Natural uranium consists for the most part of very weakly fissile U-238 and 0.7% of highly fissile U-235. In broad statistical terms, neutrons in the natural substance very largely encounter U-238 atoms which usually slow them down. When neutronic energies are reduced to between 0.1 and 1.0 eV they are captured by the pronounced resonance absorption bands [58] of U-238 nuclei, which together with surface leakage obstructs a self-sustaining chain reaction in the natural material.

Thermal[15] nuclear reactors achieve a self-sustaining reaction by embedding uranium oxide[16] fuel rods in a carefully contrived matrix of moderating material. Collisions with moderator nuclei reduce neutronic energies to below the U-238 absorption bands in a short distance after fission, so thereafter fission with very largely U-235 atoms takes place to create a self-sustaining reaction with 32 pJ of heat released per fission. Moderators are usually constructed from graphite (AGR), heavy water (CANDU), light water (PWR, BWR) or a composite of any two of these (RMBK, SGHWR). The fissile concentration (enrichment) is often increased radially to an average of around 3% in order to achieve a more uniform and therefore more cost-effective power production. Unlike heavy water, light water is both an effective moderator and absorber [58], so its conversion to less dense steam reduces both neutron moderation and absorption. By astute design of core-lattice geometry and fuel enrichment, light water reactors outside Russia have always been designed to become under-moderated with increasing steam production [61], and this policy is vindicated by the Chernobyl disaster [12]. Indeed a negative power reactivity coefficient is now a necessary prerequisite for licensing by European Regulatory Authorities.

A self-sustaining chain reaction is achieved in a fast reactor by ensuring that the average neutronic energy remains well above the absorption resonances of U-238. Inelastic scattering and parasitic absorption of fast neutrons are overcome by a typically 20% enrichment

3

with a mix of U-235 and Pu-239 oxides. The power density (W/m ) is consequently so large that individual fuel pins can have only small diameters (‘ 5 mm). To avoid a significant moderation of neutronic energies these pins are closely spaced in hexagonal subassemblies and cooled by liquid sodium. Though parasitic absorption in structures and fission products (e. g., Xe-135) is relatively less in fast reactors, damage mechanisms are clearly aggravated and careful choices of materials are necessary. For example, liquid sodium leaches out carbon from stainless steel, so this fuel cladding must be niobium stabilized. Despite higher fuel fabrication costs than a thermal reactor, research has doubled its in-service life and capital installation costs are somewhat offset by a smaller reactor core. Uranium for the Russian nuclear program and Skandia for rocket motor exhausts were found in a very remote region around Aktau. Power, desalinated water and fish (farmed sturgeon[17]) for the mining complex were provided during 1973-94 by the BN350 fast reactor: chosen perhaps for the easier transport of its relatively smaller core. The unique advantage of fast reactors resides in their better “neutron economy” which enables the production of more fissionable material than that consumed. For this purpose the core of highly enriched subassemblies is surrounded by breeding blankets of natural uranium to give

U-238 + 1 neutron! Pu-239 fissile (1.10)

If the existing stock of UK nuclear materials were to be used in this way, the estimated energy would be comparable with recoverable coal reserves [60]. However, with uranium now so plentiful, economics and strategy favor the construction of thermal reactors, whose choice is now considered.

The engineered slowing down of neutrons from around 2 MeV at fission to thermalization at 0.025 eV is almost entirely done by elastic collisions with moderator nuclei. Applying the law of conservation of momentum and assuming spherically symmetric scattering in a center of mass coordinates the mean logarithmic decrement j of neutronic energy per collision is given by [58,61]

j = 1 + — log a with a = (A — 1/A + 1)2 (1.11)

where A is the mass number of moderator nuclei. A logarithmic compression is appropriate by virtue of the wide energy range involved. Though an effective moderator has a large j, the probability of colliding with a nucleus must also be large. Hence, the slowing down power of moderator is defined as j^s where ^s is its macroscopic scattering

supplies the saturated steam component of its 2-phase output [62] to steam turbines. By allowing water to boil at the lower pressure of 7-2 MPa, a BWR needs neither steam generators nor a pressurizer. Though the lower operating pressure appears to further reduce initial capital cost by way of thinner pressure vessels, core dimensions must be increased for the lower linear fuel rating (W/m) which is necessary to prevent a damaging dryout transition from nucleate boiling [63,64]. Also as described in Chapter 3, steam in boiling subchannels can give rise to flow instability which is another potential source of fuel damage. Furthermore, power control in a BWR is patently complicated by interactions between internal steam volumes, coolant flow rate and the insertions of its cruciform control rods. Despite the additional costs of components and a thicker pressure vessel to inhibit nucleate boiling at 15.5 MPa, PWRs are more generally favored as they are cost-effective in avoiding the above design and other operational or safety issues. Specifically, in the rare event of fuel melting, the separate steam- generator units of a PWR provide an excellent heat sink24 and additional isolation between the reactor and its environment. Under these circum­stances, secondary-side steam vented to atmosphere in an accident

A residual heat removal heat exchanger is also provided.

nstrumentation
Lifting lug
Thermal sleeve
	Hold down spring
Alignment pm
Core barrel
column
Inlet nozzle
Access port
Radial support
Figure 1.4 A Pressurized Water Reactor [178]

“bleed and feed strategy” would be far less radioactive than that released from a BWR.

Public confidence in nuclear power was shattered by the reactor incidents at Three Mile Island [66] (1979) and Chernobyl [12] (1986). However, the former galvanized the start of globally intensive safety research,[18] as well as new stringent operating legislation and decom­missioning techniques [69]. In this respect the author’s own research at AEEW was abruptly shifted from intact plant control studies to investigations of explosive boiling and its potential damage to internal reactor structures. Others conducted theoretical and experimental stud­ies [67,68] into the impact of a jet fighter (Tornado) on a reinforced concrete reactor building, or a dropped flask of highly radioactive reactor fuel, or the detection before their critical length [96] of embrittlement cracks in PWR pressure vessels. By 1992 international research had confirmed the effectiveness of active accident control measures to mitigate the consequences of fuel melting in both fast and thermal reactors. This research still continues into the design of passive emergency cooling systems [108-110] based on natural circulation to avoid auxiliary power supplies.

While this book was being written, the northeast coast of Japan experienced horrendous earthquakes, tsunami and a major nuclear incident at the Fukushima BWR plants. Some brief personal comments here concerning its impact on nuclear safety and future construction appear apposite. National licensing of nuclear plant operations requires a demonstrable engineered resilience to local seismic activity. In this respect all Japanese plants escaped unscathed from the earthquakes in 2009 and 2010. Also just prior to the 2011 tsunami, an effective neutronic shutdown was effected on all the West Coast and East Coast Fukushima plants: despite the Richter-scale 9 quake. Media reports and pictures indicate that structural and emergency core-cooling systems failed at Fukushima as a result of swamping by the subsequent tsunami whose estimated 14 m height grossly exceeded the design limit of 5.3 m. Historic data on tsunamis is therefore less complete than for their earthquake precursors, and flood defenses on the East Coast clearly proved inadequate. With hindsight, Japanese nuclear stations should have been built on the West Coast which is sheltered by mainland China. It is considered here that the resulting opposition to nuclear power station construction in locations not threatened by tsunami or serious flooding is an over-reaction. Likewise, whilst inadequate UK planning has allowed homes to be built on flood plains, this is no reason to refuse their construction elswhere on suitable sites. Also it is widely believed that diagnostic X-rays are our only exposure to radiation and that any exposure materially damages our health. In fact cosmic rays and natural radioactive decay in the earth’s crust cause a continuous pandemic exposure[19] [20] and though citizens in the granite city of Aberdeen receive about three times the background exposure of Londoners, no statisti­cally significant increase in pertinent cancer cases occurs. Though I-131 ingressed into Tokyo’s drinking water from Fukushima it was around only 1/5th the UK’s safe limit for all ages, and the Japanese advocated consumption by adults only. Moreover, very conservative exclusion zones and contamination limits on farm produce were imposed. However, it was the admission of falsified plant safety reports [170] in February 2010 that created the material loss of public confidence.

During 1993 numerous early retirements of UKAEA professional staff left a minority to develop successful decommissioning and waste glassification technologies that are now deployed worldwide by AMEC plc [70]. At Winfrith the Zero Energy Facilities and the High Tempera­ture Dragon Reactor have now been properly decommissioned, and the buildings of the SGHWR demolished. However, water reactors them­selves take relatively longer due to their thicker corrosion deposits which hold up larger quantities of radionuclides. Nevertheless progress to date suggests that the entire site will be outside nuclear regulations during 2039-48. While the storage of glassified radioactive waste raises public concern in some quarters, the Swedish towns of Forsmark and Oskarshamn actually competed [320] for a high-level waste facility to be built in their respective neighborhoods [72]. Construction at Oskar­shamn was approved in 2009 with a start date in 2013, and on completion concreted waste in 25 tonne copper-sheathed stainless steel drums embedded in impervious Bentonite clay cushions will be buried in stable igneous rock tunnels. A further safeguard is that these containers are to be retrievable for inspection because the waste itself has potential industrial or medical applications. Unlike the environ­mental release at Bhopal of dioxin with an indeterminate active life, the radioactivity in nuclear waste reduces to that of mined ores in about 7000 years [321]. The natural fission reaction at Oklo some 1800 million years ago provides evidence that igneous rock formations alone can contain fission products for well over this period.

Since the Three Mile Island incident in 1979 the worldwide deployment [73] of 269 PWRs has operated at high-capacity factors [25] and without a major failure of the nuclear technology itself. A contributing factor is the shared experience within the PWR Operators Club that has led to safety-enhancing retrofits and procedures. It is therefore contended that PWRs offer a technically sound and safe solution to an impending electric power deficit. Vindication of this strategy is further offered by the willingness of populations around existing AGR sites to accept replacement PWRs: particularly when endorsed by the families of local plant staff. However, technical and safety issues alone are insufficient for renewing the UK nuclear power program in the now privatized electricity industry. Specifically, the huge capital investment must be largely met by private equity rather than as previously from public funds. In this context a commensurate return on shareholder funds must be incipiently visible, and towards this end an appropriate financial framework must be pre-established by the government and its regulator OFGEM. Some pertinent factors for consideration are now described.

The lifetime cost breakdown [74] for new PWR plant is shown in Table 1.7 which reveals the largely dominant construction cost. Build — times, and therefore costs, vary between 4 to 7 years depending on national working practices and the number of repeat orders. No revenue is evidently forthcoming during construction, but capacity factors once operational are as high [25] as 90% over a design life [74] of 60 years with insignificant carbon emissions and highly efficient land use (see Section 1.6). However, due to the initially high capital expenditure and delayed revenue, an estimated payback period of 30 years is required [52]. On the other hand as explained in Section 1.7, CCGT generation

is radically different and more favorable to private equity investment by virtue of its timely income and assured profit. Though nuclear appears to be the lowest cost source of low-carbon electricity generation [52], it has to compete under the present UK regulatory framework with the base-load costs of CCGT units, even though these are not compliant with the mandatory 2020 UK emission targets. Indeed the Chief Executive Officer of RWE Nuclear argues [75] that the government’s renewable obligations tariff should be changed to a low carbon obliga­tions tariff in order to fairly characterize the role of nuclear power. The financial similarity between wind and nuclear power investment clearly supports his argument. However, levelized costs for new-build nuclear are estimated [52] as $92 to $123 per MWh, which are well below[21] $246 to $308perMWh for offshore wind turbines: even before the cost of the necessary backup systems is included.

The first thermal reactor for commercial electricity was completed in 1956 at Calder Hall. Despite its description as “The Peaceful Use of Atomic Energy,” there remains public apprehension that commercial nuclear power stations are also sources of weapons material. In essence, nuclear weapons create an explosive growth of the neutron population in a mass of largely fissile material as an end in itself, or as the initiator for a fusion device. For the Hiroshima A-bomb an appropriate mass of uranium was highly enriched with U-235 to restrict parasitic absorp­tions by U-238. The Nagasaki weapon was designed around plutonium recovered from a specially contrived fuel cycle which ensured very high concentrations of fissile Pu-239 relative to that of the Pu-240 created by another neutron absorption. Because the higher mass isotope is an unstable a-emitter [76], a sufficiently high concentration would induce the partial triggering of a plutonium-based weapon. Accordingly weapons-grade plutonium has a specified Pu-240 concentration of less than 7%. During the in-service life of thermal reactor fuel, fission of created Pu-239 forms a partial and immediate replacement for “consumed” U-235 atoms, but burn-up of Pu-240 proceeds at a slower rate. Consequently the relative concentration of Pu-240 increases with increasing fuel burnup. Fuel pins for power reactors are precisely engineered fabrications that embody years of research to enhance safety and the economy of electricity generation. Burn-up targets[22]

for commercial power reactors have always been determined by these considerations so that Pu-240 concentrations in recovered plutonium are too high for weapons purposes. Nuclear power generation has been and is therefore still divorced from nuclear weapons.