by Ray Monk
This close collaboration between theorists and experimentalists at Berkeley was to be mutually beneficial in the mid- to late 1930s, but in the crucial year of 1932 there are signs that communication was not all it should have been: not only was Oppenheimer wrong-footed by experimental developments, but Lawrence seemed to be held back by a lack of appreciation of certain crucial theoretical points.
Most of Lawrence’s efforts were directed at building bigger and bigger cyclotrons and accelerating protons to greater and greater energies. A key moment in this development came in February 1932, when his latest cyclotron succeeded in accelerating protons to an energy of one million volts. ‘I wrote the figure on the blackboard,’ Livingston later remembered. ‘Lawrence came in late one evening. He saw the board, looked at the microammeter to check the resonance current and literally danced around the room.’ Ever the publicist, Lawrence lost no time in spreading the news, and the next day Livingston recalls: ‘We were busy all that day demonstrating million-volt protons to eager viewers.’ The San Francisco Examiner sent a reporter to Berkeley to see what was going on, and announced excitedly that Lawrence and Livingston ‘are setting about trying to break up the atom and release its terrific energy’. Reporting on plans to build an even bigger cyclotron, the Examiner went on: ‘With the greater magnet, they hope to shatter the atom completely with an ultimate 25,000,000-volt impact.’
But, as quantum theorists had known for many years before this, and as Oppenheimer must surely have been aware, there was a very good chance of splitting the atom with much lower energies than these. Indeed, the key theoretical point in understanding this centres on a phenomenon that Oppenheimer himself had been among the first to consider: the mysterious process known as ‘quantum tunnelling’. Due to the fact that, according to quantum theory, protons, along with electrons, and so on, are both particles and waves, there is a significant possibility that the particles that make up a nucleus can suddenly appear outside the electrical barrier (the ‘Coulomb barrier’) that surrounds the nucleus. Rutherford had known about the phenomenon, but had pictured it as electrons pulling protons out of their shell, like tugboats pulling liners out to sea. It was the Russian physicist George Gamow who first realised that it was a direct consequence of the wave-like nature of subatomic particles, and that quantum mechanics offered the means to quantify and predict the probability of such ‘tunnelling’ occurring.
Gamow’s analysis appeared in print twice in quick succession in the autumn of 1928: in German in the Zeitschrift für Physik on 12 October, and in English in Nature on 24 November. Actually, in terms of being the first to publish such an analysis, Gamow had narrowly been beaten to it by an article by Edward Condon and Ronald Gurney that appeared in Nature in September, but, from the point of view of the history of splitting the atom, Gamow’s account is more important for two reasons: first, unlike Condon and Gurney, Gamow was less concerned with how protons might get out of the nucleus than how they might get in; and second, it was Gamow’s account that stimulated the work that led to the world’s first splitting of the atom.
Shortly before it was published, an advance copy of Gamow’s Zeitschrift article was sent to John Cockcroft at the Cavendish Laboratory, who saw immediately its implication that atoms might be split with relatively low-voltage protons. Rutherford, when he made a public appeal for progress in artificial acceleration of particles, had thought that one needed eight million volts or more. However, in a series of calculations that Cockcroft wrote out, showed to Rutherford, but did not publish, he demonstrated that, if Gamow’s analysis was correct, a ‘mere’ 300,000 volts would probably be enough.
Encouraged by Rutherford, Cockcroft worked together with his Cavendish colleague Ernest Walton to design and build a machine capable of accelerating protons to the required 300,000 volts. By May 1930, they had a machine capable of 280,000 volts and felt able the following August to publish an account of their work in the Proceedings of the Royal Society, in which they took the possibly risky step of making public Cockcroft’s calculation that 300,000 volts would be enough to penetrate the nucleus and so split the atom. Around the same time, however, Rutherford gave a speech in which he claimed: ‘What we require is an apparatus to give us a potential of the order of ten million volts.’ Whether Rutherford believed this, or whether these words were a smokescreen, is difficult to say, but, knowing perfectly well that Lawrence and his team were constructing machines capable of more than one million volts, Rutherford continued to encourage Cockcroft and Walton in their endeavours using more modest machinery.
In January 1932, Cockcroft received a letter from his old friend Joseph Boyce, who told him: ‘I have just been on a very brief visit in California and thought you might be interested in a brief report on high-voltage work there.’ The ‘place on the coast where things are really going on,’ reported Boyce, ‘is Berkeley’:
Lawrence is just moving into an old wooden building back of the physics building, where he hopes to have six different high-speed particle outfits. One is to move over the present device by which he whirls protons in a magnetic field and in a very high frequency tuned electric field and so is able to give them velocities a little in excess of a million volts.
Lawrence, Boyce added, ‘is a very able director, has many graduate students, adequate financial backing, and his work so far . . . has achieved sufficient success to justify great confidence in his future’.
In the light of such reports, Rutherford’s encouragement became more urgent. On 14 April 1932, after they had been told by Rutherford to ‘stop messing about and wasting their time’, Cockcroft and Walton, without any great hopes of success, fired some accelerated protons at a sample of lithium, a very light metal with an atomic mass of 7. The result was so astonishing that Rutherford and Chadwick were called to the laboratory to verify that there was no mistake in their observations. What they all saw were the familiar scintillations that told of the emission of alpha particles.fn33 There was only one conclusion to draw: the protons had caused the lithium nucleus to break up, forming two alpha particles (that is, helium nuclei, which have an atomic mass of 4, which makes perfect mathematical sense, since the combined atomic masses of the lithium nucleus plus the proton is 8, equivalent to two helium nuclei). In other words, what Cockcroft and Walton had achieved was the world’s first splitting of the atom by artificial means. What is more, they had done it with protons accelerated to a voltage significantly below the 300,000 that Cockcroft had calculated. In subsequent tests, they discovered that lithium nuclei could be disintegrated at 125,000 volts, which was far below what anybody had thought possible.
When Cockcroft and Walton measured the energy of the alpha particles emitted from the reaction, the results provided both dramatic confirmation of the most famous equation in science – E = mc2 – and a startling illustration of the kind of energy that can be released by an atomic reaction. For the answer was eight million volts. Since from each lithium nucleus there emerged two alpha particles, this means that, from the collision of a single proton, travelling with an energy of 125,000 volts, with a single lithium nucleus, sixteen million volts of energy had been released (two alpha particles, each with an energy of eight million volts). Little wonder, then, that people immediately started to wonder how such tremendous energy releases might be used in explosives.
Partly because he was acutely conscious that this kind of speculation would inevitably follow the announcement of their achievement, Rutherford – after helping Cockcroft and Walton write up their experiment for Nature – urged upon them the importance of keeping quiet about it until a sober account had appeared in print. But, as Walton spelled out in a letter written to his girlfriend, Freda, this was not Rutherford’s only reason for not letting the news leak out. ‘We know,’ Walton wrote, ‘that people in the States are working along similar lines and Rutherford would like to see any credit going to the Cavendish. He is not fond of American physicists in general on account of their tendency to do a great deal of boasting ab
out very little.’
And yet Rutherford was evidently itching to announce the news. On 28 April 1932, two days before the report for Nature appeared, he chaired a meeting at the Royal Society in London on ‘The Structure of Atomic Nuclei’, which had been organised primarily to allow discussion of Chadwick’s discovery of the neutron. Rutherford arranged for Cockcroft and Walton to be present at this occasion, and then, before introducing Chadwick, announced their achievement in disintegrating the lithium nucleus.
Two days later the Nature piece appeared, but, meanwhile, Rutherford’s announcement at the Royal Society had attracted the attention of the press, and on Sunday 1 May 1932, the Reynold’s Illustrated News, under the heading ‘SCIENCE’S GREATEST DISCOVERY’, reported:
A dream of scientists has been realised. The atom has been split, and the limitless energy thus released may transform civilisation . . . This is the greatest scientific discovery of the age.
The same day the Sunday Express went with: ‘The Atom Split. But World Still Safe’, while the Daily Mirror pleaded: ‘Let it be split, so long as it does not explode.’ The idea that atom-splitting would lead to extremely powerful bombs had been around since the 1920s. Bertrand Russell mentioned it in his 1923 best-seller The ABC of Atoms, and it formed the central idea in a play called Wings over Europe, in which scientists threaten world leaders that they will use atom bombs to destroy the major cities of the world unless an international policy is agreed to use the tremendous energy released by nuclear reactions. The play had premiered in New York in 1928, but, by a strange coincidence, was showing in London at the very time that news broke about the splitting of the atom by Cockcroft and Walton.
As it turned out, of course, the fear of the energy that might be released from within a nucleus was, in general, well founded. However, for the moment it was premature; the kind of energy release observed in the disintegration of lithium, though extremely impressive, could not be used to make explosives. This is for two reasons. First, though there is a dramatic difference between the energy of the penetrating protons and the energy of the released alpha particles, one has to bear in mind that, at 125,000 volts, only one proton in about ten million will penetrate the nucleus. Thus, the total energy needed to release the sixteen million volts of the two alpha particles is about 1.25 billion volts. Second, an explosive requires a chain reaction, which had not yet been witnessed and which could not possibly occur with the disintegration of a light element like lithium. Rutherford’s famous dismissal as ‘moonshine’ of the idea that atomic-splitting might be a source of energy in the future was, therefore, perfectly reasonable in the light of what was then known. However, it was while pondering a report of that remark in The Times that it suddenly occurred to the Hungarian scientist Leo Szilard, that, if one could find an element that disintegrated when bombarded with neutrons, and if that element emitted two neutrons for every one it absorbed, then a chain reaction could occur that would be a source of enormous energy.
It would take a few years for the rest of the world to catch up with Szilard. Meanwhile, the story of Cockcroft and Walton’s achievement was picked up by newspapers all over the world, including the New York Times, which in its Sunday edition on 8 May, under the heading ‘The Atom Is Giving Up Its Mighty Secrets’, described the experiment and commented: ‘Never was a result more unexpected obtained.’ This was no doubt read by Ernest Lawrence, who was at that time preparing to get married to Mary Blumer in New Haven, Connecticut. The wedding took place on 14 May, by which time, it is certain, Lawrence knew that he had been pre-empted as the first person to smash the atom. It is therefore not true, as legend has it, that he heard the news while on honeymoon. It is possible, however, that the telegram his assistant James Brady remembers receiving – ‘Cockcroft and Walton have disintegrated the lithium atom. Get lithium from chemistry department and start preparations to repeat with cyclotron’ – was sent by him while still on honeymoon.
The news from the Cavendish did not stop Lawrence from building bigger and bigger machines, nor did it damage his ability to attract huge funds for these projects – indeed, if anything, it helped by stirring up interest – but it must have made him aware of the importance of being well informed about theory, just as Chadwick’s discovery of the neutron had made Oppenheimer aware of the importance of keeping up with experimental developments.
The fourth and final major development in experimental physics during the annus mirabilis of 1932 – the discovery of the positron by Carl Anderson of Caltech – exhibited a lack of communication between theorists and experimentalists that seems nothing short of bizarre. As Graham Farmelo, Paul Dirac’s biographer, has said: ‘Many of the characters in this strange denouement, including Dirac, behaved in ways that are now barely comprehensible.’
Among those characters whose behaviour seems inexplicable was Oppenheimer himself. There are several reasons why one might have expected Oppenheimer to have been in close contact with Anderson while he conducted the research that led to the discovery of the positron. In the first place, Anderson, as we have seen, had been a student of his. Indeed, during his first lecture course at Caltech in 1930, Anderson had been his only student. Second, though Anderson’s postdoctoral research was conducted with Millikan rather than with Oppenheimer, it was on a subject in which Oppenheimer had a deep interest: the nature of cosmic rays. Third, the hypothetical existence of the positron, a particle with the same mass as an electron but with a positive rather than negative charge, had been discussed by Oppenheimer in print, when he showed that Dirac’s theory of quantum electrodynamics – the very theory he had tried to explain in a lecture attended by Anderson – demanded it. And yet, despite all this, when Anderson discovered the positron, he did so apparently unaware that the existence of such a particle had been predicted by Dirac, or indeed that the possibility of its existence had been discussed at all by anybody.
Anderson had started his research on cosmic rays in the autumn of 1930, after he had completed his PhD. Though he worked with Millikan, he regarded Millikan’s theological view of cosmic rays as mere wishful thinking, and certainly did not feel himself obliged to provide evidence for it. Rather, he wanted to gather hard evidence about the nature of cosmic rays, and so developed a method of photographing their activity inside a cloud chamber, which allowed him to make visual records of the paths of charged particles emitted from cosmic-ray collisions. By the autumn of 1931, Anderson had about 1,000 such photographs and, in November, he wrote to Millikan, who was then in Cambridge in order to give a paper at the Cavendish, sending him some photographs that puzzled him. What the photographs appeared to show were collisions that resulted in the simultaneous emission of a negatively charged particle, which was surely an electron, and a positively charged particle, which Anderson assumed to be a proton.
Millikan could shed little light on these photographs, but exhibited them at the Cavendish anyway, presenting them merely as evidence of the tremendous energies of cosmic rays, which he thought explicable only by adopting his own theological interpretation of them. Among Millikan’s audience at the Cavendish, however, was Patrick Blackett, who was deeply intrigued by Anderson’s photographs and resolved to find an explanation for them. In fact, the explanation for the phenomenon Anderson had photographed had already been given by Dirac in his lecture at Princeton in October, in which he had said that it should be possible to detect positrons – or, as he was calling them at this time, ‘anti-electrons’ – experimentally. In collisions between pairs of ultra-energy photons, Dirac explained, sometimes the photons should disappear and in their place should appear a pair of particles: an electron and an ‘anti-electron’, a process subsequently named ‘pair production’.
Clearly, this is what had happened inside Anderson’s cloud chamber, but when Millikan addressed the Cavendish in November, Dirac was still in Princeton, and nobody else seems to have made the connection between Dirac’s prediction and Anderson’s photographs. Why did the connection not occur to Oppenheimer? O
r, if it did, why did he not mention it to Anderson? Late in life, Anderson recalled that around this time he ‘talked to Oppenheimer quite a bit’, but also that ‘I found it hard to talk to Oppenheimer because his answers were usually, at least to me, encased in some sort of mysticism. I couldn’t understand what he was saying, but the idea of pair production, if he had said that, I would have understood.’ As Farmelo remarks: ‘It beggars belief that Oppenheimer never pointed out the connection between Dirac’s theory and Anderson’s experiment to Dirac, to Anderson or to anyone else. Yet that appears to be what happened.’
Several possibilities suggest themselves. One is that the idea of pair production simply did not occur to Oppenheimer. After all, he was not at Dirac’s Princeton lecture and the lecture had not been published, so he might well have remained ignorant of Dirac’s latest thoughts on the question. But even if the specific notion of pair production did not occur to him, it still seems odd that he did not mention what he had already said in print – namely that Dirac’s theory demanded the existence of a positively charged particle with the same mass as the electron. Another thought is that he was reluctant to help someone working on cosmic rays with Millikan, because he assumed that the point of his research was to lend support to an analysis of cosmic rays that he thought was mistaken. Most likely, though, is that he was still so convinced that Dirac’s theory was wrong that the last thing he thought Anderson could possibly have photographed was evidence that it was right. This does not entirely resolve the puzzle, since it raises the question: if Oppenheimer did not think Anderson had photographed the positron, what did he think he had photographed? A proton?