Valentine L. Telegdi wonders (Physics Today, October 2000, page 25) why Leo Szilard abandoned his early patent applications on the linear accelerator, cyclotron, betatron and synchrocyclotron. Perhaps, he suggests, Szilard “lost interest in pursuing them,” or the patent examiners “may have raised questions of novelty” if they knew the work of Gustaf Ising or others.
This has been one of the mysteries of Szilard’s life. Why did he seemingly abandon so many astonishing and career-making inventions? Was it erratic and eccentric behavior, as is usually assumed?
In my talk at the Szilard Centenary in Budapest in 1998, 1 I argued just the opposite—that Szilard was a logical and determined man who has been misjudged. I published an account of his refrigeration inventions with Albert Einstein, 2 and I am grateful to Telegdi for this occasion to discuss Szilard’s accelerators.
The German patent examiner’s response to Szilard’s 1928 application on the linear accelerator still exists. Szilard gave a copy to Einstein, and it is preserved in the Einstein Archives. The examiner rejected the invention as unpatentable with this classic statement:
Patents can be given only for inventions that permit a commercial use. However, the submitted procedure apparently has only a scientific value. Whether, in accordance with the invention, any commercially useful material can be produced by accelerating artificially-produced positively-charged corpuscles, appears from our present knowledge ruled out. In the whole application, no hint is found that the applicant has produced, or can produce, such material. Obviously the yield would be so tiny, as with atomic disintegration from the natural alpha rays of radioactive substances, that even in the future the prospect of using the invention in commerce has the highest degree of improbability. 3
Priceless! What was Szilard to do? To prove the patent office wrong, he needed to build the devices. But without a patent, what company would support such a project? Szilard turned to his friend Dennis Gabor, as Szilard recalled in an unpublished letter:
It was my intention to build some of the machines and I turned over my patent applications to a colleague, Dr. D. Gabor, who at that time was with the Siemens Company and who thought that he might enlist the support of that company for this task. Nothing came of this, however. 4
Szilard could have stopped there, but he did not. Telegdi notes that in 1934, after fleeing Germany, Szilard filed an application in the UK on betatron and synchrocyclotron designs that were even more sophisticated. Telegdi suggests that this was Szilard’s last work on accelerators, but that is not so.
At Oxford University, while searching for an element that might sustain a nuclear chain reaction, Szilard collaborated with James Tuck to build such a betatron. Frederick Lindemann, director of Oxford’s Clarendon Laboratory, agreed to fund betatron construction, and plans were moving forward when history intervened. Donald Kerst, who built the first successful betatron, later called the Szilard–Tuck design the “most promising and most complete in technical detail” of early designs. Kerst believed it “would surely have succeeded were it not for the war in Europe.” 5
