Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Ted Taylor by Kenneth W.
Ford on 1995 February 13,Niels Bohr Library & Archives, American Institute of Physics,College Park, MD USA,www.aip.org/history-programs/niels-bohr-library/oral-histories/23198
For multiple citations, "AIP" is the preferred abbreviation for the location.
Taylor's work on fission weapons at Los Alamos (included weapons boosted by thermonuclear reactions), and the relation of that work to John Wheeler's work on thermonuclear weapons in the 1950-53 period. Observations on John Wheeler, Conrad Longmire, Carson Mark, Marshall Rosenbluth, John Reitz, George Bell, Walter Goad, Preston Hammer, Edward Teller, Stan Ulam, Francoise Ulam, Jim Tuck, Nick Metropolis, Enrico Fermi, Norris Bradbury, John von Neumann, and others. Comments on early computers.
First, a very specific question: Do you know who was the main designer of the fission trigger for the Mike shot? Were you involved in that, or was it somebody else?
There were quite a few. Specifically, the Mike shot?
Any time there was any kind of a fission bomb proposed for testing or tested, I did some sort of quick and dirty estimates to get a sense of what the yield would be and whether to use plutonium or highly enriched uranium of both. But I was not the designer of the Mike trigger. Conrad Longmire was probably involved, and Carson Mark. I can't recall what the primary was, what it was like, and I haven't tried to think of that for a long, long time, but I'm surprised that I can't recall it. I think I worked on it, because I did work on the George primary. That was an experiment before the Ulam-Teller breakthrough.
That was for Greenhouse?
That was for the Greenhouse test, but it was a fission implosion of a cylinder that had at least one end exposed to a channel in which the radiation flowed and then compressed a very small secondary. That was a peculiar sort of fission explosive, the likes of which we'd never made before. So I spent a lot of time trying to figure out how to use the more standard predictive techniques we had for predicting the alpha—that is, the time constant of the explosion as a function of time C and the yield. I probably did that on the primary for Mike, but I was not working on the H bombs at all in the sense of doing any calculations of how things might work. I didn't pay any attention to things thermonuclear until a couple of years later when I got heavily involved in boosting of fission bombs. But never any work on H bombs as such—not because of any convictions. I do vaguely remember saying to myself and maybe to other people, "Wouldn't it be neat if it didn't work?" I can easily imagine having done that because I wasn't working on it myself; it had nothing to do with my success or failure personally.
One thing that made Matterhorn work two thousand miles away is the fact that the geometry of the device lent itself to the division of labor. Roughly speaking, there was the fission trigger, there was the flow of energy from the fission bomb to the thermonuclear material, and then there was the burning of the thermonuclear material. What I'm a little vague on is who the principal people were at Los Alamos, at the other end of this axis, that handled the fission trigger plus the energy flow, which was at Los Alamos, while the burning, the thermonuclear reaction, was at the Matterhorn.
I'm guessing that Conrad Longmire was involved directly in that and Marshall Rosenbluth. And possibly John Reitz and George Bell. There were some aspects of the coupling that I know George was involved in maybe later, and that was when we wanted to avoid excessive neutron heating of the secondary, and he was sort of T Division's expert on neutronics of that sort. Walter Goad may have done that too. I think the reason I don't recall what I did is that there was a lot of work that had to be done to make sure that the primary bomb was appropriate for getting fairly efficient radiation flow through it. I didn't know anything about that, so I think probably all that happened was I got the parameters that whoever did this chose and did my own thing in estimating the yield. But the creative part of that, I'll bet anything, was a combination of Conrad Longmire and Marshall Rosenbluth—and Teller, who always got his finger in the pie.
Switching gears to computers, this period, of course, '50 to '53, was in a sense the birth of the automatic computer, and I just wonder what your recollections were of the early computers—for instance, the MANIAC: When did it actually become a useful tool and did you use it? And card programmed calculators. What's your general recollection of computer power at Los Alamos in that period?
It's very simple and straightforward, and that was that just down the hall about five doors from my office was the computer part of the laboratory. This was before the MANIAC was even designed. The head of the computer group was Preston Hammer. I would go to him with a little piece of paper with some proposed new dimensions or new masses for an implosion calculation. This was within a couple of months of the time I got there, because somehow a sense of trust developed that I wouldn't tell him to do something stupid. He would then say fine, and he'd use the IBM machine—which was not an IBM 701 or an IBM 650. I don't know what it was called.
We called it a CPC.
I think you're right. That does ring a bell.
Card Programmed Calculator.
Yes, yes. Everything went in on cards, a big stack of them. Six weeks later Preston Hammer would turn up in Jack Smith's and my office and put on my desk a stack about a foot high of IBM listings, and then I started having my fun poring through that. So that from my perspective there was a sort of a black box and some people out there. I made a vow I would never learn how to program, because I saw even people like Marshall Rosenbluth spending hours and hours and hours doing something that I was not good at, didn't understand, and thought would be infinitely boring.
Did Preston Hammer also supervise the little army of young women who ran Marchant calculators?
Not directly, no. They were part of his group, but there was someone else, whose name I can't remember, who ran that group of women. In between was Bengt Carlson, who then took over from Preston Hammer when Hammer left (I forgot why Hammer left) So Bengt Carlson took over the job of running the computer department so far as the standard computers that we were getting periodically from IBM. The 650 sort of revolutionized what we were doing there, and then the 701. I remember the MANIAC development being totally separate from all that. It was Nick Metropolis' baby. He and Johnny von Neumann spent a lot of time on that, and the objective was pretty specifically to be able to do the coupled hydrodynamics and radiation flow necessary for H bombs. I think—I assume, I don't know this—that the similar specific answer to the question, "What is the ENIAC to do in Princeton?" was the same thing. [Slip of tongue. The Institute computer was not the ENIAC. KWF] Now this may have been something that flowed out of John Wheeler's group. I don't know. I was only vaguely, peripherally aware of these things so far as the computing was concerned, except that I used the output. That was sort of meat and potatoes of everything I did, but I had nothing to do with producing it.
My recollection is that the MANIAC in fact never worked well enough to be a useful tool. It was overtaken by better machines.
I couldn't quarrel with that. I can't resist saying, as a side thing about the MANIAC, that there was a specific time when it was turned on the first time, and that was very soon after Nick Metropolis came back from a trip. It happened that while he was on his trip there was a showing of "The Man in the White Suit," the movie, in Los Alamos, at the Los Alamos movie theater. Most of us had seen it, and there was a machine in there that made this thread from which the white suit was made that made these wonderful bubbling, popping sounds. Somebody, some wag—I don't know who it was—recorded it. I think it was Jim Tuck, actually. Yes, I think Jim Tuck did this. He recorded it and set it up so [it would play back] during this ceremony. I wasn't there when it happened, but the ceremony was supposed to be when Nick would come in and turn on this switch, which then started the whole thing, and what came out was this very loud "Man in the White Suit" chemical factory that made these wonderful sounds.
I hadn't heard that story. I have another question related to computing—a very specific event. I don't know if you remember it. One summer—I don't know which summer, whether it was '50 or '51—Fermi and Ulam got into an informal, friendly competition as to who could calculate the progress of thermonuclear burning quicker and better, Fermi by hand or Ulam and Everett using the machine. Fermi got this really gorgeous young woman to help . . .
Oh yes, I remember her.
...with her Marchant calculator. I think she later became Mrs. Ray Sawyer. She married someone at Santa Barbara.
Yeah. I'm trying to remember her name. She was a beautiful woman.
Do you think that would have been the summer of '50?
Yeah. Yeah. because by the summer of '51 these calculations which they were doing—under what conditions you might get runaway in the super—were no longer of interest. So it had to have been, and it was the summer. I should remember the woman's name. I don't know how it got into this, but I wound up periodically cutting her hair with a razor blade and the fuss over her and the whole idea of these young women working the Marchant and Friden calculators was sort of a big deal.
I remember we were amused because Fermi developed this enthusiasm for calculation.
Oh yes. And people used to kid him. I remember that subject of the nice looking computer women, which was mostly that one woman, being discussed in one of Fermi's hikes, which he would organize almost every weekend, either in the Sangre de Cristos across the valley or right behind Los Alamos. Those walks they were a high point for all of us, including him. Sort of a free-wheeling conversation about what was going on.
Well I have a question on that one. Do you have any recollection of who won the bet, Ulam-Everett or Fermi?
I don't remember. I remember the bet, and I have a feeling that I remember the bet because of what Fermi won, but I don't know. That is that it was another case of the traditional way of heating deuterium. But I can't vouch for that.
Ted, everybody and his brother has some memory or some thought about Ulam versus Teller and their relative roles in the great breakthrough idea. Do you have any particular recollections of that?
No, not in terms of recalling anything at that time. Nothing that has any bearing on what the two said to each other in that meeting in Edward's office where they sort of came out triumphant. I heard about this; I didn't see them coming out. I was on the floor below them and wasn't paying attention to that anyway. What I do remember, which bears on this, having thought about it later, is that Ulam described to me this idea of using the hydrodynamic energy from a fission bomb to squeeze a much bigger charge of highly enriched uranium to make a very big explosion, and to make a very big explosion with high efficiency compressing through a large number of critical masses something on the order of a 100 kilograms of uranium, which, if completely efficient, will then be 1.7—megatons.
I have a very clear picture of his describing that, and what I remember hearing soon after the breakthrough was that they had a discussion in which Ulam kept pressing for squeezing the secondary. Now whether he did that with the key perception that then the inverse Compton effect wouldn't drain the energy, that things would be much closer to equilibrium and that at these high densities you get a fast enough reaction rate and a high enough temperature rise so that it would be very efficient. I don't know who came up with that. The key point as it was described at that time was radiation implosion, which was already in both their minds, because that had happened in the Greenhouse experiment—not because anybody was proposing to use the radiation implosion; it was just a mechanism to squeeze some deuterium with some tritium and actually make a thermonuclear reaction.
The alternative to that was the booster, which also worked. My sense of things is that this direct compression is something that they sort of both saw at the same time, that compressing it was the way to go. Then the question was: How do you compress it? The subject had been brought up by Ulam, and what I've come to believe is that Teller said, "Oh, that's terrific, but let's use the radiation, not the hydrodynamics." And then everything sort of became clear. Now whether that's really what happened, I don't know. I suspect that nobody knows, including Teller, now.
[laughs] That's probably so.
Francoise Ulam is about as close as anybody alive to what might have happened. I think she had said repeatedly that basically the whole idea was Stan's.
There exists a letter that Ulam wrote to von Neumann in which he said, "I've just told Teller about this new idea and he likes it. It must be wrong." [laughs] Well, [two interesting] personalities. I don't know. I think for me [to say more] now would be speculating. I never did know exactly what happened, and I think, as I mentioned to you on the phone, I thought for a little while that the breakthrough idea was John Wheeler's, because he was the first person I heard describe it at a presentation in front of a blackboard in the T Division conference room.
He was talking about lots of what I think he called sausages, hooked up to each other so that a primary and a secondary and a tertiary (actually in effect an infinite number eventually) by having the radiation funnel down something that first imploded something small, then something bigger, than the radiation flowed down this tube, and what would go would be 10 stages maybe. Not necessarily 1-2-3. Now I don't know when that was, except that what I remember was that it was very soon after Teller and Ulam told people [their idea].
Spring of '51 probably.
Ted, you were right there in the middle of T Division and saw a lot of Teller. I'm curious to know to what extent you and others were aware of his general level of distress; the fact that he was almost constantly unhappy with Bradbury and so on.
I wasn't aware.
That didn't propagate to his co-workers?
No. Well, at least not to his young co-workers. What propagated to me about Teller was from Teller directly. I developed an enormous respect for him and high regard. The reason was that he was very encouraging to young people, and listened to what I was doing, and I got the impression it was the same with Marshall Rosenbluth and others. Very supportive. I had a couple of ideas that I talked to him about for some hours, because it just happened that coming back from an APS meeting on the west coast we wound up on the same airplane, and it was grounded for about 6 hours in Phoenix, and there was nothing else to do.
He and I were the only people from the lab who were there, so that's when I really got to know him. He asked what I was doing. I talked about working out some ways in which we could do some very small yield experiments related to the stockpile bomb, which was nominally 80 or 90 kilotons, depending on very high compressions, much higher than we'd ever seen before. And there was a question whether we'd really get them.
So this was a proposal for doing a series of what we wound up calling a Christmas tree set of tests. He caught hold of that immediately, and right after we got back, within a few days, he called me into his office and he said, "I've set up for you to talk to Bradbury and McDougall and Carson Mark—and I think Fermi was there. Well, I'm sure he was. I'm not so sure about von Neumann. So here, just a few months after I got to Los Alamos I was, at his invitation, spilling out this idea to all these fancy people, and I was just absolutely entranced by the whole thing.
So my feelings about Teller were very friendly, very good, very full of all kinds of admiration. I loved to hear him talk and give his talks, and then everything just sort of slid clear out as far as my feelings about him when the news of the Oppenheimer hearings broke around three years later. But up to that time I just thought the world of him. I kept hearing things about troubles he was having, but I don't think I was hearing those when I first got there. My first impression of him was that he was an extremely clever, nice person, full of good humor and warmth.
My reactions were the same. We've mentioned lithium deuteride, and you pointed out in your written comments that it was a completely separate idea from the new design idea. Do you have any recollection of who first suggested lithium deuteride and when that occurred?
I really don't know, but I think a somewhat educated guess would be that it may even have come up immediately, on the grounds that whereas anything even with the atomic number of lithium would quench—there are too many electrons. The electrons were stealing energy from the main reaction, so the more electrons, the worse it would be. But not in equilibrium. So I feel reasonably sure that before Mike was set off, but probably after its design was largely frozen, it was being seriously proposed that for real H bombs the thing to use was lithium hydride.
There had been a lot of calculation work with lithium 6 deuteride and highly enriched uranium in connection with the alarm clock, because that's how it worked. Lithium was a key component of the alarm clock. That was already being worked on when I got there. You know, this is coming back now. I'm sure that must have come up immediately, because it was the way that made the alarm clock work. Lothar Nordheim was working sort of full time on that. I think also Bob Richtmyer, and Teller was interested in that to some extent.
As far as I know, there has never been an alarm clock or anything like it built, but I think there is a consensus, or close to a consensus, that the first so-called H bomb of the Soviets was an alarm clock, or similar to it. Its layers are lithium 6 deuteride and uranium outside of fission bomb.
I only wrote down one additional question, which is completely open-ended, just to ask you if you would state any general recollections you have of John Wheeler—as a person, as a colleague—and what you remember from those Los Alamos days about him.
Well, I guess everything I remember about him was from those Los Alamos days, because I didn't really see him later—with the one exception I mentioned to you a while ago. I remember him, I can see him clearly, standing in front of a blackboard with a big lunch table between him and the audience at a T Division seminar talking about three things, on different occasions.
The first one, I believe, was his giving a report on progress on refining the liquid drop model of the nucleus and in particular making predictions about the behavior of even-even and odd-even nuclei regarding fission. The thing I remember from that most clearly is his intense facial expressions coupled with use of his hands and holding this nucleus out for all of us to see and having it start to wobble because a neutron had gotten in there and started making waves on the surface, and then splitting. He was wonderful doing that, at using his hands and expressing himself.
The second was when he described this variation ... He described a device in which, as I remember, he started describing it as a series of devices imploded by radiation from the preceding device—preceding in time. And then, well, it would be a small explosion that started and produced the conditions in the secondary for which one then got 10 times the yield or so as the first one. Then some multiplication factor for each one after that. Then he was speculating that that might just be made a continuous process in which you made the thing as long as you chose, and if you wanted a thousand megatons that's what you'd get. So what came through to me hearing what he was saying was that performance of the classical super had been achieved conceptually by a really different method.
The third was his describing a project that I picked up on later and actually spent a lot of time on. That was using the fact that an H bomb produced a lot of excess neutrons, and if you figured out how much each neutron cost or how much a mole of these neutrons cost, it was tiny compared to neutrons at Hanford. So he was proposing to outproduce Hanford in the production of plutonium by a huge factor by setting off H bombs deep in the ocean, surrounding the H bombs with an appropriate blanket that would then stimulate the capture of neutrons in uranium 238. These explosion products would then rise to the surface in a huge bubble that would oscillate and burst at the surface, and would then be caught by some huge sheets of plastic floating on the surface, which could then be connected with each other to keep the solution of this debris from the explosion in seawater getting more diluted. Then you had a big ship, a floating processing plant, that would extract chemically the plutonium from this stuff, seawater, at concentrations of a few parts per million. He called that whaling, and he went through his same thing with his hands, describing what this would all look like.
I remember we were quite fascinated by that. It just happened then that later on, probably stimulated by his recent idea, I started asking questions: Suppose we did that in a much more confined environment like the bottom of an ice cap. It was very easy to figure out how much ice you'd melt and what the concentration of the stuff would be. John von Neumann got extremely interested in that, and set up a number of meetings after he became a Commissioner of the AEC. He set up meetings which, among others, included one in which the Boy Scout who had been with Admiral Byrd in Antarctica was present, told us all about the ice in Antarctica and so on.
The idea was to make use of this same fact, that one output of an H bomb was very cheap neutrons. Plutonium was still really expensive. The ice sort of automatically, not only contained it, but also contained the explosion, producing a product which was water with, in a practical sense, a very high concentration of plutonium. The project died with von Neumann, and I got interested in other things and was getting ready to leave Los Alamos. The only time I remember spending any time with John Wheeler, aside from brief times at meetings, attending the APS meeting or something, was during a policy examination set up by an Air Force committee looking at what the United States should do after the Russians—the Soviets—started extremely vigorous testing again, in the late '60s.
When John Wheeler gave a briefing to this group, he said, "Well, one thing that we need to do is to increase the scale of production of tactical nuclear weapons to the point where we are aiming at having a million of 'em." His justification which I remember most of the people in the room were sort of aghast at—his justification, as he put it, sounded quite reasonable. He said everybody was saying before World War I started that everybody now had enough dynamite bombs to kill an inconceivable number of people. Then the war broke out, and within a few weeks all sides [had used up their] reserves of ammunition. [They had to] settle down making thousands of times more than anybody had ever imagined.
He said the same is going to happen with nuclear weapons, and if the Soviets recognize this and we don't, we'll be in for it. He was very concerned about Russian domination of the world. Beyond that, the last time I saw him was just a few months ago, briefly, at the big get together at the Institute for Advanced Study honoring Freeman Dyson. John was there and we talked for a while. I don't think he remembered me until we started talking, and then he said he did. I think he did actually, after we talked more. And he seemed fine.