Inside the Centre: The Life of J. Robert Oppenheimer

by Ray Monk

  Despite the many difficulties and uncertainties, by January 1944 the ‘gadget’ had been designed and a suitable name, the ‘Thin Man’, had been chosen for it. All that remained, so it was believed, was for Parsons and his Ordnance Division to test the dropping of it and work out the details of its internal ballistics. A few months later, however, in April 1944, Segrè finally received some samples of reactor-produced plutonium and, to everybody’s horror, discovered that the rate of spontaneous fission was five times that of the cyclotron-produced samples he had measured earlier. Just as Seaborg had warned, the plutonium had far more Pu-240 in it than that produced by a cyclotron. The alarming but inescapable conclusion was that the ‘Thin Man’ was a non-starter. The whole idea of a gun-assembly plutonium bomb – the idea that up until then had formed the central focus for almost all the work done at Los Alamos – would have to be abandoned.

  This was devastating news, but in Segrè’s earlier measurements of spontaneous fission in uranium there was a silver lining: a gun-assembly bomb made with uranium would work and, in fact, was even more straightforward than they had thought. The uranium bullet could be fired at a mere 1,000 feet per second and the length of the gun could be reduced from seventeen feet to six. Thus, in place of the plutonium ‘Thin Man’ bomb, there emerged the uranium ‘Little Boy’, the bomb that would be dropped on Hiroshima. So confident were Oppenheimer and his colleagues that ‘Little Boy’ would work that they did not see any need to test it. The bomb was designed and built and then left to one side, waiting for the U-235 that would form both the bullet and its target.

  Until recently, every book about the history of the atomic bomb has contained a false description of Little Boy’s design. Perhaps misled by the Frisch–Peierls memorandum and by The Los Alamos Primer, the published accounts of the bomb invariably state that the gun assembly worked by firing a small uranium bullet into a slightly subcritical mass of U-235, thus forcing it to go supercritical. In fact, the material was split almost in half: at one end of the gun was a group of rings of U-235 that formed 40 per cent of the supercritical mass, and at the other end another group of slightly larger rings that formed 60 per cent. And it was this latter, larger group that was fired onto the smaller group. At the same time, neutrons were emitted from a polonium-beryllium ‘initiator’, thus causing the supercritical mass to explode.

  For more than fifty years this was an official secret, known only to those who worked on the bomb. Then, in 2004, a truck driver from Illinois called John Coster-Mullen published a book that contained the first-ever publicly available accurate description of the Little Boy bomb. Coster-Mullen’s hobby was model-making and he had set himself the task of producing an accurate model of the Hiroshima bomb, to accomplish which he made a close study of every photograph and every document available. His research convinced him that the accounts published up to that time were wrong, and he set about correcting them. That he, a man without a university degree in physics (or, indeed, anything else), was able to reverse-engineer the bomb and produce an accurate and detailed account of its design shows, perhaps, the wisdom of Bohr’s remark that there never was any secret about how to make a fission bomb. Or, as Coster-Mullen himself has put it, what his research has shown is that the real secret of the atomic bomb is how easy it is to build one.

  At the beginning of July 1944, Oppenheimer broke the news to the scientists gathered for the Los Alamos weekly meeting that the plutonium ‘gadget’, as it had been conceived up until that point, would have to be abandoned; there was absolutely no prospect of producing a gun-assembly plutonium bomb. The reason it took three months to make this announcement is that Segrè, Oppenheimer and Groves, so appalled were they at the consequences of Segrè’s results, kept hoping that further experiments, further measurements, would show those results to be incorrect. Alas, no matter how many times Segrè and his team counted the spontaneous fissions produced by reactor-produced plutonium, the result was always the same: too high for gun assembly to be workable. Groves had even been reluctant to inform the scientists at the Met Lab in Chicago of the results. When, at the beginning of June 1944, Robert Bacher told him of his intention to report Segrè’s findings to the scientists there, Groves replied: ‘Do you think that needs to be reported to them?’ ‘Of course,’ Bacher replied, ‘it’s a fundamental fact of the material they’re working on.’ When he did tell them, Bacher recalls that Compton ‘went just as white as that sheet of paper’.

  In the late spring of 1944, then, it looked extremely doubtful that the Manhattan Project would succeed in achieving the goals set for it. For, if there was no prospect of a gun-assembly plutonium bomb, neither, given the agonisingly slow progress at Oak Ridge, was there any hope of producing two gun-assembly uranium bombs by the following summer. If they were to meet the target set for them, therefore, they would have to start from scratch and design a plutonium bomb fundamentally different from the ‘gadget’ they had had in mind during the previous year.

  That the laboratory was able to do just that is astonishing and demonstrates, as well as the resolute determination of everyone involved, the foresight and adaptability that Oppenheimer brought to his task as director. Thanks to his foresight, an alternative to the gun-assembly plutonium bomb lay ready to hand: the implosion method of detonation, first suggested by Richard Tolman, elaborated upon by Robert Serber and pursued with tenacious dedication at Los Alamos by Seth Neddermeyer. Oppenheimer’s adaptability is shown in the way that, in the summer of 1944, he reorganised the whole laboratory, turning it away from the ‘Thin Man’ and towards what became known as the ‘Fat Man’, the plutonium implosion bomb that would – thanks to an almost unimaginably intense effort – be ready for military use just a year later.

  For the first six months of work at Los Alamos, implosion was very much a side issue, comparable in that respect to the work Edward Teller was pursuing on the ‘Super’. Like the Super, it was seen as something that was potentially interesting, both scientifically and militarily, but, compared to the gun-assembly gadget, of marginal concern to the project. To pursue implosion, Neddermeyer had a team of just eight people, who spent their time in a remote canyon performing experiments with explosives that most people on the Hill thought were leading nowhere. Among those sceptics was L.T.E. Thompson, a naval ballistics expert whom Parsons consulted and whose judgement he trusted above that of any other advisor. ‘Dr Tommy’ (as he was known to Parsons and his family) came to Los Alamos in the summer of 1943 and observed a demonstration given by Neddermeyer of the basic idea of implosion. ‘It seems to me,’ Thompson announced afterwards, ‘there is a fundamental difficulty with the system that makes it quite certain not to be satisfactory.’

  The ‘system’ he was commenting upon was, in some important respects, quite different from the implosion device first envisaged by Tolman and described by Serber. What Tolman and Serber had imagined was a way of assembling a critical mass of plutonium (or uranium) that brought together several pieces of the metal arranged in a circle. What Neddermeyer had in mind, rather, was something more subtle, which would exploit the fact that critical mass is affected by density. For a fairly straightforward reason, the critical mass of a dense piece of material is lower than that of a less-dense piece. The reason is that, in a dense piece of matter, the distance – and therefore the time – that a neutron has to travel before it causes fission is smaller, and so, the denser the material is, the smaller it can be while still being able to undergo the eighty generations of fission needed for an explosion.

  Neddermeyer’s concept of implosion exploited this fact in a novel way. Instead of having a bomb that assembled a critical mass by bringing together two or more subcritical masses – which was the idea behind both the gun-assembly design and the Tolman/Serber version of implosion – Neddermeyer proposed turning a subcritical mass of material into a supercritical mass by squeezing it. His design called for a subcritical hollow sphere of uranium or plutonium to be blown inwards, imploded, uniformly, so that its densit
y increased to the point at which it would go supercritical. The squeezing would be achieved by explosives arranged around the sphere. The ‘fundamental difficulty’ that Thompson identified was that the design required the external pressure on the sphere to be exactly symmetrical. If it were not, the sphere would not be transformed into a denser sphere, but rather flattened, ‘in about the manner of a dead tennis ball hit with a hammer’, as Thompson put it in a letter to Oppenheimer of June 1943.

  Until September 1943, Neddermeyer was almost alone at Los Alamos in thinking that this problem could be overcome and that implosion would turn out to offer a practical method of creating an atomic bomb. In their efforts to solve the problem, Neddermeyer and his team conducted experiments in which they surrounded stove pipes with TNT, which they then detonated, trying to get the pipes to collapse symmetrically. The results were not encouraging. Those who were called upon to witness these experiments, and the non-symmetrically flattened pipes they invariably produced, were unimpressed. Then, in September 1943, Oppenheimer, Groves and some of the leading theoreticians at Los Alamos suddenly began to take implosion more seriously and to regard it as a centrally important part of the laboratory’s work. This was six months before Segrè’s devastating news about the spontaneous fission of reactor-produced plutonium, and was completely unrelated to any perceived problems with the gun-assembly design. The renewed interest in implosion was, rather, stimulated at this early stage by the problems in obtaining significant amounts of enriched uranium, in the light of which a method that offered the possibility of building a bomb that required a smaller critical mass of uranium and did not need such pure uranium seemed well worth exploring.

  Fresh hope for the implosion programme was provided that September by a visit to Los Alamos of John von Neumann. Von Neumann was a Jewish émigré from Budapest who was considered, among some extraordinarily stiff competition (Teller, Szilard, Wigner, to name but three), to be the most brilliant of the Hungarian scientists associated with the Manhattan Project. Having, while still a young man, made fundamentally important contributions to a wide variety of disciplines (including logic, mathematics, quantum theory and economics), von Neumann moved to the United States in 1930, and in 1933, at the age of twenty-nine, was appointed professor of mathematics at Princeton’s Institute for Advanced Study. He had the kind of mind that could turn to anything and, as luck would have it, during the war he became interested in the mathematics and physics of explosives. This led to a series of consultancies, mainly for the US navy, in which he demonstrated time and time again the military usefulness of a powerful intellect. Oppenheimer and Parsons were desperate to get him to Los Alamos, but were unable to tempt him to come on a full-time basis. The best they could do was persuade him to make ‘an occasional visit to Santa Fe’, as Parsons put it in a letter to him of August 1943.

  On his first such visit, at the end of September 1943, von Neumann, during the two weeks that he was at Los Alamos, put the implosion programme on a completely different footing, replacing the sceptical indifference towards the project that had prevailed up until that point with a lively, intense and optimistic interest. This was largely because of the seriousness with which he himself treated Neddermeyer’s research. Such was the awe in which von Neumann was held that the leading scientists at Los Almaos began to think that, if he was interested in implosion, there must be something to it. With regard to the technical problem that Neddermeyer was attempting to solve, von Neumann’s initial contribution was to suggest two things: 1. increase the amount of explosive used to implode the fissionable material so as to increase the speed of the implosion; and 2. use ‘shaped’ explosive charges, which make better use of the physical properties of shock waves (in the mathematics of which von Neumann was by this time probably the world’s leading expert).

  Shaped charges or (‘hollow charges’ as they are known in Britain) had been discovered in the nineteenth century, but were not exploited militarily until the advent of armour-piercing shells in the Second World War. The basic idea is that, instead of having a solid explosive charge – like, say, a stick of dynamite – one hollows the charge out, leaving an empty space. It was found that this concentrates the energy produced by the explosion (because the energy released rushes to fill the empty space), allowing more penetrative weapons to be designed and built. Von Neumann was an expert on this kind of charge and realised that if, instead of simply surrounding his target material with TNT, Neddermeyer arranged a series of shaped charges around it and was then able to ensure the charges all went off at the same time, implosion might just work.

  Charles Critchfield, who was a member of Neddermeyer’s team, has described how von Neumann’s suggestions ‘woke everybody up’. After von Neumann’s visit, he remembered, Teller called him, saying: ‘Why didn’t you tell me about this stuff?’ At a meeting of the Los Alamos Governing Board of 28 October 1943, Oppenheimer gave reasons for giving high priority to the implosion programme – reasons based largely on the interest that von Neumann had shown in it. He mentioned in passing that von Neumann believed that the speed of implosion (if the charge was great enough and arranged well enough) was so great that ‘there is less danger of pre-detonation’, but at this stage what really excited Oppenheimer, Groves and Teller about implosion was the promise it offered of reducing the quantity of enriched uranium needed to make a bomb. In his report to the Governing Board of 4 November 1943, Oppenheimer remarked that ‘both Groves and Conant seemed very much in favour of pushing the implosion method . . . the only one which offers some hope of justification for the electromagnetic method’. In other words, the only remaining hope for the hugely expensive Calutrons at Oak Ridge was not that they could produce enough enriched uranium for a gun-assembly bomb – that clearly looked a false hope – but rather that they might produce enough, slightly impure, uranium to make an implosion bomb.

  If implosion looked more interesting, attractive and promising after von Neumann’s visit, it also became more urgent – too urgent to leave in the hands of Neddermeyer and his tiny group. Oppenheimer thus began to enlarge the programme and recruit onto it people with more experience of explosives. His first target was George Kistiakowsky, the Ukrainian-born professor of chemistry at Harvard, who was director of the National Defense Research Committee’s Explosive Research Laboratory at Bruceton, Pennsylvania, and probably the most eminent expert on the chemistry of explosives in the US. Kistiakowsky was at first reluctant to come to Los Alamos, ‘partly’, he later said, ‘because I didn’t think the bomb would be ready in time and I was interested in helping to win the war’. He nevertheless agreed to visit Los Alamos in October 1943 as a consultant. What he found dismayed him. ‘The situation is a mess,’ he wrote to Conant after his visit. ‘The real difficulty is that there is a serious lack of mutual confidence between Parsons and Neddermeyer.’ Furthermore, though Parsons ‘is now committed to a vigorous prosecution’ of implosion research, it ‘is doubtful that he believes in its success’.

  Kistiakowsky recommended an enlargement and a reorganisation of the implosion programme, with the appointment of a new leader for it, someone who could work with Parsons. Though he was still reluctant to commit himself to the task, it was clear to everyone that he himself was the right man for it. However, he took some time to see what was staring everybody else in the face and did not join the laboratory on a full-time basis until February 1944. In the meantime, while still acting as a consultant, he set about reorganising the research programme, providing it with a more rigorous scientific method and a detailed list of particular experimental studies to carry out. When he eventually joined Los Alamos, it was as a deputy leader of the Ordnance Division, with responsibility for implosion (with Ed McMillan as the other deputy, with responsibility for gun-assembly). Kistiakowsky and McMillan were also made members of the Governing Board, as was Kenneth Bainbridge, who was at the same time appointed leader of a newly created division dedicated to the problems of bomb assembly.

  Though much research s
till needed to be done, certain features of the design of the implosion bomb were fixed in the autumn of 1943. Its maximum size, for example, was determined by the size of the bomb bay in the B-29 bomber that would be used to deliver it: five by twelve feet. Also, it was realised from the beginning that an implosion bomb could not be long and slender, but would have to be round and large, hence the name ‘Fat Man’. Apparently, the hope was that anyone listening covertly to discussions about modifying B-29s to accommodate ‘Thin Man’ and ‘Fat Man’ would interpret them as referring to plans to transport, respectively, President Roosevelt and Prime Minister Churchill.

  Directing the programme to design and build ‘Thin Man’, which lasted from March 1943 to July 1944, had been a demanding job, but its demands paled before those required to complete ‘Fat Man’ by the deadline of the summer of 1945. This latter was a truly gargantuan task that involved, among other things: mastering, and in some cases, inventing new mathematical techniques to describe and predict the behaviour of shock waves; determining, by experiment and observation, the right shape for the explosive charges that would be used to implode the fissionable material; inventing a method of initiating the chain reaction in an implosion device; developing a new branch of physics (the hydrodynamics of implosion);fn53 and designing and constructing a kind of bomb that no one before the war had even envisaged.