Natural uranium is found in ore deposits in many places around the world. It is predominantly a mix­ture of the two isotopes uranium 238 (99.27%) and uranium 235 (0.72%), with a trace of uranium 234. All three isotopes are radioactive.

Therefore of the three fissile materials mentioned in Section 3.2.3, natural uranium is a direct source for one, U-235, and an indirect source for a second, Pu-239 via the fertile U-238. This underlines the im­portance of natural uranium in nuclear power. The third fissile material, U-233, is of little significance, at present, although of possibly important potential because of the large ore reserves of the fertile thorium.

Before considering the fission event further it will be useful to summarise some of the key properties of natural uranium and its isotopes:

• Natural uranium consists of

99.3% U-238 a emitter half life 4.5 x 109y

0.7% U-235 a emitter half life 7.1 x 108y

0.1% U-234 a emitter half life 2.5 x 105y.

• U-238 is a fissionable material; it can undergo fission provided the absorbed neutron has a kinetic energy of at least 1.1 MeV, Fig 1.6 (c).

• U-238 is also a fertile material, forming fissile Pu-239 following capture of a neutron. Neutrons of intermediate energy are very readily captured in the resonance capture peaks of U-238, Fig 1.6 (b). •

3.3 Properties of the fission event

The fission reaction generally gives two primary fis­sion products, a number x say of neutrons and some energy as gamma rays. Figure 1.7 illustrates the fission event; the fission reaction of, for example, U-235 can also be represented thus:

“35u + ln A1(fp, — A2(pp) xn +

92 0 Z1 Z2

Although it is possible for the nucleus to break into more than two pieces (especially when the incoming neutron has high energy) the most likely event is binary fission, giving two fission products.

A number of questions arise:

• What sizes are the fission fragments?

• How many neutrons are emitted?

• What are their energy?

• How many are prompt and how many are delayed?

• How much total energy is released in the fission? Where and how does it appear?

Section 8.2 of this chapter illustrated that the neu­tron and heat flux in a radial direction across the core follows a first order Bessel function distribution usually approximated as a cosine. In a gas cooled re­actor such as magnox or AGR, similar gas flow rates in each channel would result in a temperature distri­bution at the channel outlets also approximating to a cosine, A gas temperature geometry of this sort would result in many of the peripheral channels operating appreciably below the optimum outlet temperature with a resulting loss of thermal power.

The thermal distribution profile can be flattened by varying the gas flow rates across the core, i. e., maximum at the centre, minimum at the periphery. In both magnox and AGR designs this variation in flow rate is achieved by fixed ‘gags’ at channel inlets. The position of each gag is adjusted during the reactor commissioning to optimise on the thermal power ex­tracted from the core.

In addition to the fixed gags the AGR channels have variable flow gags which can be adjusted on­load. These are described in detail in Chapter 2 of this volume.

8.11 Fluid flow

When discussing the various factors affecting the sur­face heat transfer coefficient, the importance of high flow rates and high turbulence was emphasised par­ticularly with gaseous coolants. Unfortunately whilst an increase in the value of both these factors will enhance the rate of heat transfer it will also increase the circulator power necessary to achieve those values. The design of the core flow conditions must be opti­mised to achieve a high (thermal output to circulator power) ratio and a study of fluid flow is necessary to evaluate the principal factors affecting the circulator power.

The gas circulator must be designed to achieve the required flow rates and also to overcome the various pressure drops in the gas circuit due to fuel element geometry, gas duct bends, boiler tubes, etc. We can obtain a good indication of the various factors af­fecting the level of pressure drops in a channel by considering the results obtained from what is known as Reynold’s Experiment. If in a horizontal pipe of length (L) we increase the flow velocity and measure the pressure drop (Др), a plot of pressure drop per unit length against velocity (v) results in Fig 1.26.

AB is a straight line showing that up to point B, which is a definite transition point, the pressure drop is-proportional to the velocity. BC is an approximate parabolic curve and over this range of flow the pressure drop is approximately proportional to the (velocity)2. A study of. the flow regime in the pipe would show that A-В represents a smooth streamline flow and

B-C represents turbulent conditions. For gaseous cool­ants in particular it is essential to operate with turbulent now conditions in order to achieve an eco­nomic rate of heat transfer; for most practical reactor problems it is therefore convenient to assume that pressure drop is proportional to the square of the flow velocity.

Using this assumption an empirical expression has been developed which will enable the pressure drop per unit length of pipe or duct to be determined:

др — 4 f 6 v2 L 2 de

where Др = pressure drop over the length L

f = the friction factor which is dependent upon the geometry of the pipe or duct and which is determined by experiment

6 = fluid density

de = equivalent hydraulic diameter

The presence of a density term in this expression (6) means that it can only be used accurately if the density is constant over the length of the channel, i. e., the flow is incompressible. In applying it to a reactor fuel channel therefore, allowance must be made for changes of density due to variation in tem­perature and pressure along the channel. These varia­tions arise principally from the heat transfer processes which we have discussed. Thus an expression which will give an accurate determination of channel pressure drop taking these thermodynamic changes into ac­count is complex, but an approximate expression useful for illustrating the principal factors affecting pressure drop is:

Др = G2/6m [(2fL/de) + (At/tm) + (Ap/pm)] where G = mass velocity (density x velocity)

6m = mean density Дt = (t] — t2) = temperature rise

tm = (t і + t2)/2 = average temperature

Др = (P] — P2) = pressure drop

pm = (P і + P2)/2 = average pressure

The assumptions made in arriving at this approxi­mate expression are usually sufficiently accurate for first stage design calculations and the expression il­lustrates that the channel pressure drop is a function

of:

• The coolant properties and the flow rate.

• The channel geometry.

• The channel temperature distribution.

• The ratio between the channel pressure drop and the mean pressure.

Having established the pressure drop in the reactor channel, similar calculations are necessary to determine the pressure drop in the total coolant circuit, i. e., boilers, gas ducts, etc. This total pressure drop now becomes a major design parameter for the gas circu­lators. Circulators are described in detail in Chapter 2 of this volume. At this stage it is sufficient to indicate two expressions which can be used to deter­mine the gas circulator power:

• Circulator power = (pressure rise over the circu­lator) x (mean volume flow of gas).

• Circulator power = (mass flow) x (specific heat) x (circulator temperature rise).

G. A. W. Blackman, CBE, FEng

Chairman, Central Electricity Generating Board and Chairman, British Electricity International L td

For over thirty years, since its formation in 1958, the Central Electricity Generating Board (CEGB) has been at the forefront of technological advances in the design, construction, operation, and maintenance of power plant and transmission systems. During this time capacity increased almost fivefold, involving the introduction of thermal and nuclear generating units of 500 MW and 660 MW, to supply one of the largest integrated power systems in the world. In fulfilling its statutory responsibility to ensure continuity of a safe and economic supply of electricity, the CEGB built up a powerful engineering and scientific capability, and accumulated a wealth of experience in the operation and maintenance of power plant and systems. With the privatisation of the CEGB this experience and capability is being carried forward by its four successor companies — National Power, PowerGen, Nuclear Electric and National Grid.

At the heart of the CEGB’s success has been an awareness of the need to sustain and improve the skills and knowledge of its engineering and technical staff. This was achieved through formal and on-job training, aided by a series of textbooks covering the theory and practice for the whole range of technology to be found on a modern power station. A second edition of the series, known as Modern Power Station Practice, was produced in the early 1970s, and it was sold throughout the world to provide electricity undertakings, engineers and students with an account of the CEGB’s practices and hard-won experience. The edition had substantial worldwide sales and achieved recognition as the authoritative reference work on power generation,

A completely revised and enlarged (third) edition has now been produced which updates the relevant information in the earlier edition together with a comprehensive account of the solutions to the many engineering and environmental challenges encountered, and which puts on record the achievements of the CEGB during its lifetime as one of the world’s leading public electricity utilities.

In producing this third edition, the opportunity has been taken to restructure the information in the original eight volumes to provide a more logical and detailed exposition of the technical content. The series has also been extended to include three new volumes on ‘Station Commissioning’, ‘EHV Transmission’ and ‘System Operation’. Each of the eleven subject volumes had an Advisory Editor for the technical validation of the many contributions by individual authors, all of whom are recognised as authorities in their particular field of technology.

Ail subject volumes carry their own index and a twelfth volume provides a consolidated index for the series overall. Particular attention has been paid to the production of draft material, with text refined through a number of technical and language editorial stages and complemented by a large number of high quality illustrations. The result is a high standard of presentation designed to appeal to a w-ide international readership.

It is with much pleasure therefore that I introduce this new series, which has been attributed to British Electricity International on behalf of the CEGB and its successor companies. I have been closely associated with its production and have no doubt that it will be invaluable to engineers worldwide who are engaged in the design, construction, commissioning, operation and maintenance of modern power stations and systems.

March 1990

The nuclear reaction rate between moderator and neutron is given by:

R = Nas<t> (see Section 2.4.1 of this chapter)

number of neutrons colliding, m2/s

moderator microscopic cross-section for scattering

elastic scattering, as there cannot be inelastic scattering in this case

neutron flux

number of neutrons, m2/s

For a given neutron flux a high reaction rate value for R requires large N and large cts. The first require­ment — large N — means the moderator must be a solid or liquid; solids and liquids have much the same number of nuclei per unit volume in contrast to gases which have several orders of magnitude fewer nuclei.

The second requirement — large as — is self — explanatory. The value to put to os however is not simple as it represents a whole spectrum of scatter values over the 2 MeV/0.025 eV range.

As stated previously 7 rays are very high frequency electromagnetic radiation. They are highly penetrating: 10 cm of lead are required for example to attenuate 1 MeV 7 rays to one tenth of their initial intensity. All electromagnetic radiation, including 7 rays, may be regarded as coming in ‘packets’ or ‘bundles’ of discrete energy called photons whose energy E is proportional to the radiation frequency v. We have: E = hv where the constant of proportionality ‘h’ is known as Planck’s constant. Photons behave in many ways as a particle but, in contrast to the a and 0 radiation which are gradually slowed down as they interact with matter, photons must always travel with the speed of light. 7 rays interact with matter in three ways, as follows, in which the photons are either annihilated and disappear or are converted into photons of lower energy:

• Photoelectric effect The 7 rays interact with the orbital electrons of the atoms of the absorbing

material. A photon disappears in transferring all its energy to an electron, ionising the atom. The photoelectric effect, and thus the attenuation of 7 radiation, is much more likely to occur in the heavier elements and for 7 rays up to 1 MeV.

• Compton effect This is also an interaction with orbital electrons but here the photon transfers only part of its energy to an electron and is converted into a lower energy photon. The 7 ray continues at the speed of light but at a different angle and frequency from previously. Mono-energetic incident 7 rays will thus be scattered by the Compton effect and converted into 7 rays with a range of lower energies.

• Pair production In interacting with matter a 7 ray may disappear in producing an electron positron pair. To do this the photon must have a minimum energy equivalent to the mass created: hv ^2 meC2 ^ 1.022 MeV, where mt is the rest mass of the electron.

It should be noted that the expression keff — k« Pnl (Section 7.2 of this chapter) implies the assumption, that all the diffusing neutrons have the same energy and hence a single factor for the non-leakage proba-‘ bility — ‘the one group diffusion theory’. Clearly, this is not so and a refinement is the ‘two group dif­fusion theory’ where all the neutrons are taken to be either slow or fast neutrons and each group has the appropriate non-leakage probability factor. Thus:

kcrr = кос PNLs PNLf = pfTjf PNLs PNLf

where Psjls and PNLf refer to the slow and fast neu­tron group respectively. This expression is sometimes called the six factor formula.

Further refinements can be made by dividing the neutron energy spectrum into more and more distinct energy groups. This approach is called the ‘multi group theory’. The results quoted here are for the ‘one group theory’.

There are many different ways for a binary fission nucleus to split but in every case the mass number and atomic number on each side of the above equa­tion must be the same:

A1 + A2 +- x = 236 Z1 + Z2 = 92

Symmetrical fission with no release of neutrons would give

A1 = A2 = 118 and

Z1 = Z2 = 46 which is the element palladium.

However, experiment shows that symmetrical fission does not occur often, about one per 20 000 fissions. Asymmetrical fission is much more likely, see Fig 1.8. In general thermal fission of U-235 results in one product having a mass between 90 and 101 and the other between 132 and 143. Each of these masses occur in 6 or 7% of the fission yield and about 70% of all fissions give products within this range.

The fission yield distribution is similar in shape for other fissile materials but with slight differences in the position of one or other of the peaks. The dis­tribution also depends on the energy of the incoming neutron; higher energy neutrons give a broader spread of masses with a greater chance of symmetrical fission.

Fission products are generally highly radioactive /3/7 emitters, see Table 1.5. This radioactivity gives

rise to complex shielding problems that must be over­come in the handling and reprocessing of irradiated nuclear fuels. It also results in decay heat, see Section 3 of this chapter. Two other significant properties of fission products are that some may be precursors to delayed neutrons, Section 3, and some may very readily absorb neutrons — i. e., poisons with high neutron capture values.

4.10 Introduction

This section gives a brief description of the more important commercial nuclear power reactors used in the western world. For detailed information on their design and performance there is a wealth of excel­lent textbooks, reviews and reports readily available in the literature.

Commercial nuclear power reactors (Fig 1.27) are principally thermal reactors — that is, the fast neu­trons released by fission are slowed down in a mod­erator to speeds which correspond roughly to the moderator temperature; the neutrons are thus in ther­mal equilibrium with the moderator. Thermal neutrons are much more likely than fast neutrons to induce further fission enabling natural, or perhaps only slight­ly enriched, uranium fuel to be used. This contrasts with the alternative design approach of the fast re­actor in which the chain reaction is sustained by fast neutrons. There is, therefore, no requirement for a moderator but the fuel used must have a high pro­portion of fissile material, usually plutonium.

All the three possible moderator materials — carbon in the form of graphite, ordinary light water H20 and heavy water D20 — have been exploited commercially. The fuel used is either metallic uranium or the dioxide U02 to give much superior thermal and dimensional stability in service. The coolant is either gaseous (carbon dioxide, or helium where a high degree of inertness is required) or water (light or heavy). The fuel is encased in a cladding to prevent the highly radioactive fission products from entering the coolant and also to act as a chemical barrier between the fuel and coolant. Stainless steel or mag-

nox, an alloy of magnesium, is used in the case of the gaseous coolant and a zirconium alloy, zircaloy, for water coolant.

Table 1.8 lists some of the main design features of a number of commercial nuclear reactors. The table includes the fast reactor because of its future potential. Also included is the steam generating heavy water reactor SGHWR because of its UK interest, although there is now little likelihood of any commer­cial application of this design.

The con; :$ of this new edition of the nuclear volume of Modern Power Station Practice reflect the considerable expansion and development of nuclear power generation in the UK since the initial volume was published in 1964. During that period the AGR programme has been completed, the approval of Sizewell В launches a new generation of PWR stations and the first of the Magnox has completed its useful life. Thus the current volume presents a comprehensive picture of the design, development and operation of the majority of the nuclear station designs currently being operated throughout the world.

In rewriting the volume, the original format has been completely changed to accom­modate the now compendious technical nature of the subject with four independent but complementary chapters, Nuclear physics and basic technology; Nuclear station design; Nuclear station operation and Nuclear safety. Although each chapter is complete within itself, some overlap of technical matter between the chapters is inevitable and indeed essen­tial, reflecting the co-operation of widely differing technical disciplines necessary to ensure the safe and economic design and operation of nuclear stations.

In order to keep the size of the volume to reasonable proportions, consideration of plant and procedures which are common to both nuclear and fossil-fired stations have been omitted. For instance those aspects of once-through boilers which are not specific to AGR or PWR designs have been confined to Volume B.

The presentation of such a wide-ranging review of nuclear power has been made pos­sible only by the expertise provided by some fifty-one authors, all practising engineers and scientists of national and international reputation from the CEGB and the National Nuclear Corporation. The editing of such a wide variety of expertise has been a formidable but enjoyable task and the resulting volume represents the distillation of many years of experience in the design and operation of nuclear plant which must be unrivalled both in the UK and overseas. I should like to record my thanks to all the authors, many of them colleagues from the CEGB, for their co-operation in providing the editorial material and in meeting the production deadlines. I should also like to thank the staff at the National Training Resources Unit of the CEGB, in particular Peter Reynolds without whose hard work in the production of the illustrations, tables and references, the publication of the volume would not have been possible.

P. B. Myerscough Consulting Editor — Volume J

The conditions for a good moderator of high energy transfer and high collision rate (Section 5.1 of this chapter) are combined in the slowing down power. This is defined as:

slowing down power = £ Nas — *

where 1$ is the macroscopic cross-section (see Section

2.4.2 of this chapter).

Values of slowing down power for three important moderating materials are given in Table 1.7 and in­dicate their relative effectiveness to moderate neutrons. However, a more significant term in judging a suitable moderator is as follows.

4.2 Moderating ratio

The moderating ratio incorporates the slowing down power and the third requirement given in Section

5.1 of this chapter, low neutron absorption. It is defined by:

w. slowing down power

Moderating ratio = —————————————

macroscopic absorption

cross-section

Study of Table 1.7 shows that, as one would expect, the most effective material for slowing down neu­trons is ordinary ‘light’ water, H2O, followed by heavy water, D2O, and then graphite. However, when the need to minimise neutron losses is included, as it is in the moderating ratio, the best material for use as a moderator is clearly heavy water, then graphite and thirdly light water — but see Sections 5.4 and

5.5 of this chapter.