CHAPTER 3

Heisenberg’s Pioneering Paper

AS A CONSCRIPTED reservist, Professor Heisenberg had served . two months in the previous two summers with the Gebirgsjäger. Obeying a mobilization order late on 25 September 1939, he travelled by train from Urfeld to Berlin and reported to Hardenbergstrasse 12 at ten next morning. There he was informed that his call-up telegram had been a deception and that he should attend next door for a conference of nuclear physicists considering the possible applications of nuclear energy.

The Heereswaffenamt had seized the Kaiser Wilhelm Institute (KWI) for Physics at Berlin-Dahlem as the scientific centre for its uranium research project. Professor Erich Schumann was its coordinating head and he had appointed Dr Kurt Diebner, a second-rank scientist engaged on conventional explosives research, to direct it. The aim was to concentrate and coordinate at Berlin-Dahlem the secret activities of the Uranium Project. In Diebner’s appointment there lay the danger that the project was vulnerable to penetration by political functionaries, as had happened elsewhere. Diebner saw the weakness of his position and agreed that the survival of the Institute depended on it having an authoritative Director. Accordingly he consented to Professor Heisenberg being invited to join the Institute as scientific adviser, travelling from Leipzig University to Berlin once a week. On these visits Heisenberg could intervene to forestall undesirable developments in research in Berlin and elsewhere36.

Professor Schumann’s address to the group emphasized the defence aspects of the enquiry. The German Reich being at war, it was of the greatest importance that Germany should be forewarned of all possible eventualities; this was the purpose of the technical appraisal they would be undertaking. Even a negative conclusion was valuable, for the military could then be reassured that no unpleasant surprises were in store.

The experimentalists were commissioned to undertake a variety of materials measurements in specified areas of research, while Heisenberg was given the written task:

“to consider whether, under the known circumstances of the characteristics of fission processes in uranium, a chain reaction is at all possible, and if so please commit your ideas to paper.”37

Whereas the majority of physicists were willing to affiliate with the Heereswaffenamt group, a large number was not prepared to relocate under one roof at Berlin-Dahlem, and thus from its inception the programme was structured with a Headquarters and three provincial satellites at Leipzig, Heidelberg and Hamburg. From about 1942 onwards there were other groups in Czechoslovakia, Germany and Austria, mainly SS who kept themselves aloof from the professors, but as early as 1941 the Reich Post Office nuclear project in Berlin had equipment for nuclear research including high voltage installations and a cyclotron.

Heisenberg completed his assignment within two months, and on 6 December 1939 he submitted his findings in the first of two pioneering papers, G-39 The Possibility of Obtaining Energy from Fission in Uranium38.

Throughout the Second World War the spectre of a German atom bomb haunted many people, but from the outset Heisenberg made no attempt to disguise the possibility that one could be built.

“If a chain reaction is possible, then the bomb is possible. Its intensity would depend on the rate of liberation of energy before the chain reaction collapsed,”

he advised the Heereswaffenamt in his paper. As part of the deliberate process to denigrate the German nuclear project, various historians have stated that the Germans appear not to have considered the question of the fast fission chain reaction. The official British UKAEA historian Margaret Gowing39 added for good measure that the critical size of the U²³⁵ bomb appeared not to have been investigated either. Piecemeal transcripts of the German physicists’ secretly tape-recorded private conversations in internment at Farm Hall, England, in 1945 were included by General Leslie Groves, head of the Manhattan Project, in his book40 published in 1962. Following the release of the full documentation by the London Public Record Office in February 1992, it became apparent that Groves had lied when reporting what was supposed to have been said in these conversations. Groves stated that the transcripts proved how Heisenberg

“had not thought of using the bomb designs we had used: ours took advantage of fast neutrons … the Germans thought they would have to drop a whole reactor.”

In fact Heisenberg was recorded in conversation as saying in 1945:

“I knew it could be done with U²³⁵ using fast neutrons. That’s why U²³⁵ alone can be used as an explosive. They can never make an explosive with slow neutrons, not even with the heavy water reactor.”

The purpose of Groves’ lie was to pervert history by proving falsely that there never could have been a German atom bomb because the top scientist did not know the principle. This had to be done because a small atom bomb actually had been built and tested by Germany.

In his pioneering paper of December 1939 Heisenberg spoke of “enrichment”. Natural uranium consists essentially of two isotopes. 99.3% of the material is U²³⁸. This isotope captures free neutrons in the uranium, and this is why natural uranium cannot explode. The ‘fissile’ isotope U²³⁵ exists in natural uranium in the proportion of 0.7%. If the ratios can be changed, and the major isotope physically reduced in the material, then neutrons will be more free to act on the U²³⁵ atoms. If the material is sufficiently rich in U²³⁵ atoms, say above 50%, then it can be arranged for an explosive chain reaction to occur, although even 7% will be sufficient for an explosion of some sort.

In his report Heisenberg explained:

“An increase in temperature results from enriching the U²³⁵ isotope. If the U²³⁵ were to be enriched sufficiently to obtain a temperature corresponding to a neutron energy of 3.5 million degrees C, … the mass for the release of all available fissile atoms at once, would be: R=10∏ l cms = 31.41 cms.

“This explosive transformation of the U²³⁵ atoms can only occur in almost pure U²³⁵, because the capture bands of the U²³⁸ isotopes, even when present in reduced quantities, still absorb the neutrons.”

This is not a formal expression of critical size, but merely a general statement for discussion based on the idea that a fast fission atom bomb is possible provided one can obtain huge amounts of the U²³⁵ isotope. In Heisenberg’s equation, the unknown element is l, the diffusion length or mean free path. This term is the mean distance travelled by a neutron between release from an atom at fission to absorption and fission in another U²³⁵ atom. It can be a variable: the American physicists, when first putting their minds to the problem, came up with estimates of critical mass ranging from Feyman’s 50 kilos to Robert Oppenheimer’s 100 kilos, and the Americans had far better tools at their disposal for making the measurements than Heisenberg.

The Implosion Method

In the implosion method of detonating an atom bomb, the bomb core is a subcritical mass surrounded by a uniform layer of high explosive. When the HE detonates, a massive uniform pressure of millions of pounds p.s.i. is created which compresses the core material to a supercritical density, thus causing an implosion. Obviously, since the fissile material is compressed into a much smaller volume, the diffusion length is much reduced. If it were to be reduced by a factor of three, i.e. uniformly compressed by the explosion to a third of its original volume, then the critical mass required for implosion is smaller by the power of nine. This might have put U²³⁵ within Germany’s capabilities with a stupendous investment and the help of the electrical giants. The diffusion length is a variable depending on the quality of material and other factors. The more efficient the implosive force for the compression, the less U²³⁵ bomb material needed.

A crucial question is whether Germany had an effective implosion fuse in 1939. It is claimed that they were close to it, and by 1941 had made such advances in the technology of implosion fuses that they were already working on an ultra-violet type. Sources allege that an efficient implosion fuse which could set off all detonators around the bomb sphere at the speed of light was invented by Prof. (Ing) Friedrich Lachner in Vienna pre-war and a model exhibited at the Wiener Technische Hochschule (University) to a gathering of home and foreign physicists at a lecture evening. Professor Lachner was a colleague of Professor Adolf Smekal and had obtained his inspiration for the work after hearing one of Smekal’s talks. Later Lachner joined Professor Stetter’s team of SS physicists at Innsbruck. This group was studying the possibilities of the plutonium bomb and allegedly the implosion fuse was perfected in the course of the work.

Knowledge of the existence of this fuse in 1939, if only at the design stage, would explain Heisenberg’s persisting concern about the U²³⁵enrichment question, since once the fuse had been developed it cut the bomb material required to a few kilos of U²³⁵. If explosives technologists could improve on a fourfold compression factor the amount of material would be reduced by substantially more.

Professor Carl-Friedrich von Weizsäcker explained:

“Certainly we made no attempt to build a bomb. This decision was made easy because we recognized the impossibility of manufacturing a bomb in Germany under war conditions. If people now say that we set out to avoid or obstruct the building of a bomb that is a dramatization, since we knew that we were not in a position to do so.”41

That is what we are supposed to believe. The more important aspect of Heisenberg’s G-39 paper, however, concentrated on what was to occupy him for the rest of the war, the atomic pile that never was and the saga of the heavy water moderator.

Reactor Theory

It has to be stressed that uranium work did not lie at the centre of Heisenberg’s interests. When confronted with the assignment, he succeeded in familiarizing himself with the semi-technological field and during the war came to be regarded as the leading expert in the German Uranium Project. If, as he seems to have claimed, he was proposing, initially at least, to be a saboteur of Nazi nuclear science, Heisenberg was in a unique position to direct the uranium work along a false path, since, as the acknowledged senior theoretical physicist, he had been entrusted with the task of formulating the theory from the outset and, having set the guidelines, continued to influence the experimental side of matters until Germany’s final surrender. At the end of 1939, having been cut off from most foreign literature since the outbreak of war, he was entering new territory in attempting to establish a theory for the working uranium pile or Uranbrenner. His summaries in the two papers dated 6 December 1939 and 29 February 1940 respectively are still accepted today as completely correct, or at least in so far as what they actually say. But there is a material omission in these two reports. One cannot at this late stage discount the possibility of an error in his theoretical workings, but all along one has the impression that Heisenberg did not want a working nuclear reactor, and it does not seem unreasonable to suspect that he deliberately drew a false conclusion on which he was to rely later.

In the preamble to the report, Heisenberg cited as his principal source of reactor theory an article published in the 9 June 1939 edition of the scientific periodical Die Naturwissenschaften 42 under the title Can Technical Use Be Made of the Energy Content of Nuclei? written by the physicist Dr Siegfried Flügge of the KWI for Physics at Berlin-Dahlem. Under a sub-title The Control of Chain Reactions Flügge had stated:

“The decisive question for the technical application of the mechanism is manifestly this: is it possible to slow the chain cascade? Adler and Halban (Nature, Vol 143, 1939, p.739) have entered the debate and suggested the addition of cadmium salts to the mixture beforehand. In the absence of cadmium, the reaction would soar straightaway to a stationary temperature of 100,000 degrees C.”

In the mentioned article Adler and Halban had warned:

“The danger that a system containing uranium in high concentration might explode once the chain starts is considerable.”

The idea of the instantaneous explosive chain reaction in a reactor is grounded in an error of theory caused by failing to take into the mathematical reckoning the small fraction of relatively long-lived neutrons which are emitted up to a minute after fission. What should have been done in the mathematical theory was to average the slowing down and diffusion time of the lifetime of the prompt neutrons liberated within a micro-second of fission added to the mean lifetime of the 0.75% of neutrons which emerge up to 80 seconds after fission occurs. That calculation would have shown that while neutron density increases exponentially with time, the stable Period of the Reactor is not “less than 1 second” but is about 54 seconds.

The delayed neutrons play the decisive role in the safe control of modern atomic energy plants and without them nuclear power reactors would not be feasible. Heisenberg may have been genuinely under a misapprehension. On the other hand, he may have realized that this would be a useful error to have in hand if he wanted to obstruct the development of a nuclear reactor.

In later reports he was never challenged when he relied on the argument as a reason for proceeding at slow ahead with the interminable low-level experiments which filled the next five years. Knowingly or not, Professor Heisenberg accepted that the Period of the Reactor was less than one second in length, after which the reactor blew up. After the war, in his reproduction report respecting the Haigerloch B8 experiment published in FIAT Review of German Science 1939–1945 Heisenberg acknowledged that his theory had been at fault, admitting:

“American work shows that the Period of the Reactor is substantially extended by the delayed emergence of a number of those neutrons liberated during the fission process.”

And in a report about the German project prepared by A. Weinberg and L. Nordheim for A. H. Compton on 8 November 1945 the authors were of the opinion that the importance of delayed neutrons for the stability of a nuclear reactor had probably not been considered. Even if Heisenberg knew all along, however, he could hardly say so in 1947. So, throughout the Second World War, Heisenberg believed, or let it be thought that he believed, that the Uranbrenner – the atomic pile for power – was impossible because the reactor would explode one sixth of a second after it went critical. He did not explain this fact in writing when setting down the theory originally, although one would think he must have informed his superiors at the Heereswaffenamt of his fears confidentially. To make some sense out of the fact that Heisenberg and the Uranium Project spent the war years performing interesting experiments of subreactor geometry, and obviously had no intention of actually bringing an experiment beyond the critical point since there was sufficient heavy water available in aggregate to moderate a working reactor by 1944 but no enthusiasm for doing so, Heisenberg must have convinced Hitler of the impossibility of building a working pile. Hitler did not want a nuclear reactor in any case because it was Jewish Physics. Probably he just waved a hand in dismissal, allowing Heisenberg and the reactor project to appear to be doing something useful to keep enemy Intelligence on the hop. That really is the only logical conclusion to be drawn from the manner in which the project was conducted.

The Basis of Reactor Design

The surest method of realizing energy production from the fissioning of uranium lay in enriching the U²³⁵ isotope, Heisenberg explained: the more the enrichment the smaller the reactor would be. If the proportion of theU²³⁵ isotope in the uranium material were to be enriched by 50%, from 0.7% to 1%, success was practically certain. However, such a proceeding was prohibitively expensive.

Natural uranium could be used in the reactor vessel in conjunction with another substance, a ‘moderator’, which slowed down the neutrons in the reaction without absorbing them. The deceleration increased the chances of a neutron finding a U²³⁵ isotope to fission. Ordinary water and paraffin were not suitable as a moderator, since, being rich in hydrogen atoms, they absorbed neutrons. On the other hand heavy water and very pure carbon satisfied the requirements. Slight impurities in them could spoil the reaction, however.

Heavy water (D²O, deuterium oxide) is four times more efficient at slowing neutrons than the purest graphite and thus a much smaller reactor is required. Surrounding the reactor vessel would be a ‘reflector’, a wall of material enclosing the core of a nuclear pile against which escaping neutrons are scattered back into the reaction. Heisenberg indicated that graphite blocks would be suitable for this.

He then described a number of possible reactor arrangements. The most important was a configuration three cubic metres in size consisting of 30 tons of pure carbon in the form of graphite and 25 tons of uranium oxide which, according to his calculations, would reach the critical point and supply energy. In the supplementary paper to G-39 of 29 February 1940 Heisenberg confessed to some misgivings regarding his design for a graphite reactor and this may have been prompted by Professor Harteck’s interest in it.

Professor Paul Harteck (1902–1985) had graduated in chemistry at the University of Vienna and at the age of 26 had obtained his PhD at the University of Berlin. He is credited with the discovery of parahydrogen.

In 1933 he studied nuclear physics at the Cavendish Laboratory and during this period was set the task of producing a quantity of heavy water, which he achieved by spending several weeks passing an electric current through a small electrolytic cell. The amount was minute in comparison with all the gallons of water used in the process. Later he would have charge of Germany’s heavy water production process. Following his return to Germany in 1934, Professor Harteck was appointed Director of the Institute of Physical Chemistry at Hamburg. He was a Nazi Party member and his team of five co-workers were known as “the Hamburg Bomb Group”.

Heisenberg had remarked that the uranium machine would shut down automatically at certain peaks of temperature, and then only resume when the temperature had fallen again. This would occur because of the expansion of metals on heating, resulting in a lowering of density and an alteration of the various cross-sections. This same increase in temperature would cause an increase in the width of the capture bands formed of U²³⁸ isotopes. This was due to the nuclear Doppler Effect. The widening of these U²³⁸ capture bands caused many more neutrons to be absorbed, resulting in a lessening of fissions until the chain reaction collapsed altogether.

In earlier conversations with Heisenberg, Professor Harteck had suggested that uranium and moderator should be segregated into a heterogeneous design more favourable for the production of an efficient reactor. When he read the mention in Heisenberg’s two pioneering papers of the problems of heat, Harteck realized that there was a better way of building a nuclear reactor altogether. If a pure carbon moderator was used at extremely low temperatures, the nuclear Doppler Effect would ensure that the width of the U²³⁸ capture bands would shrink and the reactor would produce no heat. If it did heat up, the chain reaction would collapse. Thus all the troublesome engineering arrangements inherent in an energy-producing reactor, such as heat transfer, core and fuel cooling and temperature control, would be obviated. We may infer from his obvious disinterest that Professor Heisenberg was not honestly in favour of building a working nuclear reactor at all, for this simple experimental zero-energy design would have been a good way to look at the problem of reactor stability. But he knew the terrible danger it presented. In his initial report he had observed:

“An extraordinarily intensive neutron and gamma radiation goes hand in hand with energy producion. Even in achieving only 10kW power, 10¹⁵ neutrons and gamma rays are created every second. The radiation is, therefore, 100,000 times greater than that produced in a large cyclotron. Even if a substantial amount of this radiation is absorbed in the core of the pile, nevertheless the working reactor would obviously require the provision of the most comprehensive biological shielding against radiation. This applies especially at the ‘switching-on’ of the machine, i.e. at criticality. At the moment when the temperature reaches the stationary value of 100°C, 10⁸ calories are used to produce heat leaving an excess of 5 to 10¹⁹ neutrons and gamma rays liberated.”

The very low temperature uranium pile would produce nothing but radioisotopes and the intensely radioactive decay products of nuclear fission. Radioisotopes do have modern applications in medicine, biochemistry, biology and industry, but Professor Harteck saw another use for them.

After the war Harteck admitted43 that his idea in proposing to build a sub-zero uranium pile was to obtain nuclear waste for use against the populations of enemy cities. This seems to have been the first time that radiological material was being seriously suggested for military purposes. Such weapons are not outlawed by international treaties, since they are not classified as chemical. The evidence suggests that Hitler was prepared to entertain radiological warfare to stave off defeat but might not have resorted to it early on in the war unless he thought it would guarantee him victory.

Professor Harteck set about building an experimental sub-critical pile immediately. His idea was simple. Dry ice sublimates slowly at a temperature of – 78°C and is as pure as one part in a million. Its oxygen atoms do not absorb neutrons in significant quantities at very low temperatures. Concluding that carbon dioxide ice was an ideal moderator for his proposed experiment, Harteck asked permission of the Heereswaffenamt to proceed and went ahead with it at once.

He had useful contacts with the firm of I G Farben, and on 8 April 1940 he induced the firm’s research director, Dr Herold, to make him a gift of a 15-tonne block of dry ice to be delivered at the end of May. The War Office agreed to supply a railway wagon to expedite the consignment from Merseburg to Hamburg, and Harteck wrote to Diebner asking for 300 kilos of uranium. This figure was on the low side, but Harteck thought that it was all that was available.

It must have been obvious that Harteck was expecting to perform his experiment within a week of receiving his 15-tonne block of dry ice. Diebner had only 150 kilos of uranium oxide at Berlin-Dahlem, but Heisenberg was waiting for a large delivery from the War Ministry and probably had a large hoard besides. Diebner promised Heisenberg that the large amount would arrive in June and requested him to settle privately with Harteck.

Heisenberg suggested to Harteck in a letter that he was exaggerating the urgency of his experiment, since there were a number of preparations to be made first:

“… of course if there is for any reason any urgency in your experiments, you can go first by all means. But I should like to suggest that for the time being you content yourself with just 100 kilograms.”

Heisenberg concluded in a very reasonable vein that he was quite prepared to let Diebner make the final decision. Harteck replied by return, emphasizing the obvious urgency, and begged Heisenberg to loan him from 20 May, for three weeks at the most, as much of his Leipzig stock as he possibly could allow. In the expectation that Heisenberg would relent, Harteck asked Dr Herold to delay shipping the ice until the last possible moment and spoke to Diebner twice to emphasize his need for a minimum supply of 600 kilos of uranium oxide.

At the end of May Diebner loaned him 50 kilos and Dr Riehl of the Auer Company brought him 135 kilos more. Heisenberg sent nothing. When the block of ice arrived at the beginning of June the experiment was doomed and the only useful information it yielded was criteria for the distribution of neutron density in certain arrangements of uranium oxide and dry ice44.

Heisenberg’s non-cooperation prevented Harteck from obtaining a figure for neutron multiplication. This would have enabled Harteck to calculate the quantity of materials he needed for a working pile. Both Harteck⁴⁴ and Wirtz45 made this point subsequently.

A Windfall of Uranium Oxide: Harteck Tries Again

There was no shortage of uranium oxide in German-occupied Europe. In May 1940 German forces arriving at Oolen in Belgium had discovered at the warehouses of the Union Minière Company over 1200 tons of uranium oxide and 1000 tons of other refined uranium metals. The British Government had known about this stock since early 1939 but had dropped a plan to purchase it outright and so remove it from proximity to Germany. The President of Union Minière, Edgar Sengier, appears to have made a purely business decision to leave the material for the Germans when they invaded so that his company would find favour with Hitler should he emerge victorious in the coming war in western Europe. Sengier then ordered the uranium mines at Katanga in the Belgian Congo flooded and had the mined ores shipped from Lobito to the United States. In October 1939 he transferred his offices to New York46.

The Germans controlled the Joachimstal mines in Czechoslovakia and thus held virtually all the uranium in Europe. There was in fact so much uranium in their hands that Professor Harteck set about planning his ambitious second experiment, a heterogeneous design consisting of 20 tonnes of uranium oxide in a lattice of shafts embedded throughout a 30-tonne block of dry ice. As soon as he announced his intention, he ran up against the determined opposition of Heisenberg, who argued that the experiment was so big that all Harteck would learn from it was a great deal about 20 tonnes of dirty uranium oxide and 30 tonnes of dry ice. Why Harteck thought that was something worth knowing he could not imagine. He expected that it would not work, however, or at least not unless Harteck sent the uranium oxide to a factory for purification first.

Harteck then came under growing pressure from other quarters, probably orchestrated by Heisenberg, and these argued that it was too extravagant for a first experiment to use 50 tonnes of materials to do the whole programme at once, while Heisenberg returned to the attack by remonstrating about the unprofessional approach to the experiment.

Bitterly Harteck was forced to concede defeat, refusing to accept their opinion. He resented Heisenberg in particular, commenting that, to his knowledge, Heisenberg had never contributed a single basic idea leading to the solution of the problems of nuclear fission: he found it inexplicable that a theoretical physicist who had never been involved in a large experimental venture could be appointed as leader of a technological enterprise. It was worse than merely poor judgment. Harteck attributed Germany’s failure to produce a nuclear weapon to the antagonistic attitude existing between the theoretical physicists and the experimentalists: the former considered the latter as beneath them, “a few egoists pushed the others aside”.47

What seems to have been Heisenberg’s real worry over Harteck’s proposed reactor was that since it operated at sub-zero temperatures it might be easier to stabilize it with control rods when it went critical: if this sort of primitive reactor worked, it would produce the nuclear waste which Harteck wanted to use in radiological weapons. Harteck felt sure that such a programme would have brought the war to a swift end in Hitler’s favour.

Was Professor Harteck serious about radioactive bombing? One must profess astonishment and credit him for having come clean on the matter after the war. Few others were honest. Considering how the dry-ice low-temperature reactor would have placed Germany’s nuclear programme on an entirely different footing, he stated:

“You must be thankful this didn’t occur. Not that an atomic bomb would have been made. But if you have a carbon dioxide reactor and you let it run for a certain time, the cubes or rods of uranium would have become highly radioactive. Much radioactive material could have been made which could have been thrown about. That would have been very bad⁴⁷.”