科学哲学ニューズレター

No. 52, July 22, 2003

Book Review by S. Uchii, Edward Teller's Memoirs (Perseus, 2001)

Editor: Soshichi Uchii

　

Edward Teller's Memoirs

by Soshichi Uchii

Edward Teller (1908-), one of the inventors of the hydrogen bomb, published his memoirs a couple of years ago. Whether or not you like Teller, this book is a "must" for anyone interested in the history of nuclear weapons. It is very interesting and illuminating in many ways. In this brief review, I wish to focus just on one point: Teller's and Ulam's contribution to the breakthrough, which occurred in 1951, for the American Super program. Since we cannot believe everything Teller claims, I wish to contrast Teller's claims to Richard Rhodes' more balanced view, which appeared in 1995 (Simon and Shuster).

Now, the general background: On January 31, 1950, President Truman announced his decision on the hydrogen bomb, that this program should be pursued. And thus Teller began recuruiting members for this program at Los Alamos. The result was the so-called "Family Committee", including Los Alamos staff such as Carson Mark and Marshall Holloway, and including also new members such as J. A. Wheeler and others. However, Teller's initial plan, the so-called "classical Super" turned out to be unworkable on closer examinations. Teller needed desperately a new plan.

Teller's Version of the Facts

Teller tells us that it was December of 1950: a new idea emerged in response to Carson Mark's teasing him as regards the feasibility of Super.

I began a review of every idea that had gone into planning for a hydrogen bomb, looking furiously for a possible mistake or new idea. ...

In the midst of my mental review that afternoon, I suddenly realized that there was an easy way to avoid the difficulty I had described to Fermi: Particles carry energy in proportion to their number, but radiation carries energy in proportion to volume. If there were many particles and little volume, particles would carry most of the energy!

... Liquid deuterium is much more readily compressed, and such compression would make thermonuclear reactions possible after the energy transfers had reached equilibrium! The answer of how to design a thgermonuclear weapon was simple and obvious. (Teller 2001, 310-2)

For a substantial thermonuclear reaction to occur, the deuterium should be compressed. Within a few more minutes, I realized that the deuterium could indeed be compressed by the energy produced in a fission explosion so that radiation will be reabsorbed and fusion can occur.

Within an hour of Carson's derisive remarks, I knew how to move ahead--- ... thus, almost at once, the new plan appeared to be ready. (Teller 2001, 313-4)

Then, Teller claims that he discussed this idea with several people, von Neumann, de Hoffmann, Darol Froman (the head of the Family Committee), and Bradbury (the director of Los Alamos); but the latter two would not listen. Teller and Bradbury discussed the matter in the middle of January 1951, Teller recalls (Teller 2001, 315). And only after this, Ulam came to Teller's office, with his own idea.

Not long after my visit to Nevada, Stan Ulam came to my office. He announced that he had an idea: Use a fission explosion to compress the deuterium, and it would burn. His suggestion was far from original: Compression had been suggested by various people on innumerable occasions in the past. But this was the first time that I did not object to it. Stan then proceeded to describe how an atomic explosive should compress several enclosures of deuterium through hydrodynamic shock. His statement excluded my realization of why compression was important, and it also included details that were impractical. (Teller 2001, 316)

Thus even in 2000, Teller does not want to recognize any important contribution by Ulam.

Rhodes' Version of the Facts

However, Richard Rhodes tells us a quite different story.

In December 1950, Ulam thought of a way to increase implosion compression by orders of magnitude. He called the arrangement "hydrodynamic lensing." The fluid ("hydro") Ulam had in mind to power ("dynamic") his implosion system was the shock wave from an atomic bomb---primarily, he writes, the "enormous flux of neutrons." ...

To use fission for compression, Ulam realized, would require staging: one bomb, which came to be called a "primary", would set off a second, physically separate bomb, a "secondary." ... The idea was elegant but premature; Los Alamos had no immediate use for it. It stayed with Ulam into the new year. Sometime in January, he received a memorandum from Darol Froman ...

Shortly after responding to Froman's memorandum, Ulam writes, he saw a way to apply his "iterative" scheme to a thermonuclear. (Rhodes 1995, 462-3)

Ulam went to see Mark Carson but Carson was too busy to pay much attention to his new idea; so he visited Bradburry next, with more welcoming response. Then the next morning, Ulam spoke to Teller.

"For the first half an hour or so during our conversation," Ulam continues, "[Teller] did not want to accept this new possibility ...." But "after few hours," Teller "took up [Ulam's] suggestions, hesitantly at first," then "enthusiastically." Ulam mentions two reasons why Teller warmed to the staging idea: "He had seen ... the novel elements" and he had "found a parallel version, an alternative to what I had said, perhaps more convenient and generalized." (Rhodes 1995, 466)

We will ignore here the "novel elements"; what is crucial is "a parallel version", which means to use the radiation coming from the primary (fission bomb), rather than neutrons, in order to compress the thermonuclear secondary. The importance of this may be briefly explained as follows: If a bomb is with a relatively low efficiency (such as the Fatman), radiation is a rather minor product; but if a bomb has a high efficiency, i.e. smaller and higher-yield (as was actually the case with newer products at Los Alamos), energy comes out predominantly as radiation and the fireball becomes hotter. Thus Teller proposed an alternative (to Ulam's original idea) of using radiation; implosion by means of radiation may be faster and yield a longer-sustained compression of the fusion fuel. No doubt this was one of Teller's original contributions, and the result was justly called "Teller-Ulam configuration". Their proposal, worked out and supplemented by detailed calculations by Teller's collaborators, were reported as a joint paper by Teller and Ulam on March 9, 1951.

Later, in late March, Teller added another idea for obtaining a higher efficiency of thermonuclear reaction; that is, he proposed a second fission component within the secondary thermonuclear stage. In a word, a second fission explosion can be set at the center of fusion fuel, so that the outward push due to this second explosion and the initial implosion can come to equilibrium, sustaining a hot and highly compressed layer, thus burning deuterium more efficiently. The Mike shot (at Eniwetok, Nov. 1, 1952) was in fact designed by this idea. However, the construction of the Mike device was directed by Marshall Holloway, not by Edward Teller, because Norris Bradbury was wise enough to exclude erratic Teller from the team for thermonuclear project. Here is Los Alamos engineer Jacob Wechsler's witness:

Here's Edward marching along and every time you turned around he wanted to change things. It kept getting worse and worse and finally Marshall said, we've got to get him out of here. You can't tell Edward not to be there. Norris went to do it and Edward really brew his stack. That was when he said, well, either Holloway is going to run this or I'm going to. He wasn't running it anyhow, but he felt like he was. He wasn't in charge of it; he was just blowing out his ideas. Norris said, Edward, if you can't fit the mold, leave. (Rhodes 1995, 478)

From this incident alone, you can easily imagine Teller's characteristics as a colleague; back in those days Los Alamos was working for the implosion bomb (Fatman), Teller refused to work under Hans Bethe, and that was why the British team, including Klaus Fuchs, joined the implosion team.

However, Teller's personality aside, it is quite clear that Teller made a great contrinbution to the designing of Super, and this is acknowledged by everyone; and naturally, Rhodes agrees with this. But this does not preclude another question: who was responsible for making such a breakthrough? Teller expressly depreciate Ulam's proposal of compression. But Ulam's breakthrough was the combination of compression and staging; without the idea of staging, Teller's deeper knowledge of radiation would not have contributed to the new design. Of course, Teller might have come to the idea of staging someday, but the plain fact is that he had never dreamt of that until Ulam talked about it in January 1951.

Thus, it seems to me quite reasonable that Rhodes writes as follows:

Edward Teller seems to have found it intolerable that someone might share credit for the historic invention on which he had been working singlemindedly for almost ten years; he moved immediately to take over the technical breakthrough and make it his own. After he and Ulam issued their joint report, Francoise Ulam [Stan Ulam's wife] observes, "my impression is that from then on Teller pushed Stan aside and refused to deal with him any longer. He never met or talked with Stan meaningfully ever again. ..." (Carson Mark confirms Francoise Ulam's impression: "Ulam felt that he invented the new approach to the hydrogen bomb. Teller did't wish to recognize that. He couldn't bring himself to recognize it. He's taken occasion, almost every occasion he could, not every one, to deny that Ulam contributed anything. I think I know exactly what happened in the interaction of those two. Edward would violently disagree with what I would say. It would be much closer to Ulam's view of how it happened.") (Rhodes 1995, 471)

Teller's Memoirs Again

But to be fair, we still have to listen to what Teller wants to say. We already saw Teller's memoirs on this incident. But there are important sequelae. After Teller left Los Alamos, he made a campaign for a second weapons laboratory (eventually, that was successful and the Lawrence-Livermore Lab was born in September 1952). And, of course, the famous Oppenheimer hearing comes in April 1954; it is well known that Teller's testimony hurt both Oppenheimer and Teller himself, but I will skip all this, because Teller's apologetic memoirs on this point (Teller 2001, 369-384) do not seem to be persuasive at all. Anyway, these two incidents destroyed the good relationship between Teller and Los Alamos. Further, two reporters of Time magazine (James Shepley and Clay Blair) published The Hydrogen Bomb, in which they denounced all but Teller, based on interview with Teller; they praised Teller as singlehandedly responsible for building the bomb. This enhanced the anger among many people already angry with Teller, he recalls. Thus Teller decided to write an account of the hydrogen bomb "that would give credit to the many people who had worked on the program" (Teller 2001, 404).

That account appeared as "The Work of Many People" in Science (Feb. 1955). But we should carefully read Teller's comments on this paper. In a footnote on page 407, Teller remarks as follows:

In the article, I even went so far as to give Stan Ulam credit for suggesting compression, although I had come to that realization weeks before Stan discussed it with me. His contribution that day was that, by his interest, he freed me from Bradbury's dicutum not to pursue other designs. Still, I had no objection to the white lie in the article if it soothed ruffled feelings. Somewhat later, after it had become clear that Los Alamos intended to rewrite history, I was asked to sign the patent application and swear that Ulam and I had originated the design. Because months after the calculations had begun on the new device Ulam still didn't understsand my design and claimed it would never work, I felt justified in drawing the line at perjury. I refused to sign, and as a result, that patent has never been submitted. (Teller 2001, 407, note 6; my italics)

Teller in effect confessed, if the statement is true, that he lied in order to soothe ruffled feelings of many people; and it is clear that as of 2000, he wishes to insist on his own whole credit for the Teller-Ulam configuration. After all, Teller himself confirmed Rhodes' statement that "Edward Teller seems to have found it intolerable that someone might share credit for the historic invention"; so we may as well delete the word "seems to" from Rhodes' reserved statement!

Last modified Dec. 1, 2008. (c) Soshichi Uchii