The First War of Physics (31 page)

Diebner started to draw up plans for an even larger reactor, but now ran into conflict with the demands of Heisenberg’s experiments. Heisenberg preferred to continue with the layer configuration despite the evidence suggesting that the lattice arrangement might work better. At issue here was the very different experimental philosophies adopted by the two research groups. Heisenberg was content to build understanding of the physics through a series of reactor
experiments
designed to allow measurement of the values of fundamental nuclear constants. As Heisenberg later confided to Harteck, he preferred the layer configuration because the theory was much simpler.

Diebner was less concerned about the theory and wanted to build a working reactor as quickly as possible. When subsequent theoretical studies pointed to the superiority of Diebner’s lattice configuration, Heisenberg remained stubborn. Professional pride may have been a factor, but the simple truth was that for Heisenberg the nuclear project was no longer his major preoccupation.

More ominously, perhaps, Heisenberg had so far perceived no need for cadmium control rods of the kind that had been used in the Chicago uranium–graphite pile, although he understood that these would be required in a working reactor. In truth, without control rods an experimental nuclear reactor reaching criticality would precipitate a major disaster.

A lot of experience in microfilm work

‘Oh, I think that is true’, Oppenheimer said, in answer to Pash’s question concerning other groups interested in the work going on at the Rad Lab. ‘But,’ he went on to say, ‘I have no firsthand knowledge. I think it is true that a man, whose name I never heard, who was attached to the Soviet consul, has indicated indirectly through intermediary people concerned in
this project that he was in a position to transmit, without danger of leak, or scandal, or anything of that kind, information which they might supply.’

Oppenheimer explained that, speaking frankly, he was ‘friendly’ to the idea that the Russians – as allies of America in the war against Nazi Germany – be advised of the American work on the atomic bomb, but that he would not want this kind of information to get to the Soviets through the ‘back door’.

Pash was all ears.

‘Could you give me a little more specific information as to exactly what information you have?’ Pash enquired. ‘You can readily realise that phase would be, to me, as interesting, pretty near, as the whole project is to you.’

‘Well, I might say,’ replied Oppenheimer, ‘that the approaches were always to other people, who were troubled by them, and sometimes came and discussed them with me.’ He went on: ‘[T]o give more … than one name would be to implicate people whose attitude was one of bewilderment rather than one of co-operation.’

In Oppenheimer’s reply, the Chevalier incident had suddenly become one of several approaches, to several physicists working on the programme. Two of these physicists, Oppenheimer explained, were working with him at Los Alamos, and the other was a Rad Lab physicist who had departed, or was about to, for the Oak Ridge facility in Tennessee. It was, as he later admitted, a ‘cock and bull story’, designed – if Oppenheimer’s flustered response could be called that – to throw Pash off the scent.

Oppenheimer had already named Eltenton, who, he now explained, was to arrange contact with someone from the Soviet consulate ‘who had a lot of experience in microfilm work, or whatever the hell’. But Oppenheimer did not want to name Chevalier, who he believed had acted as an innocent messenger. When pressed by Pash to name his friend, Oppenheimer replied: ‘I think it would be a mistake. That is, I think I have told you where the initiative came from and that the other things were almost purely accident … The intermediary between Eltenton and the project thought it was the wrong idea, but said that this was the situation. I don’t think he supported it. In fact, I know it.’

Pash pressed him further, but other than reveal the fact that the intermediary was a member of the Berkeley faculty, Oppenheimer refused to give a name. ‘I want to again sort of explore the possibility of getting the name of the person on the faculty,’ Pash cajoled, ‘not for the purpose of taking him to task but to try to see Eltenton’s method of approach.’ Oppenheimer did not budge, and tried to downplay the significance of the incident. Surely, the transmission of information vital to America’s allies was something that should in any case be happening through formal channels. The fact that this transmission was not happening meant that information passed through the ‘back door’ was obviously treason in substance, though, perhaps, not in spirit.

These were all sentiments that had been expressed by many in Oppenheimer’s circle of ‘leftwandering’ friends and colleagues. They were not, however, the sentiments expected of the head of the Los Alamos laboratory, a leading contributor to one of America’s most secret war programmes. Worse, Oppenheimer had started to spin a web of deceit, making the classic error of elaborating a lie in the mistaken belief that it would lend it authenticity. He had not yet been caught in this lie but, unknown to him, it had been caught on tape.

The meeting ended as it had begun – amicably. Pash arranged for a transcript of their conversation to be produced and sent it to Groves with a covering note. It made no difference.

By this time the FBI had received a rather extraordinary anonymous letter. The letter was dated 7 August 1943 and was written in Russian. It named Zarubin (Zubilin), Kheifets and Kvasnikov and many others as Soviet spies. It also accused Zarubin of involvement in the March 1940 massacre of nearly 15,000 Polish prisoners of war in the Katyn forest
³
and, rather bizarrely, of spying on the United States for the Japanese. The author clearly hated Zarubin, and urged the FBI to expose him to the Soviet authorities as a traitor, whereupon he would be summarily executed by Vasily Mironov, whom the anonymous author claimed was a Soviet diplomat and a loyal NKVD agent. Inevitably, the FBI was suspicious and didn’t
know quite what to make of the letter, but there were enough independently verifiable references in it to make them pay attention.
⁴

The FBI eagerly agreed to put Eltenton under surveillance. In early September a short note was intercepted from Weinberg to ‘S’ (presumed to be Steve Nelson), requesting that he, Weinberg, should not be contacted. A sure sign, Pash argued, that Oppenheimer had tipped him off. Peer de Silva added his own voice to the growing chorus. He wrote to Groves on 2 September: ‘The writer wishes to go on record as saying J.R. Oppenheimer is playing a key part in the attempt of the Soviet Union to secure, by espionage, highly secret information which is vital to the United States.’

However, the intense surveillance of the radical young physicists at the Rad Lab had turned up no further evidence of espionage. The physicists were nevertheless removed from the programme and its proximity. Lomanitz had been drafted. Friedman was fired shortly after being given a position teaching physics to army recruits at Berkeley. Both Lomanitz and Friedman perceived their predicament to be a direct result of their union activity, and nothing more.

At Lomanitz’s farewell party, Weinberg speculated that their troubles might be the result of something else, but held back from telling them that he might actually be the cause. In the meantime, Weinberg was left on the Berkeley campus under close surveillance in the hope that he would expose more of the Soviet intelligence network.

Oppenheimer had asked Bohm to join him at Los Alamos but Groves intervened, advising Oppenheimer that the transfer could not be sanctioned, giving the rather obscure reason that Bohm had relatives in Germany. Weinberg and Bohm took positions as teaching assistants at Berkeley, presenting the course on quantum theory that Oppenheimer had once taught.

Lansdale interviewed Oppenheimer again on 12 September 1943, in Washington. Lansdale explained that in his position he could do little else but base his suspicions on past associations. What was he to make of:

… the case of Dr J.R. Oppenheimer, whose wife was at one time a member of the Party anyway, who himself knows many prominent Communists, associates with them, who belongs to a large number of so-called ‘front’ organisations, and may perhaps have contributed to the Party himself, who becomes aware of an espionage attempt by the Party six months ago and doesn’t mention it, and who still won’t make a complete disclosure.

But Lansdale also confessed that he believed Oppenheimer to be innocent of any wrongdoing: ‘I’ve made up my mind that you, yourself, are OK,’ he said, ‘or otherwise I wouldn’t be talking to you like this, see?’

‘I’d better be – that’s all I’ve got to say’, Oppenheimer replied.

Thin Man and Fat Man

By the autumn of 1943 the road to the atomic bomb was clear to Oppenheimer and the team of physicists at Los Alamos, but no less fraught with difficulty.

Two huge facilities were now under construction at Oak Ridge for the large-scale separation of U-235. One of these, called Y-12, was an electromagnetic separation plant based on Lawrence’s calutron design. Lawrence had estimated that to separate just 100 grams of U-235 per day would require about 2,000 calutron collector tanks, each set vertically between the pole faces of thousands upon thousands of tons of magnets. The tanks and magnets were organised in oval units – nicknamed ‘racetracks’ – with each racetrack consisting of 96 tanks. Groves believed 2,000 tanks – twenty racetracks – to be beyond the capabilities of the construction company, and cut the number back to 500, or five racetracks, anticipating that advances in the technology prior to completion would increase production rates and compensate for the difference.

The facility required a vacuum system and magnets that had never before been built on this, truly Lawrencian, scale. The magnets were 250 feet long, and weighed between 3,000 and 10,000 tons. Their construction had actually exhausted America’s supply of copper, and the US Treasury had loaned the project 15,000 tons of silver to complete the windings. The magnets required as much power as a large city and were so strong that workers could feel the pull of magnetic force on the nails in their shoes. Women straying close to the magnets would occasionally lose their hairpins. Pipes were pulled from the walls. Thirteen thousand people were employed to run the plant. The first racetrack – Alpha I – began operation in November 1943. It promptly broke down.

Despite the enormous scale of Y-12, Groves had remained largely ambiguous about the prospects for electromagnetic separation. This was very new technology and therefore uncharted territory. About eight miles south-west of Y-12 a gaseous diffusion plant, called K-25, was being constructed. This plant was to be housed in a huge U-shaped building measuring half a mile long by 1,000 feet wide. At the time of its construction it was the largest building in the world. The plant would employ another 12,000 people. This was, at least, more familiar technology. But it was still all a gamble. The gaseous diffusion process itself was still the subject of intense research at Columbia University, and the problems with corrosion by uranium hexafluoride had yet to be solved.

The uncertainties over the separation of U-235 were to some extent compensated for by a growing degree of confidence that the gun method would work and, moreover, that a transportable weapon could be built.

An ordnance expert acting as adviser to the project had identified a flaw in the physicists’ logic not long after the inaugural lectures at Los Alamos in April. The physicists had based their rather pessimistic estimates of the size of the gun that would be required on conventional gun designs. But these conventional designs had to take account of the need for the gun to be fired repeatedly. The gun firing the shy into the sub-critical mass of U-235 at the other end of a bomb would obviously have to fire only once, after which it would be reduced to atoms. This meant that the weight of the gun could be substantially reduced.

For U-235 the basic bomb mechanism was no longer the main problem. All they needed was enough fissile material.

But the Manhattan Project physicists were also backing another horse. Fermi’s successful demonstration of a self-sustaining chain reaction in December 1942 had led directly to the creation of a much larger-scale reactor to produce plutonium at ‘Site W’ in Hanford, south-central Washington state. Construction had started in March 1943, with a 45,000-strong labour force. The first nuclear reactor, named the B-reactor or 105-B, began construction in August based on Fermi’s uranium–graphite design. It would take a year to build the plant, and plutonium was not expected to be available in sufficient quantities for a bomb until early 1945.

However, unlike U-235, it was not at all clear that the gun method would be effective for a plutonium bomb. Too little was known at this stage about this new element’s physical properties for any conclusions to be drawn, particularly in regard to spontaneous fission and problems of pre-detonation. If plutonium exhibited a greater tendency to pre-detonate, then the muzzle velocity from the largest gun would be insufficient. The plutonium shy would be fired too slowly to prevent the bomb from just fizzling out.

In contrast to the gun method, implosion offered the possibility of assembling a critical mass much faster and more reliably. Even better, Teller now realised, a very violent Shockwave could compress a sub-critical mass of plutonium to super-criticality. Implosion would literally squeeze the mass to a super-critical density which would then explode, without the need to assemble a larger super-critical mass of normal density from a hollow sphere made up of separate components.

However, implosion would not be workable unless a spherical shockwave could be created with conventional explosives packed around the outside of the bomb core. The mathematical physicist John von Neumann had demonstrated that, to be effective, an implosive shockwave would need to be spherically symmetrical to within a tolerance of just 5 per cent. In early July, Neddermeyer had set to work with modest implosion experiments set up on a mesa south of the Los Alamos laboratory, across Los Alamos canyon. These involved detonating conventional explosives wrapped around
short lengths of pipe, so that the pipes would close in on themselves to form flat metal bars. Early results looked distinctly unpromising, the pipes emerging bent and twisted, indicating that the Shockwaves thus generated were far from symmetrical.