In contrast to gaseous coolants, liquid coolants may undergo a change of phase (i. e., into vapor) if their temperature rises high enough. However, they have a much higher heat capacity and their better heat transfer characteristics (dis­cussed in Section 3.3) allow them to be operated at much higher heat fluxes than gases. A variety of liquids have been used in reactor cooling, but only water (light and heavy), organic fluids, molten salts, and liquid metals will be considered here.

3.1.2 Light Water

From the earliest days of the development of nuclear energy, reactor cooling by ordinary water (i. e., light water) has been the most commonly adopted practice.

Y/o t#ar /’on Кл 1 1СОГІ OC noolont лп И’Л

reactor (P^WR; see Chapter 2) or in combination with a separate moderator such as graphite or heavy water. An example of the former combination is the Russ­ian boiling-water, graphite-moderated direct-cycle reactor (RBMK). Here, the fuel channels cooled with light water pass through pressure tubes set within the graphite core, which acts as the moderator.

Although light water is readily available, a number of problems are associ­ated with its use:

1. It has a relatively low boiling point (100°C), and therefore the reactor must be operated at high pressure to maintain the water in the liquid state at tempera­tures suitable for power generation cycles. Thus, in the P^^ the light water is pressurized up to 15.5 megapascals (155 bars, 2300 psia), at which its satura­tion temperature is 345°C. The average outlet temperature for the water from a P^^ is around 320°C, although the temperature may exceed the saturation temperature (giving some local boiling) in some of the channels.

2. In the presence of a neutron flux, water decomposes slightly into its con­stituent elements (hydrogen and oxygen). This radiation-induced reaction can be suppressed by having an excess of hydrogen dissolved in the water, which is the system adopted in the P^^.

3. Water is actually quite a corrosive substance, reacting with the materials of the fuel elements and reactor circuit and picking up trace amounts of the variety of elements present. These elements are in the form of dissolved or suspended material and may be activated in the neutron field to give radioactive isotopes, which remain in the water or deposit around the circuit. Thus, the primary cir­cuit of a P^^ is generally rather radioactive and requires remote mairtenance procedures. This activation can be minimized by very careful control of the water chemistry within the circuit and the choice of reactor materials. This sub­ject is of great importance in the economics of the system operation.

4. As discussed in Chapter 1, light water is a fairly strong absorber of neutrons, which leads to two problems. The first is the need for extra enrichment of the fuel, and the second is that under certain circumstances, accidental removal of the coolant water from the reactor core (e. g., by replacing the water by steam in a loss-of-coolant situation) may lead to an increase in the neutron popula­tion and an increased rate of the nuclear reaction. This is not a problem if the light water also serves as a moderator (as in the P^^ and BWR), since the moderator is also removed and the nuclear reaction stops. In the Russian RB^K reactor, where the main moderator is graphite, there are potential prob­lems with a loss of cooling water leading to a reactivity increase if the fuel channels are voided—what is often refe. red to as a positive void coefficient.

Thus, although light water is the most widely used coolant, it clearly is not ideal, but, to be fair, neither is any other coolant.

The Mihama-2 power station in Japan is equipped with a 500-^MW(e) two-loop pressurized water reactor. On February 9, 1991, this plant also experienced :> steam generator tube rupture that allowed high-pressure water from the reactor circuit to flow into the (lower-pressure) secondary circuit formed by the steam generator shell, the turbine, and the condensers.

At 12.24 h, with the reactor at full power, increasing radioactivity was sig­naled in the blowdown line of one of the plant’s steam generators. Further in­dications from the water in the steam generators and in the air extract from the condenser suggested an initially minor leakage of primary coolant water from a damaged tube in this steam generator.

Around 12.45, both the pressure in the primary circuit and the water level in the pressurizer started to decrease despite the activation of the pumps charging water into the coolant circuit. Three minutes later, a reduction of reactor power was initiated. The isolating valves on the steam line from the affected steam generator were activated but failed to close completely. Two minutes later, the reactor, turbine, and generator were automatically shut down. Almost immedi­ately, reducing pressure and level in the pressurizer signaled an emergency core cooling system (ECCS) water injection. At the same time, the reactor con­tainment was automatically isolated. Primary loop pressure and water levels continued to reduce rapidly.

At 13.52, the feed flow to the damaged steam generator was stopped and the unit was isolated. Ten minutes later, the relief valve on the steam line from the remaining undamaged steam generator (B) was lifted to allow decay heat re­moval. Attempts were also made to equalize the pressures on the primary and secondary sides by opening the pressurizer relief valves. However, this opera­tion was not successful, and depressurization of the primary circuit was there­fore undertaken via the alternative of the pressurizer spray system.

By 14.34, the HPIS pumps were stopped and 12 minutes later the primary and secondaiy circuit pressures were equalized, terminating the release. The re­actor reached a safe, “cold standby condition" at 02.30, February 10, 1991.

Analysis after the event suggested that 55 tons of coolant passed from the primary to the secondary systems and 1.3 tons of radioactive steam bypassed the main steam isolating valve and were released to the turbine hall.

After the accident, a camera was lowered into the damaged steam generator and located the fractured tube. The tube was removed for inspection. It had ruptured close to the top (sixth) tube support plate (Figure 5.10). There was ev­idence of fatigue failure due to vibration of the tube. Corrosion debris was also found between the tube and the support plate.

This type of recirculating steam generator is equipped with V-shaped antivibra-

tion bars to prevent fluid-induced tibe vibration in the retirn bend region. How­ever, fiber optic obseivations showed that these antivibration bars had not been correctly installed during constriction and did not support the damaged tibe (which was also restrained by the debris at the tube support plate; Figure 5.11).

The valve malfunction—the improper seating of the main steam isolating valve and the inoperative pressurizer valves—was also investigated.

The Mihama-2 incident was a classic steam generator tube rupture-small loss of coolant event, and the releases of radioactive rare gases and iodine were small. The plant was out of operation for a considerable period of time while its steam generators were replaced.

8.1 INTRODUCTION

All forms of energy production result in the formation of waste, the safe man­agement of which is essential for the protection of the public and the environ­ment. These wastes may be produced at various stages of the fuel cycle: extraction, refining, and utilization. In the case of the use of fossile fuels the main waste products from combustion are carbon dioxide and the “acid rain” gases: sulphur dioxide and nitrous oxides. Even in the case of “clean” renew­able energy sources, waste products associated, for example, with the produc­tion of photovoltaic materials or from geothermal systems need to be taken into account in any environmental balance sheet.

Nuclear energy is no exception and waste products are formed at each stage of the fuel cycle. Some of these waste products are radioactive, and it is neces­sary to handle, store, and dispose of these materials with extreme care. The long-term disposal of radioactive species is an integral part of the design and operation of the nuclear fuel cycle.

The classical view of the origin of the planet Earth was that it was formed from material torn out of the Sun, possibly by the gravitational pull of a star that passed close by. The material torn out would be initially in gaseous form and would then condense into a liquid, which would solidify on its outer surface, forming Earth’s crnst. This view is now considered to be unlikely because the materials of which Earth is made (iron, calcium, magnesium, aluminum, etc.) do not normally occur inside stars like the Sun. Planets like Earth are, in fact, some­thing of an oddity in the whole collection of galactic material, which consists mainly of elements such as hydrogen and helium. To create heavier elements such as carbon and neon by fusion of lighter elements, a temperature of 200 million eC is required, and even heavier elements (iron, cobalt, nickel, etc). re­quire temperatures of 4500 million eC for their formation. Such temperatures do not exist in the Sun, but they have been postulated to occur in “supernovaes,” explosions of great violence in which giant stars end their lives.

Hawking (1988) has described the process of creation of our universe (Figure 1.3). When the “big bang” occurred, the universe was infinitely small and also infinitely hot. Seconds later the temperature had fallen to 1010 degrees and the initial expanding universe consisted mainly of particles; photons, elec­trons, protons, and neutrons. After a hundred or so seconds these particles started to combine to form the nuclei of helium, hydrogen, and “heavy" hy­drogen (deuterium). This process was completed within just a few hours, and the production of helium and hydrogen then ceased. For the next million or so years the universe expanded as the temperature dropped to a few thousand de­grees. Inhomogeneities developed and some of the denser regions of the gas cloud stopped expanding and started to collapse under gravitational attraction.

The collapse caused rotation, and disklike galaxies were born. Slowly regions of higher density were formed heated by fusion reactions converting hydrogen to helium, and “stable” stars were created. However, the larger the star the more rapid the consumption of hydrogen. Heavier elements like carbon and oxygen were formed as a result of the very high temperatures, but finally a crisis oc­curred with the exhaustion of the hydrogen fuel: the star collapsed in an explo­sive way (a supernova’), and in the final stages of the supernovae explosions the heavier elements were formed. These first-generation stars lasted a few hundred million years, and the debris from their destruction together with the original gaseous cloud formed the source material for second- or third-genera­tion stars like our Sun. These were formed about 5 thousand million years ago.

Modern theories suggest that the formation of the planets (for instance, Earth) was a multistage process. First, the solar nebula was formed by the col­lapse of the dense rotating interstellar gas cloud containing the debris from

lier supernovae condensed to form dust. In the 1950s and 1960s it was assumed that the second stage consisted of the accretion of this dust into an initially solid and cold Earth —in contrast to the classical theory in which Earth was initially a molten body that gradually cooled. It was necessary to postulate a mechanism by which the core of Earth became molten, and the explanation for this was be­lieved to be in the behavior of the radioactive materials uranium, thorium, and potassium within Earth’s core. Once temperatures were high enough, melting, followed by segregation of the molten core, was assumed to occur. However, analysis of the crustal rocks has provided an estimate of the time at which this segregation must have happened. This suggests that core formation occurred relatively early and that Earth must have accreted in a relatively hot condition.

This important new finding has led to alternative theories for the stages sub­sequent to the formation of the solar nebula. These more recent theories as­sume the dust (associated with the residual gases) accreted rapidly—within a few million years—to form larger kilometer-sized bodies, referred to as “plan — etesimals.” These were initially hot, heated by the compression during the ini­tial collapse phase and perhaps also by radioactive decay of short-lived isotopes like 26Al. These planetesimals in turn slowly accreted into planets by mutual in­teraction. As the planets approached their final masses, they were impacted by larger and larger objects. Indeed it is now believed that one or more very large Mars-like objects impacted within Earth at a late stage in the accretion process. The vaporized material ejected then coalesced in orbit around Earth to form the Moon. This hypothesis can explain the absence of a metallic core in the Moon and the high angular momentum of the Earth-Moon system. This massive im­pact would also have completely melted Earth with surface temperatures as high as 16,000 K. It would also have completely changed the structure and composition of Earth’s crust compared with that which had existed during the early accretion period.

Many atomic species were formed during the cosmic processes described above, some of which are unstable. We shall now focus on the atom, and on the particular case of uranium, which is central to our story.

Figure 1.1 shows a typical impression of an atom. It consists of a nucleus made up of dense particles called nucleons. These are of two main types, namely, protons, which each carry one unit of positive electric charge, and neu­trons, which have the same mass as protons but are electrically neutral. Thus, the nucleus has a total electrical charge equal to the number of protons within it. Orbiting around the nucleus there is a cloud of electrons, which can be thought of as very small particles (compared to the nucleons). Each electron carries one unit of negative charge, and, to maintain a balance of electrical charges for the atom, the number of electrons equals the number of protons. The number of protons determines the atomic number of the particular species of atom of a given chemical element. The total number of nucleons (neutrons plus protons) determines the atomic mass. An atom of hydrogen has a nucleus consisting of only one proton and a single electron orbiting around it. The car­bon-12 atom, shown in Figure 1.1, has 6 protons and 6 neutrons in its nucleus; thus its atomic mass is 12. There are 6 electrons orbiting around the carbon-12 nucleus. At the other end of the atomic mass scale, the most common form of uranium atom, uranium-238, has 92 protons plus 146 neutrons in its nucleus and 92 electrons orbiting around the nucleus.

Most of any given atom consists of space. If a hydrogen atom was magnified until it was 100 meters across, the electron would resemble a pinhead revolving around a ball bearing 50 meters away. The actual density of the material in the nucleus (the ball bearing) is incredibly high, typically 240 million metric tons per cubic centimeter.

Stable nuclei of low atomic mass tend to have about the same number of protons and neutrons, and those of higher atomic mass have about four and a half times as many neutrons as protons. Nuclei in which the ratio of neutrons to protons departs from this value are unstable, and they may undergo a sponta­neous change in the direction of stability. During this spontaneous change, var­ious forms of radiation are emitted from the nucleus:

Alpha (a) radiation is the emission of a particle having a mass four times that of the hydrogen nucleus and consisting of two protons and two neutrons. An a-particle is identical to the nucleus of the element helium. Beta ф) radiation consists of very small charged particles, namely, electrons or positrons (positive electrons).

Gamma (y) radiation is electromagnetic radiation that is similar in nature to light or radio waves, except that it has a very short wavelength and is capable of penetrating through a large thickness of matter.

Neutron radiation is the emission of neutrons. This occurs in a number of decay processes and can help to start the nuclear fission reaction, as described below.

The radiation arising from nuclear decay is emitted at very high velocities, typically 8000 km (or 5000 miles) per second for a, P, and neutron radiation and the speed of light for y radiation. The creation of this kinetic energy results in a small decrease in the total mass of the system, as described above. The emitted particles collide with surrounding atoms, causing them to move and vi­brate—in other words, increasing their thermal energy. Thus, the decay process leads to the generation of thermal energy.

For each chemical species, corresponding to a given atomic number, there are often several possible configurations of the nucleus characterized by differ­ent numbers of neutrons, the number of protons remaining constant. That is, several different values of the atomic mass are possible for a given atomic num­ber. Each value of the atomic mass characterizes an isotope of the element in question. Individual isotopes are described symbolically by giving the atomic mass as a superscript before the symbol for the element; thus the most common isotope of carbon is described as 12C, though small amounts of 13C and 14C exist in all natural forms of carbon. Similarly, the element uranium exists naturally in three isotopic forms, namely, 234U, 235U, and ^U.

A number of the isotopes that exist in nature are unstable and are subject to decay by the process described above. When an isotope decays to another form through the emission of alpha, beta, or gamma rays, the new form may it­self be unstable and may also decay. Eventually, a stable form will be reached, but many stages may have to be gone through before this is achieved. The re­sulting decay chains can be very long for the elements with atomic mass num­bers above 200. The decay chain for uranium (235U) is shown in Figure 1.4.

The decay processes illustrated in Figure 1.4 result in the emission of heat as the emitted rays are absorbed. These uranium isotopes were present in the material from which Earth was formed, as were a number of other unstable radioactive isotopes. An example is potassium-40 (40K), which decays very slowly to argon-40 (40Ar) by emitting beta radiation. Other unstable isotopes that may have been present are aluminum-26 (26Aft) and palladium-107 (107Pd). All of these radioactive decay processes led to the release of thermal energy into Earth’s material. An important parameter governing the rate of release of thermal energy in this way is the half-life of an unstable isotope, which is the time required for half the unstable atoms originally present to decay to their new isotopic form. Thus, after a period corresponding to 1 half-life, half the original atoms remain; after 2 half-lives, a quarter remain; after 3 half-lives, an eighth remain; and so on. After 10 half-lives, only 0.1% of the original material remains. The half-lives of 238u and 235u are 4500 million and 700 million years, respectively. The half-life of 4°K is 1300 million years. Other isotopes that were originally quite abundant on Earth include 26Al, which has a half-life of 0.7

million years, and 107Pd, which has a half-life of 6 million years.

Although the decay of these naturally occurring isotopes is extremely slow, a very large amount of decay has occurred within the lifetime of Earth (4500 mil­lion years—comparable to the half-life of 238U). Heat released by radioactive decay can escape from Earth only by being conducted to the surface and being radiated away into space. However, heat loss from the interior of Earth to the surface is quite small (currently about 30,000 GJ per second).

Future development of nuclear reactors is aimed at improved performance— from both an economic and a safety viewpoint. Electricity utilities in both Eu­rope and the United States are collectively defining these requirements in detail. A key issue in the application of future designs is their ability to be licensed in those countries wishing to deploy the design, in the same way as aircraft de­signs achieve their airworthiness certificates to operate internationally.

Most attention has been directed at advanced light-water reactors (AL^TC). Two different approaches are being pursued to meet these improved-perfor­mance goals; they relate to evolutionary and passive designs, respectively. Evolutionary designs are extensions of existing P^^ and B’^R plants build-

ing upon past experiences and using proven components but with enhanced safety features designed to reduce the probability of accidents and to mitigate their consequences. Specific examples of these plants and their vendors are the European pressurized water reactor—EPR (NPI); System 80 plus(IM)/BWR 90 (ABB); ABWR (GE); and Sizewell B/AP’^TC. (Westinghouse/Mitsubishi).

Pasive designs make effective use of natural physical processes such as grav­ity (control rod insertion), natural circulation/convection (to remove heat), evap­oration-condensation, transient heat conduction (to provide heat sinks), stored energy in pressurized accumulators (to inject cooling water), and negative reac­tivity effects (to stabilize the chain reaction). Although there are drawbacks, the maximum use of such natural phenomena can simplify the design and reduce de­pendence on operator action. Specific examples of passive plants and their ven­dors are the simplified BWR (GE); PIUS (ABB); and AP600 (Westinghouse).

REFERENCE

Etherington, H 0958). Nuclear Engineering Handbook, sec. 9.3, 9-91. McGraw-Hill, New York.

^^MPLES PROB^LEMS

1 Nudearfuel center temperature

Example: Derive an expression for the temperature at the center of a nuclear fuel pel­let assuming that the internal energy generation is uniform and the thermal conductiv­ity is independent of temperature. A solid U02 pellet has a linear rating of 45 kW/m and a surface temperature of 600°C. The thermal conductivity of UO2 is 2.7 W/m K. What is the center temperature of the fuel pellet?

Solution:

Suppose the rating of the fuel pellet, i. e., the total energy supplied as heat per meter of fuel, is R (W/m). Then the rate of energy release within a radius r is since the power produced is in proportion to the volume of fuel.

At equilibrium this rate of energy is conducted away from the cylindrical surface

at r, i. e.,

, ^ (dT — k2n r — ^ dr

where k is the thermal conductivity and Tis the temperature. So

R

41tk

where To is the temperature of the outside of the pellet (r = a). Note that the difference in temperature (TMAX — T), when the energy release rate is expressed as a linear rat­ing,. is independent of the diameter of the pellet.

45___

4 X 1tx 2.7

Problem: For the U02 pellet described in the example, calculate the maximum linear rating that would be possible if the center temperature were limited to 1500°C.

2

Figure of mentfor a reactor coolant Example: A figure of merit for a reactor coolant is given as

Derive this expression from a consideration of the ratio of pumping power P to heat output Q for a constant coolant temperature rise A T.

Solution: The pressure drop Lp across a channel of diameter D and length L is

л 1 W

llpW

where W is the flow rate of the coolant, A is the flow cross-sectional area, f is the friction factor, which for turbulent flow is proportional to

fWD

and e is the coolant density.

The pumping power P ( = ApWI @) is

2ft D

The heat output Q can be given in terms of the flow rate, specific heat, and tempera­ture rise of the coolant:

Q = WCpAT

If we use this equation to eliminate W from the pumping power equation, then

but the friction factor fis proportional to

Thus P is proportional to

2L f Q2 8 } Ц0,2

a2d12 [at2S Jc^g2

Therefore, for given channel dimensions L, A, and D, heat output Q, and temperature rise of coolant A T, the pumping power will be a minimum when

is a minimum or the reciprocal is a maximum.

Problem: A new organic coolant is being considered for reactor cooling. At the condi­tion obtained in the proposed reactor, its density is 862 kg! rn3, its viscosity is 1.5 x 1^ Ns/m3, and its specific heat is 2710 J/kg K. Calculate the figure of merit for this new coolant and compare the value obtained with those for other coolants given in Table 3.1.

6.1 INTRODUCTION

In Chapters 4 and 5 we discussed the means by which loss-of-coolant accidents (LOCAs) could occur and the ways in which reactors must be designed to cope with these extremely unlikely events. We also discussed in Chapter 5 a number of actual incidents in reactors where a failure of cooling occurred with conse­quent overheating and fuel damage. Many of these conditions were anticipated in the design, but some actually went beyond the design basis. In all cases, ex­cept Chernobyl and Windscale, the “defense in depth” approach to nuclear re­actor design was effective in limiting the public consequences of the accident. However, it is important to consider what might be involved in extremely se­vere accidents having the common characteristics of leading to partial or com­plete meltdown of the fuel within the reactor.

In classifying operational states in Section 4.1, we considered a series of tran­sient events in reactors ranging from operational transients to limiting fault con­ditions. Even in a limiting fault condition, the reactor is designed so that there is no loss of coolability of the core over protracted periods. However, one can postulate a situation in which the emergency core cooling system (ECCS) itself fails and no other cooling system is available. Another possibility would be loss — of-site power over a long period, coupled with inability to actuate the alterna­tive power sources (normally on-site diesel engines). A third possibility is that of unpredicted operator faults, which may lead, as at Three Mile Island, to con­ditions beyond those designed for as limiting.

Also in Section 4.1 we described the concept of containment and the various barriers preventing the release of activity:

• The matrix of the fuel itself and the cladding around the fuel

• The reactor pressure vessel

• The containment building or system

tor was to ensure that these separate barriers are not challenged and all remain intact. That is embodied in the safety case. But suppose these systems are de­graded in some way or are inoperative and their purpose is not achieved. What then?

To answer this question, it is therefore informative to examine how each bar­rier might be challenged and the failure mode and consequences that might re­sult. That, in turn, will lead to consideration of design measures to limit the failure or mitigate the consequences.

Research into controlled fusion reactions is proceeding in the United States, Russia, Japan, and Europe. One particular configuration of magnetic fields has proved promising: the so-called Tokamak configuration. For interested readers. Figure 9.6 shows the details of the Tokamak device and explains why three separate magnetic fields are used to control the plasma. Experiments have been conducted with larger and larger devices. Figure 9.7 shows the progress made

toward the achievement of the Lawson criterion. One particular experimental Tokamak is the so-called JET (Joint European Torus) project. The scale of this experiment can be judged from the fact that the mean diameter of the torus is 6 m and the radius of the plasma will be 1.25 to 2.10 m. An illustration of what JET looks like is shown in Figure 9.8.

On November 9, 1991, at 07.44 p. m., the JET experiment produced about 2 MW of fusion power, the first time that a significant amount of power had been obtained from controlled nuclear fusion reactions. This was also the first time JET had been operated with mixtures of the isotopes ( 14% tritium-86% deu — trium). In practice only about 0.2 gram of tritium was used. For this experiment JET consumed far more power than it generated. The next target is breakeven, the production of as much fusion power as consumed in heating the plasma. Further experiments are planned at JET to obtain and study plasmas under con­ditions and dimensions approaching those needed in a thermonuclear reactor (Figure 9.7).

Plasmas are usually heated by passing a current through the electrically con­ducting plasma. This form of heating (Figure 9.9) is effective up to about 10 mil-

Predicted performance of large Tokamaks (such as JET). under construction (1983-4)

Repent

Inte^nnediate size

Toka^maks

(1977-1980)

Second generation of Tokamaks (1970.1980)

First generation of Tokamaks (1965-1970)

lion degrees, but if attempts are made to increase the current, instabilities set in. Other methods of heating include one in vhich a beam of ionized particles is accelerated up to high energies, neutralized, and fired into the plasma. Once in the plasma, the particles become ionized, are trapped, and transfer their energy by collision with the plasma electrons. Other methods of heating include radio frequency (RF) heating and compression of the plasma with a magnetic field. All these methods have been tried on JET. They are now understood and their use can be contemplated on fusion reactors with confidence.

From this short discussion of the present status of our k nowledge of fu sion reactions, it will be seen tha t we have just about re ached the point where a con­trolled fusion reaction has been demonstrated. Th us we have reached the point in the development of fusion power that Fermi achieved in J942 wit h the first

Figure 9.9: Methods of heating torodial plasmas. ( o) Resistive heating; current flow­ing in plasma causes it to heat. ( Neutral beam injection heating. ШісІкЯге-

qtiency heating: typical frequencies lie in ranges: 60-100 GHz. (dt’CU’on слчіоП’оп heating). 1-8 GHz (lower hybrid heating). 80-100 MHz (ion cyclotron headings GO Adiabatic compression heating.

Two isoto^^ of ^hydrogen-DEUTERIUM and TRITIUM

are heated theether at a very high temperature in a reaction chamher

In this blanket the ^^rods:

• React lithium to prod uce TRITIUM end HELIUM

■The heat Is removed to raise steam for e^ctiicity production

The Tritium produped in the blanket Is retu^od to system to be a fuel її

controlled fission reaction. Progress is limited by (1) the capabilities of remov­ing the heat from the neutrons deposited in the blanket and (2) the damage sus­tained due to the high-energy neutron radiation of the thin-walled vessel.

To progress to such a reactor requires the solution of many technical and en­gineering problems, a fair number of which involve the thermal engineer. (See Nuclear Engineering and Design, vol. 68, April 1982). For example, various coolants have been suggested to remove the heat from the blanket into heat ex­changers to raise steam. It is possible to use the lithium itself, or an alternative liquid metal, although the intense magnetic fields impose a very high pressure drop with consequent high pumping losses. An alternative is to use a gaseous coolant, but this limits the energy density possible through the vacuum vessel wall (first wall).

Bickerton (1993) has summarized some of the requirements for the reactor:

• Start-up, ionise gas and increase ring current to final value (- 20 ^A)

• Heat plasma with auxiliary heating, typically < 100 MW to ignition point

• Maintaining ignition, manoeuvre plasma parameters to final operating point, where required fusion thermal power (-3 GW) is achieved

• Maintain plasma current in steady state for long pulse (- 1000 sec.)

• Extract fusion ash; i. e. helium in the case of D-T reactions

• Refuel plasma (with fresh deuterium and tritium in case of D-T reactions

• Extract heat at high efficiency from blanket, tritium breeding ration >1.0

• Shield super-conducting coils from neutron induced heating and radiation damage

• Maintain system remotely, e. g. change first wall every 2-5 years (because of neutron damage)

• Decommission and dispose of wastes.

Solutions of these and other engineering problems are being sought in the design of the next generation of Tokamaks. Because of the scale of the ex­periment, all the major nations involved in fusion research are collaborating on the design of the next machine, ITER (International Tokamak Experimental Reactor).

The design of this machine is based on the scaling laws derived from previ­ous experiments that predicted plasma performance as a function of machine parameters. The main ITER parameters are given as:

The overall objective is nothing less than a demonstration of the feasibility of fusion energy for peaceful purposes. The outline design is well advanced (Toschi, 1995).

Regarding impact on the environment, fusion reactors have some advantage over fission reactors. The waste product of the fusion reaction, helium, is inert, and thus the problem of managing highly radioactive waste does not arise. The structure of the reactor itself will become intensely radioactive and will require remote maintenance. But this radioactivity will decay over periods of hundreds rather than tens of thousands of years. The tritium used in the reactor represents a radiological hazard, and since it is an isotope of hydrogen, it requires very careful containment and protection against accidents, such as fires. In summary, although the potential radiation hazards presented by fusion reactors will be less than those of fission reactors, they will require careful attention at the engi­neering design stage.

9.3 CONCLUSIONS

With the probable development of fusion in addition to fission energy, nuclear power presents humans with a virtually infinite source of energy. The central role of energy in our economic structure has been very clearly demonstrated over the 20 years since the oil crises of the 1970s. Nuclear fission energy pro­vides a proven resource for the immediate future and nuclear fusion energy a great potential resource for the more distant future. Humanity must make use of these resources, particularly if the underdeveloped world is to achieve freedom from the bondage of hunger, disease, and poverty, and the world is to sustain its development.

Of course, there are many technical problems still to be solved, and the uti­lization of nuclear power will demand continual vigilance and great attention to technical detail if it is to continue its very successful beginning. Not least of these problems are those associated with the removal of heat from the nuclear

reaction and its effective application in power generation. We thus make no apology for having written this hook from our own viewpoints, those of ther­mal engineers.

In addition, there are great institutional and organizational problems to be properly resolved before the fill potential of nuclear power can be realized. The development of international cooperation in this area may set an example to other spheres, and make more tolerable our existence on this beautiful planet.

REFERENCES

Bickerton, R. J. (1993). ‘"The Purpose, Status and Future of Fusion Research.” Plasma Phys. Control. Fusion 35, B3-B21 lOP Pub. Ltd.

Carruthers, R. A. 0981). “The Fusion Dilemma.” Interdisciplinary Sci. Rev. 6 (2): 127-141.

Toschi, R. 0995). ‘ITER: The World’s Fusion Project.” Nuclear Europe Worldscan 15 (January-February): 55-57.

Williams, L. O. 0994). ‘‘The Energy Source: Nuclear Fusion Reactors.” Applied Energy 47: 147-67.

As discussed above, the U. S.-designed P^^ and B^^ reactors require consid­erable enrichment of the uranium in order to overcome the relatively high ab­sorption of neutrons by the light-water coolant. This disadvantage can be overcome by using heavy water as a moderator and either heavy water or boil­ing light water as the coolant. If heavy water itself is used as the coolant, it is possible to operate with natural uranium. This is the principle adopted in the Canadian-designed CANDU (Canadian deuterium-uranium) reactors, which are illustrated in Figure 2.11.

CANDU reactors dispense with the massive thick-walled pressure vessel used in P^WRs and B^WRs; instead, the fuel elements are placed in horizontal pressure tubes constructed from zirconium alloy. These pressure tubes pass through a calandria filled with heavy water at low pressure and temperature. In the CANDU reactor, heavy-water coolant is also passed over the fuel elements at a pressure of approximately 90 bars (1400 psia). It then passes to a steam generator, which is very similar to that used in the P^^ (see Figure 2.7). It should be noted that CANDU reactors have not experienced the same steam generator problems as the P^WRs, possibly because of the lower operating tem­perature on the primary side. The fuel elements consist of bundles of natural

Pnt^uurtsed heavy ^water reactor (CANDU)

End v^v inside pressure tube

— ZireolO>’ end support elate

Figure 2.11: (a) Heavy water-moderated and —cooled CANDU reactor; the fuel is natural uranium oxide; (b) end and side views of a horizontal pressure tube.

U02 pellets clad in zirconium alloy cans; individual bundles are about 50 cm long, and about 12 such bundles are placed in each pressure tube. The average volumetric power density in a CANDU core is approximately one-tenth that in a P^^ (since the moderator volume is taken into account in calculating the av­erage volumetric power density) and nearly four times that in an AGR.

However, the fuel rating is comparable to that in a P^WR Furthermore, the fuel is very much cheaper since natural uranium can be used. Although the CANDU has operated with remarkable success, difficulties have been experienced with hydriding of the zirconium alloy pressure tubes, necessitating their replacement in some cases. Even though it has a lower fuel cost, CANDU needs considerable amounts of expensive heavy water, which makes its capital cost high.

In the CANDU reactor, the coolant is distributed to and collected from the core by pipes known as headers, which are connected in turn to each of the fuel channels by other tubes known as feeders. The circuit for the CANDU reactor is illustrated schematically in Figure 4.33.

If a loss-of-coolant accident occurs, emergency coolant is injected into all the

headers by a separate emergency coolant injection (ECI) system. This system supplies light water to the reactor during the LOCA as shown schematically in Figure 4.33. The system has a high-pressure injection stage in which gas pres­sure is employed to inject the water into the headers in a manner similar to the accumulators in the P^^. In some designs, this gas-pressurized system is re­placed with high-pressure pumps that draw water from an emergency water tank. When the high-pressure supply is exhausted, water is pumped at a lower pressure from a separate water tank and fed into the reactor. Finally, the water being emitted from the reactor circuit into the containment building is recov­ered and pumped back to the headers via a heat exchanger that cools the en­tering water stream.

With the CANDU reactor, there are two main disadvantages related to be­havior during a LOCA:

1. The reactor channels are horizontal. This means that if steam voids are formed on the channel, the water phase separates toward the bottom of the channel, leaving the top part of the channel in steam and relatively un­cooled. This gravitational stratification effect is of great importance in con­sidering the behavior of the fuel following a LOCA.

2. The CANDU reactor has a positive void coefficient; i. e., when voids are formed in the heavy-water coolant, the reactivity increases because the cre­ation of the voids in the fuel channel makes little difference to the overall volume of moderator in the system. Thus, the neutron absorption in the heavy water in the fuel channels is removed and the reactivity increases. In a typical transient the fuel power can increase by a factor of 2 within 1 s after the accident, followed by a rapid decrease as the shutdown systems begin to operate. In view of these positive reactivity effects, it is important, for safety, to have two independent shutdown systems, as illustrated in Figure 4.34. In the first system, cadmium shutoff rods fall under gravity from the top of the reactor. In the second a neutron-absorbing solution (poison) is injected through horizontal nozzles into the heavy-water moderator surrounding the fuel channels.

Another potential problem with the CANDU reactor under LOCA condi­tions (with a break, say, in the inlet header) is that of flow stagnation. Water is sucked out of one end of the channel by the pump and leaves from the other end of the channel toward the break. The center part of the channel can therefore be stagnant, and this leads to rapid overheating of the fuel. In the design of the CANDU reactor, careful attention must be given to these poten­tial problems. However, there are two mitigating features of CANDU reactors that help in this regard:

1. In accidents that cause the fuel and pressure tube to heat up, substantial amounts of heat can be transferred to the moderator, which can serve as an in-core heat sink.

2. Since the control rods that penetrate the cold, low-pressure moderator are operating under low-temperature conditions, it can be argued that the sys­tems are much more reliable than those which operate at high temperature and pressure. A more detailed review of the safety of CANDU reactors is given by Snell, V. G., et al. (1990).

Should the containment fail, fission products will be released into the atmos­phere. There is much discussion about the extent to which this would happen. The gaseous fission products are usually assumed to be released completely and other volatile fission products such as caesium and iodine are assumed to be partly released. Other fission products are released in very small quantities and do not usually contribute significantly to the calculated hazard. Typical fractions of caesium and iodine assumed to be released are around 10%. For the less volatile fission products, fractions around 1% are often assumed, on the basis of experimental studies of fission product retention. The ultimate disper­sal of the fission products from such a release is calculated by using computer codes and depends greatly on the weather conditions at the time of release.

REFERENCES

Baumgartl, B. J., and F. Bouteille 0994). “The European Pressurized Water Reactor (EPR): An Advanced PWR.” Re>ue Generate Nucleaire, November-December, 478-483.

Gittus, J. H., et al. (1982). “PWR Degraded Core Analysis.” Report ND-R-610(5), U. K. Atomic Enerev Authoritv.

Hennies, H. H. (1993). “Research and Development to Improve Containment for the Next Generation of Pressurized Water Reactor Plants.” Interdisciplinary Science Re­vues 18 (3): 243-51.

Pilch, M., Allen, M., and T. Blanchat 0994). “Can Containment Buildings Take the Heat?" Atom 434 (June-July).

Prior, R. 0992). Severe Accident Progression. Core and Reactor Systems Phenomena. Se­vere Accidents and Accident Management in Light Water Reactors, March 23-27, Lyon, France.

The Three Mile Island Reactor Pressure Vessel Investigation Project: Achievement and Results. (1993). Proceedings of an Open Forum Sponsored by the OECD/NEA and the US NRC, Boston, October 20-22.

Turland, B. D., and R. F. Peckover 0978). “Melting Front Phenomena." Report CLM-P- 564, U. K. Atomic Energy Authority..

———- 0979). “Melting Front Phenomena.” Eur. Appl. Res. Rept. 1 (6): 185-201.