to the fission of U‐235 can be stopped completely by the introduction of control rods. However, the heat development due to the decay of actinides cannot be stopped by the introduction of the control rods. Thus, the heat development due to the decay of actinides is not something that occurs when the control rods are introduced. This heat development is an integral part of the heat produced by a nuclear reactor when the reactor produces normally.

The reactor is placed in a building with concrete walls having a thickness of one to several meters to contain radiation. The reactor is called critical when it reaches a stationary state and the nominal heat flow is produced.

10.2.2 PWR

Figure 10.3 is a diagram of the nuclear power plant TMI‐2 at Harrisburg. An accident in this plant is discussed in Section 10.3. The power plant is no longer active. The water pumped through the reactor does not boil and exchanges heat indirectly in steam generators. The steam generated flows to turbines and brings about their rotation. Radioactive material thus cannot reach the turbines when a fuel tube is damaged. The turbines effect the rotation of electricity generators.

Typically, the pressure in the primary circuit is 155 bar, whereas the temperature of the water entering the reactor is 290 °C and the temperature of the water leaving the reactor is 325 °C. A pressurizer serves to prevent boiling of water in the primary circuit. Two thirds of its volume are filled with water and one third is vapor space. Water having a temperature of 345 °C has a saturated vapor pressure of 155 bar. The water temperature of 345 °C in the pressurizer is maintained by either heating water electrically or by adding cold water. So there is a margin of 20 K to prevent boiling of the water flowing to a steam generator. Typically, saturated steam having a pressure of 60 bar and a temperature of 275 °C is raised in the steam generators. It is passed on to the turbines, condensed, and recycled to the steam generators. The reactor, the steam generators, the pressurizer, and the circulation pumps form the primary circuit and are inside the containment. A typical PWR having a capacity of 1000 MWe(lectrical) is equipped with two or four steam generators. MWe refers to the actual output of electric power, the heat raised in a PWR is, as MWth(ermal), a factor of approximately 2.5 larger.

10.2.3 BWR

See Figure 10.4. The steam for the turbines is raised in the reactor itself. Thus, the design of a BWR is simpler than that of a PWR. Water circulation through the reactor is accomplished by, e.g. two pumps. The water boils in the reactor. There are steam bubbles in the upper part of the reactor. Moderation is hence more successful in the lower part of the reactor than in the upper part. That is an important reason why the control rods pass through the bottom of the reactor (see also Figure 10.2). Steam leaving the reactor entrains droplets and devices in the reactor's vapor space separate droplets from steam. Dry steam passes to the turbines and separated water is recycled to the reactor. Typically, the saturated steam has a pressure of 75 bar and a temperature of 290 °C. After having passed the turbines, the steam is condensed and water is recycled to the reactor. Radioactivity can reach the turbines when a fuel tube is damaged.

Event

An accident concerning the reactor of a nuclear power station occurred at Harrisburg on March 28, 1979. The nuclear power station was located on Three Mile Island, and the accident is often indicated by the acronym TMI. The site at Three Mile Island consisted of two nuclear power stations and the accident occurred in TMI‐2. The other nuclear power station is indicated as TMI‐1. Due to a loss of cooling capacity in the reactor of TMI‐2, the temperature in the reactor increased and caused the melting of a number of fuel rods. A chemical reaction between zirconium in the fuel rods and water generated hydrogen. Hydrogen entered the containment and caused an explosion. The containment was not breached. The major release of radioactivity on the morning of March 30, 1979, was caused by the controlled, planned venting of the make‐up tank into the vent header. The header was known to have a leak [5]! However, the release did neither harm the public nor the employees of the power stations [6]. TMI‐2 was severely damaged.

TMI‐2

The capacity of the two nuclear power stations together was 1700 MWe. The reactor of TMI‐2 was a PWR (see Figure 10.3), and the pressure in the primary circuit was 2150 psi (146.3 bar). Water having a temperature of 340 °C has a saturated vapor pressure of 2150 psi. The water in the pressurizer had that temperature. Probably, the water leaving and entering the reactor had temperatures of, respectively, 320 and 290 °C. The primary circuit consisting of the reactor, the pressurizer, two steam generators, and four reactor coolant pumps can be distinguished. UO₂‐pellets were present in thin tubes made of Zircaloy‐4 and having a length of 12 ft (3.66 m). The walls of the thin tubes are called “cladding”. Zircaloy‐4 is an alloy containing the metal zirconium. The primary circuit was located in the containment indicated as the reactor building. The reactor building had walls of reinforced concrete having a thickness of 4 ft (1.22 m). The control rods passed through the top of the reactor. The secondary circuit was mainly outside the reactor building. It consisted of a turbine, a condenser, pumps, and a cooling tower. Heat was exchanged indirectly in the steam generators.

Additional Remarks Concerning TMI‐2

At roughly 2200 °F (1204 °C), a reaction between the Zircaloy cladding and water could begin to damage the fuel rods and also generate hydrogen. Damage to the cladding releases some radioactive materials trapped inside the fuel rods into the core's cooling water.

At about 5200 °F (2871 °C), fuel starts to melt. Such melting could release far more radioactive materials than the damage done to the fuel rods at 2200 °F.

Description of the Accident

Reference 7 contains a description of the accident and Table 10.1 a timeline. TMI‐2 ran at 97% power prior to the accident. It started at 04.00 h by the tripping of the first pump of a series of feedwater system pumps supplying water to the steam generators. Within 1–2 s, all feedwater pumps tripped. These pumps were part of the secondary circuit. A “trip” means a piece of machinery stops operating. The cause of the tripping was probably the leaking of water into instrument lines. Such leaking had occurred at least twice earlier at TMI‐2. When the pumps stopped, the flow of water to the steam generators stopped. The plant's safety system automatically shut down the steam turbine and the electricity generator it powered.

When the pumps that normally supply water to the steam generators shut down, three emergency feedwater pumps automatically started. However, they could not deliver water to the steam generators as valves in the two emergency feedwater lines were closed. The stopping of the feedwater flow to the steam generators caused an increase in the temperature of the water in the pressurizer. The pressure in the vapor space of the pressurizer rose to 2255 psi (153.4 bar). Water having a temperature of 340 °C has a saturated vapor pressure of 146.3 bar. Water having a temperature of 343.5 °C has a saturated vapor pressure of 153.4 bar. Thus, a relatively small temperature increase leads to a substantial pressure rise. A relief valve atop the pressurizer (PORV) opened several seconds into the accident and steam and water began flowing out of the reactor coolant system to a relief tank. However, pressure continued to rise and 8 s after the tripping of the first pump the control rods automatically dropped down into the reactor core to stop the nuclear fission. The pressure continued to rise because, when the relief valve opened, the reactor was still running at 97% capacity.

The remaining heat development of the reactor after the control rods had dropped in was still 6% of the heat development prior to the start of the accident. With the reactor shut down and the relief valve open, pressure in the reactor coolant system fell. The pressure fell because the reactor was now, as the control rods had dropped into the reactor, at 6% capacity. Up to this point, the reactor system was responding normally to a turbine trip. The relief valve should have closed 13 s into the accident, when pressure dropped to 2205 psi (150 bar). It did not. A light on the control room panel indicated that the electric power that opened the relief valve had gone off, leading the operators to assume the valve had shut. But the relief valve was stuck open and would remain open for 2 h and 22 min, draining needed cooling water. In the first 100 min of the accident, some 32 000 gallons (121 m³), over one third of the entire capacity of the reactor's primary cooling system, would thus escape. Fourteen seconds into the accident, an operator in TMI‐2's control room noted that the emergency feedwater pumps were running. He did not notice two lights that could have told him that a valve was closed in each of the two emergency feedwater lines. Water could not reach the steam generators. One light was covered by a yellow maintenance tag. No one knows why the second light was overlooked.

Two minutes into the accident, pressure in the primary circuit dropped sharply. Automatically, two high‐pressure injection (HPI) pumps began pouring about 1000 gallons per minute (227.1 m³ h⁻¹) into the primary circuit. At the same time, because of the low pressure in the primary circuit, the water in this circuit started to boil. The level in the pressurizer rose and the operators took it that there was enough water in the primary circuit. However, the level was high because of the presence of steam bubbles in the water. About 4.5 min into the accident, an operator shut one HPI pump down and reduced the flow of the second HPI pump to less than 100 gallons per minute (22.7 m³ h⁻¹).

Slightly more than an hour into the accident, TMI's four reactor coolant pumps began to vibrate severely. The vibrations were caused by the pumping of both water and steam. The HPI water supply to the primary circuit was too small relative to the water removal as steam via the relief valve due to the heat development of the fuel. Even so, after 1 h and 14 min, it was decided to close down two of the four reactor coolant pumps. After 1 h and 31 min, the two remaining pumps were shut off.

As from the latter point in time, the fuel rods were no longer submerged in water and the temperature of the fuel rods increased. A reaction between water and zirconium occurred and hydrogen was generated. Hydrogen escaped through the open relief valve into the reactor building. It mixed with air containing oxygen and a vapor explosion occurred 9 h and 50 min after the start of the accident [8]. The pressure increase due to this vapor explosion was 28 psi (1.9 bar). The containment was not breached by the explosion.

Slightly more than 2 h into the accident, it was established that the relief valve was open and after 2 h and 22 min an adjacent valve was closed to stop the leaking of water and steam into the reactor building. Still 1 h later, high‐pressure water injection was resumed. By the evening of March 28 (the day of the accident), the core appeared to be adequately cooled and the reactor appeared to be stable.

The major release of radioactivity on the morning of March 30, 1979, was mentioned in the beginning of this section. Remember that the accident started at 04.00 h on March 28, 1979.

Remarks

The heart of the matter of the accident at Three Mile Island is that it has not been possible to transfer the remaining heat of the nuclear reactor in a safe way.

The protection of the nuclear reactor at Three Mile Island relied on passive, active, and procedural safety measures (see Chapter 2). A passive safety measure is, within very wide limits, not endangered by human errors or equipment failure. The building provided passive protection by preventing the emission of radioactivity. Active process protection starts working upon a signal. Procedural safety measures concern action to be taken by humans.

The first active safety measure to be discussed was the automatic shutdown of the turbine and the generator it powered. This measure functioned well.

The second active safety measure to be mentioned was the automatic starting of three emergency feedwater pumps. This action was started by the tripping of the feedwater pumps. The back‐up pumps could not deliver water to the steam generators because valves were closed in their delivery lines. This protection method failed. Fourteen seconds into the accident, an operator noted that the three emergency feedwater pumps were running. However, it was not noted that valves in the delivery lines were closed.

The third active protection method was the opening of a relief valve atop the pressurizer. The relief valve was opened by a high‐pressure signal from the pressurizer when less than 8 s had elapsed after the start of the accident. This measure functioned well. However, the relief valve should have closed when 13 s into the accident had elapsed because the pressure in the pressurizer had come down. The relief valve did not close. The third safeguarding method failed as well.

The fourth active safeguarding method to be mentioned is the automatic dropping of the control rods into the core to halt nuclear fission. This action was started because the pressure in the reactor continued to rise after the relief valve had opened. This protection method functioned satisfactorily. Indeed, the control rods fell into the core 8 s into the accident. However, this action reduced the heat flow from the reactor to 6% of the nominal value and not to 0%.

The fifth active protection method was the automatic starting of two HPI pumps 2 min into the accident. The signal starting this action is low pressure in the primary circuit. This measure functioned well. However, the action of the HPI pumps was mitigated by human action. The final result was the failing of this protection method as well.

None of the primary process protection measures was a passive safeguarding method.

Damage

Personal damage did not occur. The small radioactive releases had no detectable health effects on plant workers or the public [6]. The nuclear power station TMI‐2 was severely damaged.

Event

A nuclear accident occurred at Fukushima in Japan on March 11, 2011. It concerned the nuclear power station Fukushima Daiichi. The word “Daiichi” means “Two” in Japanese. There were two nuclear power stations at Fukushima and the accident occurred at the second station. Both stations were operated by TEPCO (Tokyo Electric Power Corporation). The Fukushima Daiichi station consisted of six nuclear power plants, each indicated by the word “Unit” followed by a number. At the time of the accident, Units 4–6 were not operating due to periodical inspection.

Following a major earthquake, a 15‐m tsunami disabled the power supply and cooling of the reactors of Fukushima Daiichi Units 1–3. The cores of these three units largely melted in the first 3 days. Unit 1 will be in focus mainly in the next sections.

Unit 1

Unit 1 was completed and started to operate in March 1971 (see Figure 10.4). It was the third BWR built in Japan, and most of the design and manufacture of the key components were done by General Electric (GE). Its electrical output was approximately 460 MWe. The operational conditions resembled the conditions outlined in Section 10.2 for a BWR. The steam passed on to the turbines was raised directly in the reactor. The reactor was equipped with two circulation pumps, indicated as Pumps A and B. The vessel shaped like a lightbulb was the primary containment vessel (PCV). The containment of Unit 1 is indicated by the manufacturer as MARK I. The suppression chamber (SC) of the PCV contains cold water and serves to condense steam that would have escaped from the reactor.

Additional Notes Concerning Fukushima Daiichi Unit 1

The melting point of the core fuel, i.e. UO₂, is 2880 °C [9]. The melting point of a mixture of UO₂, ZrO₂, and Zircaloy is 2000–2200 °C. The melting point of stainless steel is approximately 1500 °C.

Detailed Description of the Event

Reference 10 contains a description of the accident and Table 10.2 a timeline. An earthquake hit the power station Fukushima Daiichi at 14.46 h on March 11. The output of the power station was approximately 460 MWe at this point in time. The earthquake caused a loss of electric power and diesel generators started automatically to supply electric power. Also automatically, the main steam isolation valve was closed and the control rods passed into the reactor to stop nuclear fission. The reactor's heat output was thereby reduced to 7% of the heat output at the time the earthquake occurred. As the main steam isolation valve was closed, the water in the reactor was heated and the pressure rose. Six minutes into the accident, two isolation condensers (ICs) were activated by a signal caused by a high pressure in the reactor (se Figure 10.5). Unit 1 then switched to cooling by ICs A and B. Steam raised by the remaining heat flowed to coils in the ICs. It condensed in the coils and water flowed back into the reactor. ICs could continue to cool the reactor for 8 h when the reactor stopped. The cooling period could be extended by replenishing water in the ICs.

Seventeen minutes into the accident, operators of Unit 1 stopped IC‐cooling because the cooling rate was too high. The instructions were that the cooling rate of the reactor should not exceed 55 K h⁻¹. About 27 min into the accident, when the temperature and pressure began to return to normal, operators switched one IC on again. By switching the active IC on and off by using valve MO‐3A, they controlled the reactor pressure manually.

Fifty‐three minutes into the accident, Unit 1 was hit by the second wave of a tsunami, causing the loss of electric emergency power. Out of 13, 12 diesel generators stopped supplying electric power to the power plant. As a result, Unit 1 could no longer receive electric power. Just before that happened, operators had closed valve MO‐3A. Because electric power had been lost, it was not possible to open that valve again. Thus, the ICs had become inoperable. When the IC cooling stopped, the reactor's remaining heat flow had decreased to 2% of the heat flow before the accident.

The reactor pressure rose after the tsunami had hit the plant because IC‐cooling was no longer active. When the pressure rose approximately 10%, a relief valve atop the reactor opened automatically and steam was vented into the PCV. The relief valve kept the reactor pressure at approximately 70 bar, the saturated vapor pressure of water having a temperature of 284.5 °C. Thus, as long as there was water in the reactor, the reactor temperature could not exceed this temperature. Steam was condensed in the SC and the temperature of water therein rose (see Figure 10.4).

Reactor cooling was maintained by automatic opening and closing of the relief valve to release generated steam. It has been recorded that the radiation level in front of the double door of the reactor building was high at approximately 23.00 h on March 11 (8 h and 14 min into the accident). That is evidence that a substantial amount of radioactive material had leaked into the PCV. It has been estimated by TEPCO that by midnight the reactor's core had lost all water and that a chemical reaction between zirconium and water had occurred. Zirconium was present in the fuel tubes made of Zircaloy. The chemical reaction generated hydrogen and heat.

It has also been recorded that the pressure inside the PCV was 6 bara at approximately 23.50 h (9 h and 4 min into the accident). That means that an opening had been formed in the reactor and that steam had leaked into the PCV. The temperature of the water in the SC had risen to approximately 160 °C.

Molten core material fell out of the reactor on the floor of the PCV between 0.00 and 04.00 h on March 12. It can be assumed that the temperature of the molten core material was 2000 °C. The molten material sank about 65 cm into the reinforced concrete. It cooled down thereby. The thickness of the reinforced concrete was 2.6 m [11].

Pumping of seawater into the PCV by means of a fire engine started at 04.00 h (13 h and 14 min into the accident). The seawater ran down the reactor's circumference. The pressure inside the PCV was, at the time the seawater injection started, approximately 8 bara. The temperature of the water in the SC had risen to approximately 170 °C. The seawater flow amounted to, at the discharge pressure of approximately 8 bara, 5 t h⁻¹. Both the pressure inside the PCV and the seawater flow stayed approximately constant for a period of 10.5 h.

It can be calculated that the decay heat caused the complete evaporation of the water in the seawater flow. The calculation proceeds as follows. The capacity figures of Unit 1 are 460 MWe and 1500 MWth. All figures that are given in this paragraph are approximate figures. The seawater injection by the fire engine at a rate of 5 t h⁻¹ started at 04.00 h on March 12. The decay heat was 0.7% of 1500 MWth at that point in time as more than 13 h had elapsed since the reactor had been stopped. And 0.7% of 1500 MWth is 10.5 MW. On taking 3000 kJ for the heating and evaporation of 1 kg of water in seawater, the latter heat flow is able to heat and evaporate 12.6 t of water in seawater per hour. The seawater injection of 5 t h⁻¹ did not even come close to matching the decay heat. That means that the temperature of the core material kept rising. The rapid evaporation of the water in the seawater flow also explains that the pressure in the PCV stayed more or less constant at a level of 7–8 bara. The contact between the molten core material on the floor of the PCV and gaseous water was not intimate. Only a small amount of hydrogen could additionally be formed.

The pressure inside the PCV was relieved to about 5 bara by venting to the atmosphere at 14.30 h on March 12 (23 h and 44 min into the accident). This caused the seawater flow to rise to 30 t h⁻¹.

The decay heat could now no longer evaporate all water in the seawater flow. The intimate contact between molten core material on the floor of the PCV and seawater caused the formation of a substantial amount of hydrogen gas and a pressure rise. The high pressure in the PCV lifted the top of the PCV. Next, the shield plug was lifted and a gaseous mixture flowed from the PCV into the space of the fuel exchange floor. Here, it mixed with air containing oxygen. An explosion occurred at 15.36 h on March 12 (24 h and 50 min into the accident). The explosion blew off the roof and cladding on the top part of the building. The explosion has probably been ignited by a spark caused by the falling back of the shield plug. There was an open connection between the PCV and the atmosphere after the escape of the gaseous mixture from the PCV.

Context and Aftermath

The cores of the reactors of Fukushima Daiichi Units 1–3 largely melted in the first three days [11]. The molten material of the reactors of Units 2 and 3 stayed in the reactors. There were hydrogen/oxygen explosions in Units 1, 3, and 4 in the first five days. Unit 4 was not operating at the time of the accident. The explosion in Unit 4 was caused by the leakage of hydrogen gas from Unit 3. Major releases of radionuclides, including long‐lived cesium, occurred to air, mainly in mid‐March 2011. The main source of radioactive releases was a gas release from Unit 2 on March 15. An explosion did not occur in Unit 2 as it was severely damaged by the explosion in Unit 1 and gases could flow freely to the atmosphere. The electric power supply to Fukushima Daiichi was restored after 10 days. The reactors of Units 1–3 were stable with water addition after 2 weeks and by July 2011 they were being cooled with recycled water from a new treatment plant. Official “cold shutdown condition” was announced in mid‐December 2011.

Damage

Three TEPCO employees were killed when the earthquake and the tsunami hit the power station [11]. There have been no direct deaths of radiation sickness from the nuclear accident at Fukushima. Approximately 160 000 people were evacuated from their homes. Only in 2012 limited return was allowed and in October 2013, 81 000 evacuees remained displaced due to government concern about radiological effects from the accident.

The material damage of the accident was substantial.

Remarks

The heart of the matter of the events in Unit 1 of the nuclear power plant Fukushima Daiichi is that, like at Three Mile Island (see Section 10.3), it has not been possible to transfer the remaining heat of the nuclear reactor safely.

The protection of the nuclear reactor of Unit 1 relied on active and procedural safety measures (see Chapter 2). Active process protection starts working upon a signal. Procedural safety measures concern action taken by humans.

The starting up of the diesel generators to supply electric power when the earthquake occurred was the first active safety measure. That measure was successful.

The second measure was the closing of the main steam isolation valve when the earthquake occurred. The measure was successful. Steam raised in the reactor could no longer flow to the turbines.

The introduction of the nuclear control rods into the reactor to stop nuclear fission was the third measure. It was successful as well.

The fourth protection measure was the starting up of two ICs to cool the reactor. That safeguarding method was activated by a signal indicating too high a pressure in the reactor. Although the safety measure was successfully started up, its effect was mitigated by the operators of Unit 1 and the loss of electric power.

The fifth safety measure was procedural and comprised the pumping of seawater into the PCV. The measure was taken late and the seawater flow was small. The temperature of the molten core material kept rising.

The sixth protection measure was also procedural and concerned the venting of the PCV. The measure had an adverse side‐effect because the increased seawater flow caused the generation of a substantial amount of hydrogen gas and an explosion.

The primary containment was a passive safety feature. However, it failed due to the high pressure built up by the formation of hydrogen after the venting of the PCV.

10.5.1 Introduction

High‐temperature gas‐cooled reactors (HTGRs) are discussed in this section because they have better safety characteristics than LWRs and relatively large power stations equipped with HTGRs have been built and operated.. See Figure 10.6 in which two reactor types are depicted.

The functioning of these two reactor types will be described briefly. The reactor at the left in Figure 10.6 contains fuel in prismatically shaped elements. It is called prismatic block reactor. The prismatic blocks are made of graphite and are stacked one on top of the other. Graphite is moderator in this reactor. There are vertical channels in the blocks that are filled with small fuel cylinders, also called compacts. The small cylinders contain fuel kernels containing UO₂, uranium carbide, or a mixture of thorium carbide and uranium carbide. The kernels are coated with pyrolytic carbon so as to retain fission products. The diameter of the coated kernels is, e.g. 0.95 mm. Helium gas under pressure passes through channels in the fuel elements and is heated. Hot helium gas can then transfer its heat indirectly in a heat exchanger to generate steam. Helium gas is subsequently recycled to the reactor. Generated steam drives a turbine and the turbine drives an electricity generator.

The reactor at the right in Figure 10.6 is known as pebble bed reactor (PBR). The reactor is filled with pebbles having a diameter of, e.g. 60 mm. Like the small cylinders of the prismatic block reactor, the pebbles contain coated fuel kernels. The fuel kernels can have the same size and composition as described for the prismatic block reactor. A 5‐mm outer shell of the pebbles consists of graphite. Graphite is also moderator in this reactor. Helium gas under pressure passes through the bed of pebbles and is heated. Hot helium gas can be used in the same way as described for the prismatic block reactor. Pebbles are continuously fed at the reactor top and extracted continuously at the reactor bottom.

The moderator function and the heat carrier function are separated in these two reactor types. Remember that water is both moderator and heat carrier in an LWR. A further aspect is that helium does not interact with neutrons.

Typically, in a PWR, the pressure in the primary circuit is 155 bar and water leaves the reactor at 325 °C. In a PWR, the pressure in the primary circuit must always exceed the saturated vapor pressure at the temperature at which water leaves the reactor. Thus, in order to avoid extremely high pressures, it is not feasible to raise the temperature of the water in the primary circuit of a PWR substantially. In the primary circuits of HTGRs, temperatures higher than 325 °C, e.g. 800 °C, are possible.

The temperature and the pressure in the primary circuit of an HTGR can be chosen independently from each other. That is not possible in an LWR. However, helium pressure in the primary circuit of an HTGR will usually not be chosen very low to obtain the right conditions for relatively good heat transport and transfer characteristics.

Both reactor types are often claimed to be inherently safe. The development of HTGRs started in the 1950s and 1960s. The nuclear accidents at Three Mile Island (1979), Chernobyl (1986), and Fukushima (2011) had not yet happened then. The main emphasis was, at that time, on the possibility to reach high temperatures in the primary circuit. However, it was also realized that the safety properties of HTGRs were relatively good.

Back in the 1950s and 1960s, there were three reasons to consider the design of HTGRs. The first reason was that the efficiency of nuclear power stations can increase when the temperature at which helium leaves the core rises. The second reason was that it is possible to use helium having a temperature of, e.g. 800 °C, to heat certain reactors in the chemical industry indirectly without conversion to electrical power. For instance, it can heat chemical reactors in which methanol is synthesized from carbon monoxide and hydrogen. The third reason was that HTGRs have better safety characteristics than LWRs.

Concerning the temperatures at which helium leaves and enters the core, two aspects play a role. The first aspect is the Wigner effect [12]. That effect describes the phenomenon that energy is stored in irradiated graphite and can be released suddenly when temperature rises. Such a release of energy leads to a sudden additional temperature rise. The Wigner effect is important at relatively low temperatures and is less important at relatively high temperatures. To avoid the effect, the temperature at which helium enters the core has to be at least approximately 200 °C and preferably at least 300 °C.

The second aspect concerns the investment. To keep the sizes of the steam generators reasonable, the temperature difference across the metal tube walls of the steam generators between helium and water should be substantial. The relatively great difference compensates for the poor coefficient for the heat transfer from helium to the metal wall of the steam generator.

The combination of the two effects leads to, e.g. a temperature of 700 °C at which helium enters the steam generator. Helium would leave the steam generator at, e.g. 300 °C and the temperature difference between the two helium flows would in that case be 400 K.

A further aspect is that fuel is not, like in LWRs, contained in metal tubes. Helium coolant comes into contact with graphite only. A chemical reaction between helium and graphite cannot occur as helium is inert. In LWRs, water can react with zirconium of the fuel tubes at elevated temperatures and hydrogen gas can be formed.

When graphite at, e.g. 700 °C, comes into contact with air, it will be oxidized. Concerning process safety, this is an aspect to be considered. See the paragraph on PBR process safety in Section 10.5.3.

10.5.2 Safety Aspects of HTGRs

One aspect strikes immediately on comparing the safety aspects of HTGRs and LWRs. A typical power density of a PWR is 230 MW m⁻³ [13]. A typical power density of an HTGR is approximately two orders of magnitude smaller. In other words, HTGRs are, for a given capacity, relatively large and LWRs are relatively small. The core diameter of Peach Bottom Unit No. 1, an HTGR having a capacity of 40 MWe, was 2.79 m (see Table 10.5). A typical PWR having a capacity of 1000 MWe has a diameter of 4.5 m [14]. The difference in power density is caused by the different natures of these two reactor types. The heart of the matter is that light water (normal water) is more effective in capturing neutrons than graphite. The factor between these two moderators is 65 [15]. An LWR cannot function if the fuel rods are placed farther apart from each other, leading to a lower power density. The consequences of the difference in power density will be considered shortly. First, a few words concerning the PWR. If an LOCA (loss of coolant accident) occurs, it can be expected that the shutdown rods are brought in place to stop the heat development due to the fission reactions. However, one still has to deal with decay heat; see Section 10.2.1 for a description of decay heat. That heat can be removed by means of emergency cooling systems. If such systems fail, the situation cannot be kept under control.

Next, a few words concerning an HTGR. If an LOCA occurs, it can be expected that the shutdown rods are brought in place to stop the heat development due to the fission reactions. However, also in this case, one has to still deal with decay heat. The core temperature will in such a case increase to a certain value and then start to fall. Core decay heat can be carried away by conduction, natural convection, and radiation. The criterion is to keep the temperature of the fuel below 1600 °C to avoid liberation of fission products. The reasons of this good safety characteristic are, compared to the LWR, the much larger heat capacity of the reactor per MWth and the much larger outer area per MWth. Both a PBR and a prismatic block reactor should not exceed a certain size to maintain this good safety characteristic. Those sizes will be mentioned when the PBR and the prismatic block reactor are discussed.

There is one further major aspect. If, in the case of a HTGR, the shutdown rods cannot be brought in place when an LOCA occurs, the fission heat can also be kept under control. The temperature of the core then rises and the fission reactions practically stop immediately due to the Doppler effect. At relatively low temperatures, neutrons are effective concerning fission. However, they lose effectivity when the temperature rises because more neutrons are then captured by U‐238 or by Th‐232. Thus, the Doppler effect results in a negative temperature coefficient, i.e. the core's reactivity decreases when the temperature increases.

Summing up, HTGRs can be designed such that the fuel temperature remains below 1600 °C during serious accident conditions, i.e. helium pipe rupture, simultaneous loss of electrical power supply, and simultaneous failure of the emergency shutdown.

10.5.3 PBR

10.5.4 Prismatic Block Reactor

10.5.5 Comparison Between PBR and Prismatic Block Reactor