Gestation: The Programme to Explode a British Atomic Device, 1947–52

  • John Simpson


The formal political decision of the Gen 163 Cabinet committee in January 1947, that design work should start on a British nuclear bomb, left most of the technical issues involved in its construction to be resolved by those managing the project. Their immediate aim was to build and test a fission device which would demonstrate both mastery of the nuclear physics principles applicable to atomic explosions, and an ability successfully to design, develop and manufacture the non-nuclear components necessary to trigger it off. This involved ordnance experiments which ‘could be begun and completed without the need to use fissile material at any stage’.1 In parallel, work was to be initiated on a production design for the bomb’s nuclear explosive warhead, which would offer the same advantages of allowing the fissile core to be stored separately from the fully assembled non-nuclear components that had been incorporated into the United States Mark IV weapon. This activity had to be integrated with the design, testing and production of the fuses, ballistic case and other components of the complete nuclear weapon system. The bomb had many similarities to a ballistic missile, as it was intended to be dropped from several miles above the earth, though it had no direct means of propulsion.2


Nuclear Weapon Atomic Bomb Fissile Material Ballistic Missile Uranium Hexafluoride 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Notes and References

  1. 1.
    Gowing, Independence and Deterrence, vol. 1, p. 180.Google Scholar
  2. 2.
    In the absence of detailed information on the first bomb design, it must be assumed that only part of the aerodynamic casing was used to carry the warhead, the rest containing fusing and other systems, or being empty. Pictures of the 24-foot weapon suggest that it was somewhat similar to a German V2 rocket in shape. Given the height at which it was intended to be released, its speed as it approached the ground must have been substantial. See A. Brookes V-Force (Jane’s 1982).Google Scholar
  3. 3.
    Gowing, Independence and Deterrence, vol. 1, p. 171.Google Scholar
  4. 4.
  5. 5.
    Ibid., p. 189.Google Scholar
  6. 6.
    It is unclear from the official history what were the origins of the erroneous information on the Hanford reactors. Although the figures for thermal power were probably based on the original design rating for these plants, the output calculations appear to have contained errors which were not discovered until after the arrest of Klaus Fuchs in 1950. See Chapter 2, footnote 74 and ibid., vol. 2, p. 401.Google Scholar
  7. 7.
    Ibid., vol. 1, pp. 190–1. The site had been a Royal Ordnance Factory and was known as Sellafield. To avoid confusion with another nuclear site at Springfields, the name was changed to Windscale. In 1981 its name officially reverted to Sellafield.Google Scholar
  8. 8.
    Ibid., pp. 192–3.Google Scholar
  9. 9.
    Ibid., vol. 2, pp. 391–3. In 1957, the inadequate control system resulted in some graphite channels in a Windscale reactor catching fire and distorting, producing an inability to discharge some of the fuel and the latter then burning.Google Scholar
  10. 10.
    Ibid., pp. 394–5.Google Scholar
  11. 11.
    Ibid., p. 399.Google Scholar
  12. 12.
    Ibid., p. 400.Google Scholar
  13. 13.
    Ibid., pp. 347 and 401.Google Scholar
  14. 14.
    Ibid., p. 389. In practice, enrichment back to normal levels did not result in material identical to natural uranium, as some non-fissile U-236 was created during reactor operations, and thus the percentage of this material present in the resultant uranium was far greater than before.Google Scholar
  15. 15.
    The basis for these figures is that originally the reactors had a combined output of 180 Mw(th) which was later increased to 240 Mw(th). The figures assume that they operated for 314 days per year and produced 800 grams of plutonium per 1000 megawatt days. For a discussion on the source of the 800 gram figure see Appendix 3. There is also a suggestion in ibid., vol. 1, pp. 445–6 that in 1952 the output potential of the second, more efficient, Windscale reactor was 40 kilograms of plutonium per year, giving a combined output of less than 80 kilograms.Google Scholar
  16. 16.
    The figure for the amount of plutonium required for the Hurricane assembly and the Blue Danube weapon is derived from Gowing, ibid., vol. 2, pp. 347–8. The original 1946–8 calculations (ibid., vol. 1, pp. 167–8 and 217) were based on 100 kilograms of plutonium producing 15 bombs, i.e. 6.67 kilograms per weapon. For the Hurricane test, 15 per cent less than this is reported to have been requested, i.e. 5.67 kilograms, though the actual core may not have contained as much fissile material as this.Google Scholar
  17. 17.
    Hewlett and Anderson, op. cit., p. 630.Google Scholar
  18. 18.
    Gowing, Independence and Deterrence, vol. 2, pp. 405–22.Google Scholar
  19. 19.
    Ibid, vol. 1, pp. 186–7 and 214 andGoogle Scholar
  20. S. Menaul, Countdown (Hale, 1980) p. 33.Google Scholar
  21. 20.
    This work formed part of the background to a report on ‘future developments in weapons and methods of war’ submitted to the Chiefs of Staff Committee in July 1946. See The Times, 15 June 1981, p. 2.Google Scholar
  22. 21.
    Gowing, Independence and Deterrence, vol. 1, p. 216.Google Scholar
  23. 22.
    D. A. Rosenberg, op. cit., pp. 67–8.Google Scholar
  24. 23.
    In 1949 a committee of the US Joint Chiefs of Staff conducted a planning exercise into the possible course of a world war breaking out in 1957, code named Dropshot. This assumed that 7 British bomber groups comprising 210 aircraft would be available to the NATO allies, as against 19 from the United States. The implication of the figures for atomic bomb and conventional bombing aircraft is that the UK force would carry some atomic weapons. Cave-Brown, op. cit., pp. 201 and 289–90. This planning figure of 210 aircraft may not be unconnected with the target figure of 200 weapons by 1957.Google Scholar
  25. 24.
    Gowing, Independence and Deterrence, vol.1, pp. 215–17.Google Scholar
  26. 25.
    This table is based on the magnitudes suggested in ibid., p. 217 and figures given in ibid., pp. 167–8 for the output of the Hanford reactor. They suggest that it was anticipated that approximately 6.67 kilograms of plutonium would be required for each UK bomb. In practice, both output figures for the reactors and the amount needed for each weapon were overstated - see supra, refs 15 and 16.Google Scholar
  27. 26.
    Ibid., pp. 217–19 and 223.Google Scholar
  28. 27.
    Ibid., vol.2, pp. 426–41.Google Scholar
  29. 28.
    Ibid., p. 427.Google Scholar
  30. 29.
    Ibid., vol. 1, pp. 440–3, but more especially Brookes op. cit. pp. 36–9.Google Scholar
  31. 30.
    Gowing, ibid., vol. 2, p. 295.Google Scholar
  32. 31.
    A fast breeder is a reactor which utilises neutrons moving at both high and lower ‘thermal’ speeds. By building the core of the reactor from enriched uranium or plutonium, the neutrons from a fission reaction are used more effectively, a much smaller reactor is possible, and the moderator can be dispensed with. In addition, the core can be surrounded with a blanket of U-238, a waste product of the U-235 enrichment process, and neutrons generated by the chain reaction in the reactor core will convert some of this material into plutonium. In theory this could lead to more plutonium being created in the blanket than is consumed in the reactor core, hence the name fast breeder reactor. If the reactor is not surrounded by uranium it becomes in effect a method of burning up plutonium or U-235 only, and is then termed a fast reactor or fast burner. Existing power reactors use neutrons travelling at thermal speeds, and have moderators to slow down the speed of neutrons and increase their chances of producing fission in a U-235 atom. It is because of this that such reactors are often called thermal ones.Google Scholar
  33. 32.
    Gowing, Independence and Deterrence, vol. 2, pp. 274–6.Google Scholar
  34. 33.
    Ibid., pp. 265–90.Google Scholar
  35. 34.
    Ibid., p. 290 and vol. 1, pp. 445–6.Google Scholar
  36. 35.
    Ibid., vol. 2, p. 295.Google Scholar
  37. 36.
    Ibid., vol.1, pp. 445–7.Google Scholar
  38. 37.
    For a detailed recent discussion of this problem, see A. De Volpi, op. cit., Appendix L, pp. 297–327. This suggests that the presence of PU-240 complicates weapon design because it: (i) leads to the need to increase the mass and radius of the fissile core of the weapon; (ii) introduces extraneous and unpredictable sources of radiation and heat into the weapon assembly. This can lead to reductions in the yield of the weapon, the possibility of it being physically disabled and problems in storing the completed assembly. The possibility of physical disablement arises from the fact that plutonium metal changes its characteristics if it is heated. It is in its most dense state, and hence has a minimum critical mass, below 115°C (the alpha phase). If its temperature rises above this figure it expands, creating not only a need for more material to form the critical mass but also changes in physical configuration. Above 115°C, plutonium is stable in the delta phase (310–450°C): above and below these figures it shrinks. This creates two options for weapon designers: (i) minimise the amount of plutonium used by maintaining its temperature prior to detonation below 115°C and thus use only plutonium containing a minimum of PU-240; (ii) stabilise the plutonium in the delta phase by adding gallium, and accept the penalties of a greater requirement for plutonium and the impact of having very hot metal in the core of the weapon, with subsequent effects on its design. Such a weapon could, however, use cores containing significant quantities of isotopes of plutonium other than PU-239 (p. 62). Gowing indicates that the British weapons were stabilised in the delta phase (ibid., vol. 2, footnote p. 467), and thus the problems they faced were physical shrinkage if the temperature boundaries were crossed, and the impact of the heat generated by the core upon the rest of the weapon.Google Scholar
  39. 38.
    Gowing, ibid., vol. 1, p.448 andGoogle Scholar
  40. R. F. Pocock, Nuclear Power (Unwin, 1977) pp. 30–3.Google Scholar
  41. 39.
    Gowing, ibid., pp. 448–9 and vol. 2, p. 291.Google Scholar
  42. 40.
    These figures are based upon the calculations in supra, ref 16, which indicate that 25 Hurricane size cores could be manufactured from 140 kilograms of plutonium, and the data in Table 3.1 which indicates that the output of the high enrichment plant would add 20 to 21 cores to the annual output figure.Google Scholar
  43. 41.
    Gowing, Independence and Deterrence, vol. 2, pp. 462–70.Google Scholar
  44. 42.
    For an account of this test see ibid., pp. 476–95.Google Scholar
  45. 43.
    Gowing, ibid., vol. 2, pp. 472–4 fails to make it clear that the Hurricane device was similar to the US Mark in ‘laboratory weapon’ of 1945. It possessed all its operational limitations, although it benefited from advances in electronic and explosive research over the intervening seven years, and its nuclear efficiency might have been better. The Blue Danube production weapon, however, was similar to the US Mark IV assembly of 1948 and possessed many of its operational advantages over the Mark III.Google Scholar
  46. 44.
    Ibid., pp. 461–73.Google Scholar
  47. 45.
    Ibid., p. 473 and Hewlett and Duncan, op. cit., p. 673.Google Scholar
  48. 46.
    Jackson, op. cit., p. 20.Google Scholar
  49. 47.
    Gowing, Independence and Deterrence, vol. 2, pp. 448–52.Google Scholar
  50. 48.
    Ibid., p. 452 and vol. 1, p. 437. All non-nuclear component production was undertaken in Royal Ordnance factories, rather than being subcontracted to commercial firms.Google Scholar
  51. 49.
    Ibid., vol. 2, pp. 436–7, 448 and 474.Google Scholar
  52. 50.
    For a discussion on this point see Hewlett and Duncan, op. cit., p. 415.Google Scholar
  53. 51.
    Gowing, Independence and Deterrence, vol. 2, pp. 474–5.Google Scholar
  54. 52.
    Ibid., vol.1, pp. 184–5.Google Scholar
  55. 53.
    Ibid., p. 94.Google Scholar
  56. 54.
    Hewlett and Duncan, op. cit., pp. 264–71 and Science, Technology and American Diplomacy, op. cit., pp. 57–122.Google Scholar
  57. 55.
    Hewlett and Duncan, ibid., p. 272.Google Scholar
  58. 56.
    Ibid., pp. 274–5.Google Scholar
  59. 57.
    Ibid., pp. 277–9.Google Scholar
  60. 58.
    Ibid., pp. 279–84 and Gowing, Independence and Deterrence, vol.1, pp. 249–52. The document is reproduced in Gowing, pp. 266–72.Google Scholar
  61. 59.
    Gowing, ibid., pp. 254–6.Google Scholar
  62. 60.
    Hewlett and Duncan, op. cit., pp. 286–93.Google Scholar
  63. 61.
    Ibid., pp. 293–5 and Gowing, Independence and Deterrence, vol.1, pp. 258–62.Google Scholar
  64. 62.
    Gowing, ibid., p. 263.Google Scholar
  65. 63.
    Contemporary US war plans envisaged the Soviet Army reaching the Channel ports in 60–90 days, and then launching an invasion of Britain with 45 000 airborne troops and 100 000 seaborne ones, backed up by 350 000 reserves. Soviet attempts to obtain air superiority over the United Kingdom were assumed from the outbreak of hostilities in order to prevent US strategic bombers using British bases. Estimate of the Scale and Nature of a Soviet Attack on the United Kingdom between Now and mid-1952, pp.7 and 9.Google Scholar
  66. 64.
    Hewlett and Duncan, op. cit., pp. 299–304, and Gowing, Independence and Deterrence, vol. 1, pp. 275–9.Google Scholar
  67. 65.
    Gowing, ibid., pp. 283–4 and Hewlett and Duncan, ibid., p. 306.Google Scholar
  68. 66.
    Gowing, ibid., p. 289.Google Scholar
  69. 67.
    Ibid., p. 293.Google Scholar
  70. 68.
    Ibid., pp. 294–5 and Hewlett and Duncan, op. cit., pp. 308–10.Google Scholar
  71. 69.
    Gowing, ibid., pp. 296–7.Google Scholar
  72. 70.
    Ibid., vol. 2, p. 449.Google Scholar
  73. 71.
    Ibid., vol. 1, p. 298 and Hewlett and Duncan, op. cit., pp. 310–14.Google Scholar
  74. 72.
    Gowing, ibid., pp. 299–303.Google Scholar
  75. 73.
    Ibid., pp. 303–4 and pp. 325–7.Google Scholar
  76. 74.
    Hewlett and Duncan, op. cit., pp. 480–3.Google Scholar
  77. 75.
    Gowing, Independence and Deterrence, vol. 1, p. 304.Google Scholar
  78. 76.
    Ibid., pp. 304–5. Positive vetting involves attempting to discover adverse information about an individual’s loyalty and behaviour. The contemporary British system merely involved checking that no such information existed in police records, etc. and the Conservative government was opposed to extending this procedure, partly because officers swore allegiance to the Crown and ministers were Privy Councillors, and partly because in the United States many people were only second generation citizens whereas most of the British population could trace their citizenship back for centuries. After 1955 positive vetting started to be introduced for all new AEA and Ministry of Defence employees, mainly because it was seen as a necessary prelude to closer cooperation with the United States in nuclear matters.Google Scholar
  79. 77.
    Ibid., pp. 307–8.Google Scholar
  80. 78.
    Ibid., pp. 311–18.Google Scholar
  81. 79.
    494 HC Debs Col 280, Written Answers, 6 December 1951 and ibid., p. 318.Google Scholar
  82. 80.
    Gowing, Independence and Deterrence, vol.1, pp. 410–11 and pp. 415–16, and Hewlett and Duncan, op. cit., pp. 574–5.Google Scholar
  83. 81.
    Gowing, ibid., pp. 411–14, 416 and 441–2.Google Scholar
  84. 82.
    Ibid., pp. 413–14.Google Scholar
  85. 83.
    Ibid., p. 442 and 448–9.Google Scholar
  86. 84.
    Strauss, who had been the AEC commissioner largely responsible for the furore which led to the limiting of interchange under the 1948 modus vivendi, became both AEC chairman and atomic energy adviser to the President on 1 July 1953. Green and Rosenthal, op. cit., pp. 12–13. Nichols is stated to have been ‘a hostile critic of the British programme’ in Gowing, ibid., p. 294.Google Scholar

Copyright information

© John Simpson 1983

Authors and Affiliations

  • John Simpson
    • 1
  1. 1.University of SouthamptonUK

Personalised recommendations