It is hoped that readers have concluded that the design and safety of nuclear power plants are soundly established. In this respect, their antagonists sometimes overlook that researchers and operators live in reactor neighborhoods with wives and families.

Spokespersons for non-nuclear organizations frequently assure us that “lessons have been learned:” yet the same misadventures reoccur. This is not the case with the nuclear industry in that the Three Mile Island and Chernobyl [12] incidents are shown herein to have had a tangible impact. Indeed the latter provoked the shutdown of all Russian — designed RMBK reactors in Germany, and also initiated the interna­tional requirement for water reactors to have negative reactivity void coefficients. Prior to the earlier US incident, Severe Accidents with fuel melting were regarded as hypothetical or just imaginable. Thereafter internationally coordinated research began into their phenomenological dynamics, and legislation demanded a hierarchical operational structure based on technical qualifications, simulator training and plant experience [59,65,91]. In this respect, note that human error aggravated or precipi­tated both incidents. It is also contended herein that the Fukushima incident resulted from a site-planning error rather than from a flawed nuclear technology. Because these reactors lay on the unsheltered East Coast, the enormous tidal surges were able to swamp the emer­gency power supplies for the ECCS. If they had been sited on the West Coast, as some other Japanese plants, they would have been shielded from tsunamis and their successfully initiated neutronic shut­downs by scram rods likewise maintained in spite of the Richter-scale 9 earthquake.

Renewable energy sources, especially windpower, were incipiently greeted with public approbation for their perceived potential to decel­erate global climate change. Indeed a UK government ratified an international agreement for particularly low national carbon emissions believing that renewable energy sources would deliver its promise. Now however the media and an often vociferous public raise concern over renewables’ green credentials and their ability to provide a secure UK electricity supply. If the proposed 18 GW of wind power are lost by not uncommon nationwide high-pressure weather, then despite the pro­posed construction of a European Supergrid [38,39,43], it is argued that the necessary back-up would not always be available. Specifically, Northern Europe has the heaviest industrialization with peak demands in the same winter solstice as the United Kingdom, and their existing aged stations lack the requisite margins. Moreover, Germany’s intention to shut down all its nuclear plants by 2020 clearly exacerbates compe­tition for any available power by the privatized utilities: even if it were to be available. Finally, wind turbines annually produce only some 20% of their rated outputs [36] which intensifies this competition, worsens their economics and necessitates reliable back-up by gas-fired or nuclear stations whose capacity factors are between 80 and 90%.

Actual or potential environmental impacts have been identified for all commercially viable renewable energy sources. For instance the Isle of Thanet wind farm of a nominal 300 MW has a visual impact of over some 3500 ha,[117] whereas the relatively unobtrusive SGHWR generated 100 MWe with over 60% reliability on a total of just 31/2 ha which included car parking areas. The “largest whirlpool in the world” is created when sluices of the La Rance tidal barrage open to capture a rising tide. Rather than being viewed in terms of estuarial damage, it has become a tourist attraction and shortly after construction EDF quietly switched to nuclear power. To place the heroic loss of emergency workers’ lives at Chernobyl in perspective, the catastrophic failure of the Banqiao hydroelectric dam in 1975 resulted in 171,000 fatalities [11], whilst a multi-national IAEA investigation [13] concluded that no subsequent medical conditions could be directly attributed to this crassly initiated and managed nuclear incident. Radioactive releases at Three Mile Island have been calculated to induce just an additional one or two thyroid cancer presentations during the following 20 years, [66] and there were no direct fatalities.

The necessary back-up for renewable generation can be provided only by fossil or nuclear stations. Until carbon capture [56] becomes viable, neither coal nor partially fossilized lignite is likely to be deployed as fuel. Over the past 2 years, there has been a marked shift in the prospects for gas as a result of the enormous quantities of shale gas made available by fracking. Though exploration in the northwest of the United Kingdom has apparently identified commercial quantities, its large-scale exploitation is under review due to concerns over minor earth tremors and possible aquifer pollution. These anxieties appear less relevant in the far less densely populated United States, which has become a potential major exporter of liquefied gas rather than a previous importer. Since 1992 the high thermal efficiencies (‘50%) and favorable cash flows of combined cycle gas turbine (CCGT) plants have resulted in their increasing UK deployment [51,52]. However, a KPMG report [52] concludes that the supply of UK electricity by a continuing investment in CCGT plants alone could not meet government obliga­tions on carbon emissions. In addition, the country is not self-sufficient in gas and so is vulnerable to external political unrest. On this basis nuclear power is argued herein to be a necessary component in a “mix” to guarantee the United Kingdom a secure and reliable electricity supply within its emissions obligations.

As well as largely predictable daily and seasonal changes, there are rapid unpredictable power variations on a Grid network such as the start-up of a 1 MW main-line locomotive or the disconnection of a 1000 MW station due to lightning, a bird strike or component failures [80]. Transient differences between instantaneous power generation and that consumed are shown to create network-wide common frequency fluctuations about the nominal UK value of 50 Hz, and for the explained technical and statutory reasons these fluctuations must be within ± 0.5 Hz. Due to operational rate constraints rapid unexpected changes in power demand cannot be met by modulating the primary energy sources, but only from the thermal energy stored in components of the load-following (Coupled) stations and the rotations of all synchronized motors and generators [80,117]. In this respect gas-fired CCGT stations, preferably with steam drums, are more flexible than PWR and BWR designs whose slower power changes are imposed to achieve economic fuel cycles[118] and the intervention of safety systems. Accordingly, nuclear stations operate in the decoupled mode to supply the more slowly varying and largely predictable base load: as consistent with their larger initial capital costs. The UK mix of CCGT, nuclear and wind power generation as outlined by the government on May 20, 2012, hinges at the time of this writing on consumer price guarantees to attract private equity investment.

An atmospheric release as aerosols of radioactive Iodides and Caesium in the size range 1 to 5 mm is identified as the principal hazard in nuclear power production. However, the dispersed mass would be markedly reduced by dissolution in the large quantities of reactor coolant present [104]. For instance just some 16 curies of the 3 to 5 million in the Three Mile Island reactor were released due to absorption in the water or vapor present in the containment building and its sump [66]. Since then enhanced safety systems, statutory operating protocols and pre-planned public evacuation schemes have been put into place [59,65,91]. It is also required that the aggregate probability of all Severe Accidents with or without an atmospheric release must be no greater than 10“7 per operating year throughout a plant’s lifetime. Reactor safety assessments such as Farmer’s [157] specify a particular station to be safe if and only if the statistically expected increase in cancers from all Severe Accidents during its operational life is orders of magnitude less than from natural causes. Radiological dose rates in such assessments involve variations in local population density and wind-direction statistics. However, they over-predict fatalities because

i. Induced cancers from a radiation dose are calculated from linear extrapolations of data from Japanese A-bomb survivors. Consequently there is no allowance for the natural repair mechanisms of the human body that become effective at much lower doses [66]. For example, the background radiation in the granite city of Aberdeen is three times that in London: yet there is no statistically significant increase in comparable cancers.

ii. Some 1770 cases per million of natural thyroid cancers present annually in the United Kingdom of which 80 to 90% are successfully treated by surgery [163,164].

It is also contended that safety assessments are biased against nuclear power because

a. Risks are assessed without attendant benefits: like some medical vaccinations [169].

b. Humans peculiarly accept much larger self-imposed risks than risks externally imposed, and a cancer risk is to be avoided no matter how slender. Specifically nuclear power is often rejected despite the orders of magnitude greater fatal risks from natural cancers, road traffic accidents and tobacco smoking (see Sec­tion 4.3).

Public concern remains over the storage of nuclear waste. Reference [320] details its classification by activity levels and the appropriate storage technologies. Unlike the atmospheric release of dioxin at Bhopal with an indeterminate active life, radioactivity in nuclear waste reduces to that of mined ores after about 7000 years [320]. To obstruct any environmental release during and beyond this period, a well-researched multi-barrier technology has been developed for high level (highly active) waste. After an initial storage in “ponds” to reduce decay heat the contents of intact fuel pins are glassified before being concreted within copper-clad stainless steel drums. Finally these are embedded in Bentonite clay before being stored in deep igneous rock formations that allow their subsequent inspection or retrieval. The natural fission reactor at Oklo [321] about 1800 million years ago demonstrates that impervious igneous rock strata alone can retain fission products for well over the required time span. With regard to public opinion, 80% of the populations in Sweden’s Forsmark and Oskarsham towns voted in favor of local storage facilities in their neighboring igneous rock tunnels [32].

Commercial nuclear power generation is frequently wrongly asso­ciated with offensive weapons due to the military antecedents of nuclear fission. In fact the economic viability of nuclear power necessitates the fission of just so much U-235 (fuel burn-up) that fissile Pu-239 transmuted from U-238 becomes itself transmuted into the «-emitter Pu-240 in concentrations greater than 7%. Consequently the inseparable plutonium mixture is then unsuitable for weapons. Indeed nuclear power was described in the pioneering days of Calder Hall as “The Peaceful Use of Atomic Energy”.

Power generation by thermal reactors involves less complicated (expensive) engineering than by fast reactors, which for example require intermediate heat exchangers to isolate further the primary circuit’s sodium from steam generators. There is presently no foresee­able shortage of mined uranium, so the fertile-to-fissile fuel-breeding feature of fast reactors has neither strategic nor immediate economic benefit. Accordingly thermal reactors have become the worldwide choice. Section 1.8 shows that water reactors are intrinsically the most compact designs, and because 60 to 70% of a nuclear station’s capital costs reside in civil engineering [74], they are also more cost — effective than grosser AGRs. Though BWRs need neither separate steam generators nor pressurizers, Section 2.3 shows that boiling channels can be unstable, so this apparent advantage is eroded by their lower linear fuel ratings that are necessary to present fuel damage by burn-out [63,64,297]. Other relative disadvantages of BWRs are iden­tified so that PWRs have become the more favoured worldwide choice. Finally, SGHWRs require enrichment of both fuel and moderator (deuterium) with the former necessary to achieve a negative reactivity void coefficient [61]. Consequently, economics have led to their discontinuation despite their reliability and safety having been proved over 1966-90.

[1] See Chapter 4 for the nuclear power plant case. For the Banqiao Dam, the probability of a storm created overflow was assessed [11] as 0.001 p. a., so it was considered safe for 1000 years.

[2] Experience indicates that semiconductors are most likely to fail in a short period after fabrication; hence a manufacturer’s “burnin”.

[3] Fossil-fired stations rank in a merit order corresponding to their annually averaged generation costs per kWh.

[4] Wave devices so far appear unable to meet the sporadic violent storms in the United Kingdom and Australia.

[5] Two 40-minute periods per day around Alderney.

[6] About 80% national capacity.

[7] Output steam density for a PWR boiler is some two decades larger.

[8] Apart from its scheduled annual overall and an insignificant number of short duration trips, its actual capacity factor was 60%.

[9] By using forced draught cooling towers, for example.

[10] With ac transmission, the above P is the average power per phase per cycle (i. e., in one conductor). However, three-phase cables with a constant instantaneous power compare even less favorably as the peak voltage between phases is 3V.

“After commissioning Isle of Thanet, 2010.

The stochastic meanderings of high and low barometric pressure zones across the planet are patently beyond human control. Also a high-pressure region, in which there is little or no wind, can blanket a sizeable portion of a European country for as long as a week thereby disrupting electric power generation. Correspondingly reduced annual capacity factors and installed wind powers [36,37] are illustrated in Table 1.4 for 2005-07.

Though large countries like the United States and Russia can “hedge” by spatially distributing their wind farms, smaller Northern European nations can face major disruptions which are especially critical during their peak winter demand periods. Accordingly, if wind power is to enable a significant reduction in European carbon emissions, a solution to its intrinsic intermittency must be found in order to preserve the security of national power supplies. Due to the withdrawal of govern­ment support from the Severn Barrage scheme, the remaining option for a materially sized renewable UK energy resource is wind power, which is potentially required to be the largest in Europe [42].

Though electrochemical batteries or diesel generators are reliable backups for isolated low-power applications, they are totally nonviable for sizeable offshore wind farms in a national Grid. For instance wind power developments in the United Kingdom have installed, either in construction or in planning a rated 18 GW for operation by 2020. Because this power equates to the output of about 18 large fossil or nuclear stations, radical measures are necessary to secure the national power supply. Toward this end a memorandum of understanding [38,39] was signed in 2009 for the creation of a European Supergrid network at an estimated cost of D30,000M, but construction plans remained under discussion in 2012. In particular, the French-UK ac connector of 2 GW is to be supplemented by dc links from both Norway

[11] To access existing grid connections between Germany, France, Belgium, etc.

[12] See Chapter 3.

[13] Refer to the Basel CH experience in Section 1.2.

[14] See Figure 1.1.

[15] So-called because fissions largely occur at neutron energies around the thermal vibrational energies of the fuel molecules.

[16] Allows a higher fuel temperature than the metal and a longer life.

[17] Using the warm condenser outflow into the Caspian Sea.

[18] Coordinated by the IAEA of the United Nations Organization.

[19] About 100 mrem = 1 mSV.

[20] Half-life 8 days.

[21] At$1 = £0.65.

[22] Presently over 60MWd per kg of U in pristine PWR fuel — Ref. [77].

[23] See Ref. [108] for details of European Utility requirements.

[24] Section 1.8 contends that Fukushima is a site-planning error.

Nuclear Electric Power: Safety, Operation, and Control Aspects, First Edition. J. Brian Knowles.

© 2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.

[27] The classical SISO system notion that transfer function zeros are the roots of its scalar numerator polynomial is in fact equivalent to definition (2.12).

[28] Because compensating networks should move them further still to the left.

[29] A linear partial differential equation over all t > 0 has a countably infinite set of eigenvalues, whose eigenvectors are linearly independent [110-111].

[30] Control rods or a start-up source, for example.

[31] Neutronic energies quantized to replicate their slowing down, etc.

[32] Though Table 2.2 lists just five, a sixth results from the interaction of g-rays with heavy water in an SGHWR or CANDU reactor.

[33] Doppler broadening alone affected restabilization of a growing neutron population in the University Argonaut “zero power” reactor at Risley.

[34] Fuel enrichments in thermal and fast reactors are some 3 and 20%, respectively. The inadvertent insertion of a fuel rather than a U-238 breeder pin in a fast reactor could cause a potentially serious neutron excursion, so extreme care is necessary in their refuelling.

[35] Similar situations occur in domestic hot water cisterns and in salt fingers around river estuaries.

[36] Some 2 MW in a 1000 MW(e) station.

[37] Allowances for in-service erosion and tube-to-tube variations.

[38] Some aircraft systems use 400 Hz to reduce core size (i. e., weight), but hysteresis losses are correspondingly greater [35].

[39] Electric intensity is inversely proportional to conductor diameter, so each phase of an overhead supergrid line in the United Kingdom is a tightly bunched bundle of four conductors.

[40] Allowable variations in the axial temperature profile of the SGHWR steam turbine corresponded to between 2 and 3 MW per min.

[41] Used as a means of Network power factor correction via their dc excitation [35].

[42] See Figure 3.9.

[43] For a coal-fired plant.

[44] Coal combustion is initially endothermic.

[45] An electrical machine carries a nameplate stating its (nominal) safe continuous rating, e. g., 33 KV; 500MVA.

[46] Refer to the fast reactor drum water-level control by Hughes in Section 3.1.

[47] If an increase in the scalar controller gain causes instability, the system is termed Conditionally Stable.

[48] Refer to Section 4.1

[49] CFR was never built, as attention moved to a European design (EFR) which also never materialized.

[50] A simulation of Jointed Construction for Boiler and Rig Kinetics, which was subsequently extended to include reactor, steam drum, feed train, and steam turbine modules. It was developed by Drs. A. Robins, D. Farrier and the author several years earlier.

[51] Given an electrically powered feed pump.

[52] This viewpoint is adopted in Reference [57].

[53] An aircraft impact on the containment is one exception; see Missile Studies Section 6.4.

[54] A double-offset shear-break of a cold leg of a PWR with all emergency cooling systems functional is one example of such a limiting event. US federal regulations require statements of modelling assumptions and risk assessment criteria.

[55] All exposed to a radiation hazard wear an obligatory film badge or monitor while on duty.

[56] Current evidence suggests that caesium iodides are the forms present in a fuel element, but their stability on release depends on temperature and the reducing or oxidizing character of the environment [104].

[57] Principally the b and g emitter I-131 with a half-life of about 8 days.

[58] Discoverer of the poliomyelitis vaccine.

[59] E. g., 0.1 ± 0.001 indicates a very well-understood phenomenon.

[60] 1 curie = 3.7 x 1010 Bq; 1 Sv = 100 rem and 1 man-rem is an individual dose of 1 rem.

[61] Clearly dependent on location.

[62] Very conservative; see statement (4-2-4).

[63] This is evidence of our body’s self-repair mechanisms, and the pessimism of a linear extrapolation of A-bomb casualty statistics.

[64] A natural convective heat exchanger with a NaK primary side and an atmospheric secondary side.

[65] Satisfactory reactor siting, as described in Section 4.4, involves local population density, weather patterns, and seismic activity.

[66] Neutron absorptions create Co-60, which is a penetrating g-emitter with a half-life of 5 years.

[67] As Low As Reasonably Practical.

[68] Earlier PWRs like Three Mile Island have just two.

[69] Calculations indicate no less than 1 hour, so there is time for emergency reconfigurations of a plant [65,93] or even attaching civilian fire pumps. For the fast reactor situation, refer to Reference 213.

[70] By virtue of the complex physical processes and uncertainties about physical properties, Severe Accident calculations are required to be only reasonably conservative, rather than the ±10% for normal engineering design.

[71] For example, civilian fire pumps [65].

[72] A detonation (rather than a deflagration) is characterized by a pressure wave spatially in front of a reaction [203]. ”Pinking” in petrol engines is a well-known example of an unwanted detonation rather than a deflagration.

[73] By increasing surface contact areas and decreasing heat diffusion distances.

[74] Refer to Section 5.6.

[75] Hereafter simply “coarse mixture” denotes the detonatable portion of melt.

[76] Thermal radiation flux is proportional to the 4th power of absolute temperature.

[77] Established by repetitions of this recovery procedure. Also individual debris particles are microscopically smooth with a water coolant, but they are pitted in the case of a sodium coolant, which suggests a chemical attack.

[78] See Section 5.5.

[79] Thermal diffusivities of liquid sodium and water are 50 x 10~6 and 0.14 x 10~6m2/s respectively.

[80] That is, without a shock wave.

[81] See Section 5.4.

[82] In a water reactor.

[83] See Ref. [206].

[84] Equation 4-5 in Ref. [219], which assumes the local speed of light is that in vacuo [218].

[85] — wavelength; r — electrical resistivity of the conductor m — magnetic permeability; c — speed of light in the conductor

In the case of liquid sodium, equation (5.24) reduces to

lab ‘ 13лА mm

which shows that a radiated heat flux entering the liquid is effectively absorbed at the actual interface. Similar to equation (5.20) the thermal radiation flux entering the interfacial liquid sodium [219] is derived as

(5.25)

Due to surface contamination the total emissivity of a sodium surface in Severe Accident conditions is certainly larger than the value measured [228] under clinically clean laboratory conditions as

el = 0.05

The total heat flux entering and absorbed by the interfacial liquid sodium with respect to the coordinate system in Figure 5.4 is therefore

[86] The Lax-Wendroff solution [238] scheme introduces artificial viscosity and thermal conductivity to achieve numerical stability.

[87] Thermophysical properties of molten urania vary markedly with temperature [245].

[88] For a source TM ^ TL; h ‘ aTM where a is the Stefan-Boltzmann constant.

[89] About 60 ms for a complete discharge from the MFTF thermite container.

[90] An unknown, but see later.

[91] Forty percent of fine fragmentation is found in some CORECT2 experiments [86], but this is more likely to result from multiple incoherent interactions and/or the debris recovery process from a sodium coolant.

[92] Refer to Section 5.2. Portions of a melt solidify or fail to develop the detonatable morphology.

[93] The ratio £(.D3)/£(D2) is often termed the Sauter diameter [308].

[94] The embedded BURST subroutine was developed by Dr. Kier at UKAEA, Culcheth.

[95] Hence the use of deaerators in boiler feed trains.

Nuclear Electric Power: Safety, Operation, and Control Aspects, First Edition. J. Brian Knowles.

© 2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.

[97] An exceptional use of sodium is described in Ref. [271].

[98] A three-dimensional calculation is crucial for a reinforced concrete impact [275].

[99] Visualize shock propagation in terms of a cascade of lossy spring-linked point masses.

[100] At speeds as high as 100 m/s (216 mph).

where F is a homogeneous function. Based on the above, Buckingham’s Pi-Theorem asserts that

“The physical measure Pi of a quantity which depends only on other quantities with measures P2, P3,… Pn reduces to one with the minimum number m of their non-dimensional combinations.”

Its interpretation in physical terms is that two realizations of a process with the same complete set of Pi-terms have the same dynamic behavior.

Though a systematic procedure [280] exists for deriving the dimen­sionless Pi-terms of a physical process, those describing the interaction between a missile and a structure can be determined by physical insight when there is a negligible temperature rise in the latter from its plastic deformation. Table 6.2 lists the determining variables under this condi­tion, and the pertinent dimensionless terms can be deduced by adopting missile diameter, target density and target strength as the reference parameters. These can then be combined with impact velocity to provide the additional Pi-terms to characterize the target response as

F = (VyfpJS; h/d; L/d; r/d; w/d; a; b; Pm/Pt; s/S, e) = 0 (6.3)

[102] Different materials from a prototype are sometimes used: see Ref. [297], for example.

[103] Same Pi-terms.

[104] Percent EWEF — % of a cross-sectional area occupied by the same square (EW) steel mesh reinforcement just below the surface of each panel face.

Nuclear Electric Power: Safety, Operation, and Control Aspects, First Edition. J. Brian Knowles.

© 2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.

[106] Reactor and boiler channels are routinely cleaned chemically during regular maintenance to minimize pumping power so as to promote efficiency.

[107] Output demand rather than reactor power to provide anticipatory control; see Section 3.1.

[108] A UK terminology when observed in the SGHWR around 1967 [301].

[109] See Refs. [63,64,117,297]. Burn-out can also occur as a result of nodules of higher enrichment appearing in a later manufactured batch of lower enrichment [301] fuel, or if porous magnetite deposits (crud) on fuel elements have their capillaries blocked by copper salts released from steam condensers [95]. After Dr. G. R. Hewitt identified the mechanism at AERE Harwell, it became known as dry-out, which is a more graphic description than CHF or burn-out for boilers.

[110] Temperature-induced density changes in water increase with decreasing pressure [208,209].

[111] Some PWRs have their lower head submerged in water to preserve its creep strength [315].

[112] See Section 5.8 regarding heat transfer with fine droplets.

[113] No “intelligent” signal inputs (i. e., activation and operation within the system itself).

2. Neither external power supplies nor forces required.

3. No moving mechanical parts.

4. No moving working fluids.

[114] Amounts of corium were also found in two steam generators and the pressurizer of TMI-2 [69].

[115] Higher porosities might have resulted from “Hanging fuel assemblies and control rods strewn about like pick-up sticks on top of a bed of rubble;” see Ref. [69] for the video image.

[116] Noble gas, iodine, and cesium isotopes vaporize out and so are excluded in this calculation.

[117] 1 ha approximates to 1 football pitch.

[118] By restricting differential thermal expansion between fuel pellets and clad.