Oral History Transcript — Dr. Marshall Rosenbluth
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.
See the catalog record for this interview and search for other interviews in our collection
Interview with Dr. Marshall Rosenbluth
Marshall Rosenbluth; August 11, 2003
ABSTRACT: Rosenbluth discusses his participation in the development and testing of hydrogen weapons, as well as the historical/political context within which he operated. Interviews includes discussion of his relationship with Edward Teller, his participation in events such as the "Atoms for Peace" Conference, and international exchange programs that took him to the Soviet Union. Numerous references to the Soviet science and their atomic program are made throughout the interview. Particular attention is paid to developments in plasma physics. Rosenbluth also discusses his time working in private industry at General Atomics.
Session I | Session II
Barth:I would like to start out with questions about your early scientific and personal interests. Could you give us some idea how you became interested in a career in physics?
I was always what I guess you would call a ďnerdyĒ child who was quite interested in a wide range of cultural and scientific things. My parents would tell me to go out and play baseball, and instead I would go down to the public library and read some books. So I did very heavy readingóIím now talking about the first six grades or so. Then I quickly found that I was interested in history, and I was interested in science and math. When I was growing up in New York I went to Stuyvesant High School, which was sort of a scientific public school for which you got in by entrance exams. I found that extremely stimulating. It was a very bright group of students. I had my circle of friends. This was right at the beginning of the war. In 1942, I graduated, and I hadnít really thought much about what my career would be at that point. In fact, I would say I didnít know there was such a thing as a physicist even by the time I graduated from high school. Indeed, at that point, the job opportunities for physicists were almost nil. They were extremely low paid academic jobs. I went off then to college, to Harvard, actually thinking more likely I would become a historian.
But a couple of things happened in the first year. I studied calculus for the first time, which to me was an amazingly empowering experience which I could really see how you could understand all sorts of things, and I decided that chemistry and biology just had too much memory for me to be interested. Physics was very easy. I should say, when I took high school physics, which was without calculus, I found it pretty boring. In some sense Iím not a well-rounded physicist. I donít particularly like doing hands-on experiments myself, though I do spend quite a bit of time talking to experimentalists trying to understand what theyíre doing. So then in fact I majored in physics at Harvard, and when I got out in í44 one strange incident was that it turned out that to major in physics you had to be in the School of Fine Arts or something and for that you had to have Greek or Latin, which I hadnít studied, so they gave me an engineering degree instead Actually, the form is in Latin so I canít even read it. But anyway, then I went off in the Navy for a couple of years.
Barth:This was still during the war?
Rosenbluth:During the war, yes, in í44. I went off in the Navy at the beginning of í44 and was in until the end of the war, and a couple of months later, until I got demobilized in June of í46. During the war I was a radar technician, radar and sonar, so I did actually learn some experimental physics doing that, which I found quite interesting, repairing gear on ships and so on.
Barth:You were deployed in the theater?
Rosenbluth:Yes, but I never got really beyond the fringes. Only in the Pacific, because by the time I was trained the war in Europe was pretty well over. So in any event, immediately becoming demobilized, I went into graduate school at the University of Chicago. I guess thatís when my real physics career got started.
Barth:Maybe before we go to Chicago, do any particular individuals come to mind that shaped you along the way? Does anybody come to mind who played a particular strong role in supporting your interests in physics?
Rosenbluth:Well, I would say it was more my fellow students who also had a strong interest.
Barth:Anybody we might know from your fellow students who became very active in the field?
Rosenbluth:Well, yes, there are a couple. There was Rolf Landauer from IBM. Peter Lax, who is at NYU. Well, I think about half a dozen of my classmates have since become National Academy Members. So it was a very bright group and we learned from each other. I liked the teachers but I canít say that I found any of them, per se, totally inspiring.
Barth:So there was nobody at Harvard?
Rosenbluth:Well, at Harvard there were a couple who I liked and encouraged me. I wouldnít say that I learned—There was Professor [Edwin C.] Kemble, who taught the first year Quantum Mechanics. He was very encouraging. I was quite young and he would encourage me; if I was getting discouraged to stick with it. There was Saunders Mac Lane, who was in the Math Department who taught me this freshman calculus course which I said was sort of my epiphany. Most of the faculty there during the war, the regular faculty, was basically off at war, so they had a lot of substitute people. In fact, I spent most of my courses in History and English literature and politics and took the core of physics courses and math.
Barth:You mentioned that you were very interested in history and a wide diversity of other subjects. But at one point or another it seems that most of your energy funneled into fusion and plasma physics.
Rosenbluth:Yes. Well, partly thinking what my career might be, I decided I couldnít really understand what I would do as a historian. I mean, you have to write books, and it seemed to me an awful lot of good books had already been written. Already, at that time, I was seeing some sort of revisionist books which I thought were sort of crazy, just people trying to get a new and improbable twist on history, so my take was I really enjoyed reading about it and it was fascinating, but I didnít quite know what I would do about it. Also, I may be basically too lazy to write books. Writing physics articles was more fun. What Iíve enjoyed doing as a physicist, and what I have done throughout my career, is find some problem and work on it for a few months and finally find the solution or a road set for the solution and you would know that just as you would build on previous works that other people would build on your work, so you were part of a continuous chain of knowledge, which was increasing, rather than sort of getting off into perhaps interesting sidelines or different unverifiable points of view about, say, historical topics. And of course, the mathematics content obviously appealed to me.
Barth:Maybe you can talk about your time at Chicago, your Ph.D. work. You worked with Edward Teller?
Rosenbluth:Edward Teller was my supervisor. I certainly learned a lot from the faculty there but, again, I probably learned more from the other students than from the faculty. Youíll be seeing Murph Goldberger, youíve seen Dick Garwin, or will be seeing him. We were fellow students. Also the first collaborative work I did was with [C. N.] Yang and [T. D.] Lee.
Rosenbluth:Yes. Actually, I guess Teller had first sent me a problem of trying to calculate what would happen when a negative mu-meson got absorbed in a nucleus what kind of nuclear temperatures might be excited and what the spectrum of neutrons you might see outside the nucleus would be. I did the work, he really just suggested the problem. But he was, I always found, very generous with credit. I mean, I wanted to put his name on the paper and he wouldnít hear of it. From then on the next step was that I had noticed, of course, that this problem was similar to ordinary beta decay. At that time the mu-meson decay was just being observed as well as the pi-meson decay, and it occurred to me that these could all be related phenomena and I wasnít sophisticated enough to really— you know, there are lots of divergent integrals and things. I wasnít sophisticated enough to know how to really evaluate those, so I made a rough estimate. I was very friendly with Frank Yang and I talked to him. T.D. Lee had just arrived from China, and we had a very lively three-way collaboration for six months or a year. I certainly learned a vast amount from that.
Barth:At this time you seemed to be a straight forward theoretical particle physicist.
Rosenbluth:Yes, I started out as a particle physicist.
Barth:Was this one of your major interests at this time?
Rosenbluth:At that time, yes, it was one of my major interests.
Barth:Did you find the meson topic yourself? Your dissertation was on mesons and the Yang and Lee collaboration was also on mesons. For the first year, in í49, all your papers were meson papers.
Rosenbluth:Yes. Well, this was what was in the experimental era then. It was the era when people discovered there were pi-mesons and mu-mesons and all sorts of new experiments were being done after the war. New cyclotrons were being built. At that time, and maybe even now, particle physics was considered the keystone, the center of physics, and certainly the center of physics interest in Chicago, though I did also take some courses with Maria Mayer.
Barth:On nuclear physics?
Rosenbluth:No, actually on statistical mechanics. So I never actually got too much into either solid state physics or nuclear physics per se.
Barth:There seems to be a cultural gap between these communities. When you talk to solid state physicists they have a very different approach to doing physics than high energy physicists. So itís a really interesting perspective to see how different these various physicists operate.
Rosenbluth:Yes. And of course, unfortunately itís growing more different over the years. People have gotten more specialized.
Barth:Is there anybody in Chicago, besides your student colleagues, Dick Garvin and Murph Goldberg, who comes to mind?
Rosenbluth:Well, of course, there was Yang and Lee.
Barth:They were then on the faculty already?
Rosenbluth:No. In fact, it was a wonder to me because they had just arrived. They had been stuck in Southwest China during the war under terrible conditions. And particularly Yang had gotten training such that he was clearly the most sophisticated, I would say, student there. Many of the others had worked on the Manhattan Project. There were only two or three sort of ďvirginsĒ like me who had really not been doing war work and were just coming out of college or the Navy.
Barth:Looking at your date of birth, you would have been too young for the Manhattan work, wouldnít you?
Rosenbluth:Yes, I think so. I was a year or so too young.
Barth:Could you fill us in a little bit about how it was working with Edward Teller at this point?
Rosenbluth:Thatís a complicated question. I liked Edward very much. He was an inspiring person. He was easy to talk with for brief periods. He would have ideas and then would switch his interest to something else. I think even from the time I started at Chicago his real interest was H-bombs, so he paid somewhat fleeting attention to what I was doing on my thesis.
Barth:Did he suggest the thesis topic or were you permitted to look for your own topic at this time?
Rosenbluth:I looked for my own topic. It was very academic and a laborious waste of time, I think. But anyway, I learned a lot of the techniques. Then there was [Enrico] Fermi, with whom I had some interactions. Not too close, but certainly some revealing conversations with him. And I got to know [Subrahmanyan] Chandrasekhar a little bit. Probably astrophysics is close enough to plasma physics, and that probably sparked my early interest in the high density, high temperature materials. The faculty member who was probably the most help on my thesis was Gregor Wentzel. He was Swiss and a very nice, very sweet man, a field theorist. He wrote what at least at that time was the Bible of quantum field theory. I think he has a couple of children who are still physicists; Iím not sure where. Also, there was Maria Mayer, I enjoyed talking to her. So I had a pretty fair amount of faculty contact, and the courses were generally very good I thought.
Barth:These were very unsettled and unsettling times. Were you able to focus on physics? How strong was your interest in politics at this time?
Rosenbluth:Not terribly strong.
Rosenbluth:I was still quite interested. It was the immediate post-war period, and politically I was a pretty convinced New Dealer. The student politics at Chicago was pretty well run by communists, I would say. I was Vice President of the Students for Democratic Action, which was a left wing but anticommunist political group. So I was moderately active, but I didnít spend very much time at it.
Barth:In other words, concerns about the Soviet Union didnít play a major role at this time for your motivation to go into particular areas of physics?
Rosenbluth:No, it did later on.
Barth:This only came essentially after the first Soviet nuclear test Joe-1 in 1949?
Rosenbluth:Yes, in that period.
Barth:Letís move from Chicago to Los Alamos. You worked at Los Alamos for the summer of 1948 and 1949. Could you fill us in on this experience?
Rosenbluth:Teller, with whom I was quite friendly, was also really primarily interested in his duties at Los Alamos. He asked me if I would like to come work there in summers. As I mentioned, I was beginning to be sort of worried by the communist threats and I thought the work sounded interesting physically and intellectually. So I did go there for the summers and found it quite interesting. I was a little shocked in that just about all the senior people, all the physicists that I ever heard of had left except Teller, who came back. So it wasnít the first team by any means. Which in a way was good for me, because a junior person could make a sizable contribution.
Barth:It was about this time that Teller wrote his paper, ďBack to the Laboratories,Ē where he basically said, ďWeíve played enough with mesons. Itís time to go back to the real work.Ē Were you part of this wave of coming back?
Rosenbluth:Probably. Iím not familiar with this paper. But I probably read it and Iíve forgotten it. He was somewhat persuasive. On the other hand, in 1949, when I had actually gotten my degree, I went off to Stanford despite his urgings that I stay at Los Alamos.
Barth:So he suggested ďPlease stay with us, we need theoreticiansĒ?
Barth:And you went to Stanford instead of staying at Los Alamos?
Barth:Could you explain what the motivation was behind this move?
Rosenbluth:The motivation was that particle physics was still the center of the physics universe, and Stanford was just building their first linear collider, their linear accelerator, which seemed to offer exciting opportunities. I had also been stationed in California for a bit during the war. In fact, I was on a Navy base which many years later became the site of Livermore Laboratory. I was probably the first physicist at Livermore. So that was an appeal. Leonard Schiff was there, and he was a very nice mentor. This happened to be the time when [Richard] Feynman was just coming out with his seminal papers, particularly the Physical Review paper with the Feynman diagrams and told you how to calculate everything, which I spent a month or so studying. Itís probably the only paper I ever studied that carefully, and I absorbed it quite well, and it seemed to me the obvious application was to use those techniques to see what happened when these fast electrons they were creating would hit protons. So really elementary particle interactions. I worked out all the gory algebra of that. Some of the more general principles are useful to this day. The form factors of the nucleons, the details of the mesonic interactions and the ones that supersede them. It was well recognized and satisfying work, which is basically what I accomplished at Stanford.
Barth:But somehow high energy physics didnít really seem to have satisfied your curiosity enough to stay in this field, and you moved away from it over the years. Is this a consequence of your work at Los Alamos from 1950 to 1956?
Rosenbluth:Pretty much. There was Joe-1, there was the [Klaus] Fuchs incident. Actually, parenthetically, there was one of my fellow students at Chicago, Ted Hall, who I was quite certain had been a communist spy at Los Alamos and he worried me. So I knew there had been a lot of spying done at Los Alamos. And knowing the nature of Stalin and [Lavrenty] Beria and so on, I couldnít imagine that they werenít going full bore on the H-bomb. We were wondering whether he could do it or not. So I really felt it was a duty to go back and participate at that time in the weapons program. I remained motivated that way until really after Stalinís death [March 5, 1953].
Barth:You left Los Alamos in 1956?
Rosenbluth:I think í56.
Barth:This is going to be a whole chapter of our conversation, your Los Alamos work. I wanted to ask you about your recollections of the first Soviet nuclear test, Joe 1, for example. Your major reaction was, ďItís time to go back to the labsĒ?
Rosenbluth:Yes. Well, the combination of Joe 1 and the spy revelations. I was convinced that in a totalitarian country like Russia, the scientists werenít going to have their choice of whether they should be doing high energy physics or weapons physics. They all would be doing weapons physics. And I knew from my readings that at least on the physics side, they had extremely smart people.
Barth:Who comes to mind? Physicists such as Igor Tamm and Igor Kurchatov?
Rosenbluth:And [Yakov] Zeldovich and [Lev A.] Artsimovich.
Barth:And these scientists were known to young physicists in the US at this point, from reading their work?
Rosenbluth:Right. Of course not their weapons work.
Barth:Sure. Maybe we can move into the 1950s a little bit and go into some detail concerning the work you did at Los Alamos. Maybe you can give us some general remarks about your six years at Los Alamos from 1950 to 1956. You left Los Alamos in 1956 to work for General Atomics. At one point during your six years at Los Alamos you must have made a switch towards [non-weapon related work].
Rosenbluth:During the last couple of years at Los Alamos actually I sort of switched away from the weapons program and went into plasma physics.
Barth:There was a plasma physics group at Los Alamos at this point?
Rosenbluth:There was a small one, yes. Jim Tuck was the boss. He was a holdover British participant from the Manhattan Project. And he had a pinch device. I guess, Britain, Russia, and the US all had their own secret pinch devices, and at that point, nobody really understood how they worked or what scaling was. Then the first thing I did in plasma physics was work out the implosion times, the voltages you would need and so on. And I found this was very interesting because I was quite interested in electrodynamics and hydrodynamics, statistical mechanics, the instability theory, and all these things that came into play.
Barth:And you met Tuck as part of the theory group at Los Alamos?
Barth:So thereís a theory group at Los Alamos in the Ď1950s of 10, to 20 people?
Rosenbluth:Something like that. Maybe a little more. Tuck was probably in an experimental group. But there was enough intermingling.
Barth:Like seminars, and one would see each other?
Barth:And he introduced you to the pinch work and suggested, ďHereís something that is decidedly not weapons relatedĒ?
Rosenbluth:Yes. Well, Iím not sure he would put it that way. But, ďWhich is very interesting, which I need help on.Ē Dick Garwin had also gotten involved at that time.
Barth:Weíll come back to this in a minute. You came to Los Alamos in 1950 for the six year period. You had done already two summers there so you knew some of the laboratoryís workings and you moved then right into the theory group?
Rosenbluth:Yes, thatís right.
Barth:And this came with a contract or assignment so to speak, ďHere, please help us with the booster or help us with the preparation for the first super testsĒ?
Rosenbluth:Yes. I mean, it wasnít the specific assignment. Carson Mark was the voice of the theory group. He is a very nice man and very, very helpful. He tried to steer people in a non-authoritarian way.
Barth:At Los Alamos you worked on DT reactions. Was it well known at that DT fuel would boost the yield of a nuclear weapon?
Rosenbluth:Well, no. It was certainly recognized as a possibility. But there were all sorts of physics questions which people didnít really know the answer to.
Barth:Who comes to mind? Who was in the group? You mentioned Garwin. Did Gamow or Ulam or Teller participate in discussions on a regular basis?
Rosenbluth:In the case of the booster we were mostly talking with Teller and communicating with Carson [Mark], I guess, to spread the word.
Barth:So itís not that there was a whole group constantly working together to work on these issues?
Rosenbluth:No. Of course, beyond Carson [Mark] there were all these engineering groups that had to actually make the thing in a usable form. But there are a lot of complicated physics issues when the matter interface between this heavy fissioning metal and the much lighter deuterium gas starts to interact and radiation flows and hydrodynamics flows and the interface goes unstable. Itís quite a delicate problem— I recognized right from the beginning that the stability of the interface was the main problem— that was probably recognized even before I got into it, but it became clear that was the main worry, and as far as I can tell, itís still the main worry with detailed design.
Barth:In simple terms, this means that you need homogeneous shockwaves, so to speak, but thereís no valve where the thermal nuclear fuel can fly out?
Rosenbluth:Yes. Or mix up with the heavy metal and thereby not be able to raise temperature.
Barth:So you worked on the booster principal. Obviously much of this work is still classified.
Barth:Is there anything you can still tell us about some of the details?
Rosenbluth:Well, thatís the problem. I donít know exactly what remains classfied — Well first of all, Iíve forgotten most of it. And second of all, Iím not sure what remains classified. I donít know what remains sensibly classified.
Barth:Were you present at the GEORGE shot?
Rosenbluth:No. The GEORGE shot was, of course, different from the booster. That was also something I was involved in. That was a test to conduct radiation from a fission bomb down a channel to a little pocket of DT on the outside, to watch the DT burning, but not particularly for it to interact with the fission bomb.
Barth:So it was a separate system?
Rosenbluth:Yes, where you had to transport. And it was very well set up for diagnostics because there were mile-long vacuum pipes where the neutrons and charged particles produced by the DT reaction, would fly. This was on an atoll where these products would fly to a different island and be analyzed there. And nobody knew what was going to happen. We had calculated. But I would say that maybe Tellerís greatest contribution was ó he had many great contributions ó in sort of persisting and proposing and persisting in this experiment even though it wasnít clearly on the path to [an H-bomb]. It was on a path to exploring physics involved in an H-bomb, but nobody had a clear design or idea how it might work out.
Barth:It seems then that Teller pushed for this experiment to have a better grasp of the fundamental physics underlying DT burning.
Rosenbluth:DT burning, yes. And the radiation transfers and things like that.
Barth:Does this imply that the experiment was not actually that necessary for a weapons program?
Rosenbluth:Well, my viewpoint is that it wasnít on a direct line to a weapons program. But in my viewpoint it played a key role in getting everybody, all those working on it, knowledgeable in all the detailed physics. And at least as I analyze the thinking that was going on, I donít see how they would have gotten to their solution without having this test. But I think Tellerís genius was in realizing that collecting some what would be certainly relevant physics information along the way was worth doing, even if you didnít see exactly how it lead to the final design.
Barth:He mentioned your name repeatedly in his paper, ďThe Work of Many People.Ē
Rosenbluth:I read it many years ago.
Barth:Teller also mentioned some of his other students at this time who were working at Los Alamos: Harris Mayer, and John Reitz.
Rosenbluth:Yes. I think Harris Mayer is still at Los Alamos. Yes, I guess he is a few years older than I. He is also a theoretician.
Barth:Also a theoretician?
Rosenbluth:Also a theoretician.
Barth:And John Writes?
Rosenbluth:John Writes, yes. He, John Reitz, was a student at Chicago who went to work for Ford Motor Company as part of their research laboratory.
Barth:Was Robert Richtmyer also in the theory division at this point?
Rosenbluth:Yes, theory and computation.
Barth:And Lothar Nordheim?
Barth:And John Wheeler, of course, and Conrad Longmire.
Rosenbluth:Yes. Conrad still lives up in Santa Barbara.
Barth:To give you a quote from Tellerís paper: ďRosenbluth eventually became one of the key men who carried the calculations to their successful conclusion.Ē Would you like to elaborate on this?
Rosenbluth:Oh, heís a smart man! [Laughs] You know, I think I did a lot of the detailed calculations.
Barth:Which is, of course, the time before fast computers.
Rosenbluth:Yes, they were largely done on Marchant [and Friden desk calculator] machines.
Barth:So you did the actual calculations on these machines? Or did you rely on the human computers at Los Alamos?
Rosenbluth:I relied pretty much on women working on the Marchants. They had some IBM punch card machines.
Barth:How did this calculation process look like?
Rosenbluth:I would prepare a spread sheet and they would do the calculations to fill it in.
Barth:They would punch in a couple of numbers and see what the results are?
Barth:And this, of course, was very tedious.
Barth:The theory group would design these kinds of equations and ask for this feedback on the calculations?
Rosenbluth:Yes. Later on in the early Ď50s these electronic computers started to come on line. There was the computer at Princeton at the Institute for Advanced Study. The driving force behind all these computers was John Von Neumann at the Institute. People informally called the machine there the JOHNIAC. That wasnít its official name. Then there was more or less a carbon copy built at Los Alamos, which was the MANIAC, which stands I think for Mathematical Analyzer and Numerical Integrator And Calculator. And neither of these two machines actually got finished quite in time to do the definitive MIKE calculations. MIKE was the first super shot. Of course, these being booster shots. So those were mostly done on the Marchants. And I guess six months before the actual MIKE shot, my previous wife and I went to the Bureau of Standards, which had the first real electronic computer machine up and running, and we took over the midnight shift for that for a couple of months and did much more detailed calculations which all came out very close to the cruder estimates. Partly, of course, the design was very over-designed because of the uncertainties: much more material than would have been needed. It was in a very high degree of symmetry, so it was possible to do quite good, almost analytic calculations. And Dick Garwin was involved in those too. But anyway, we actually went to the [the National Bureau of Standards ďStandards Eastern Automatic ComputerĒ] SEAC and we found similar results. This was I guess maybe six months before the actual MIKE shot. At least the theorists, were all very confident that the physics was well enough understood and that it would work as it was supposed to do. And indeed it did.
Barth:Were you actually present at the MIKE shot?
Barth:After receiving the MIKE results, would you then begin to work on the BRAVO shot?
Rosenbluth:Yes. Well, I had already been doing some work for the BRAVO shot. For the MIKE shot, only a few people went out. Even Teller didnít go out, I think. So a group of us was sitting at the Los Alamos Directorís Office at the time of the shot, and the Navy Commander clamped down a blackout restriction, so no news could be sent back. But Teller was sitting at a seismograph in Berkeley and he was able to detect what had gone on and sent us a message saying something like, ďItís a 10 pound baby boy.Ē
Barth:Thatís a famous story. Richard Rhodes actually mentioned this too in The Making of the Atomic Bomb.
Barth:You were actually close to the BRAVO explosion. You were on a ship not too far away and you also received some radiation.
Rosenbluth:Yes, quite a bit actually. Who knows, maybe thatís responsible for my present condition, but I doubt it. There we made a physics mistake — I say ďwe,Ē because Iím not sure who was responsible. We overlooked the lithium 7 (n, 2n) cross section, which greatly enhanced the yield over what we had calculated. As I said, I never learned much nuclear physics and that was my reward for having neglected it. But Iím not even sure whether it was well known in the physics community.
Barth:BRAVO was a big surprise to everyone involved. Nobody expected this high yield?
Rosenbluth:Thatís right. And certainly nobody inquired about whether the lithium 7 (n, 2n) cross section had been included. But it became apparent within a few days that thatís what had happened.
Barth:What was your reaction when you saw the BRAVO explosion? Were you surprised, ďGee, this looks bigger than expectedĒ?
Rosenbluth:Iím not sure whether I could have told the difference between 10 and 20 megatons. We were on a ship maybe 30 miles away. Of course, the other thing that happened was the wind shifted so the fallout came more over the ship than it was supposed to have done. It was very frightening, this multi-colored roiling around thing up in the sky. The image that came to my mind was a diseased brain up there. It was very scary. They locked us all below decks because of the fallout. It was probably 130 or something degrees, so it was extremely uncomfortable. But they estimated that I got about 10 rem out of it.
Barth:10 rem, thatís quite a few x-rays.
Rosenbluth:Oh yeah. Well, 400, over 300, is an immediately fatal dose.
Barth:So you had your radiation badge with you, and everybody on the ship got about a similar [dose]?
Barth:Did this influence you at all to move to a less dangerous occupation?
Rosenbluth:No, I wouldnít say so. As I said, my politics sort of changed. I felt by that time that the H-bomb problem was solved. That there was certainly a lot of work to make them smaller and cheaper.
Barth:Ted Taylorís work essentially?
Rosenbluth:Iím not sure if it was even Ted Taylorís. But to make them, to use less explosives.
Barth:Get the pits smaller?
Rosenbluth:Yes, get the pits smaller. And then, of course, within a few years after that missiles came in. But it seemed to me these were probably— you know, it was a lot of work but it probably wasnít a make or break phenomenon. Our knowledge of H-bombs was such that there was no way the Russians or anybody else was going to get qualitatively ahead of us in the matter. And so I felt that from that point of view it was a solved problem. And also I was eager to get out of the weapons business. After Stalin died and [Stalinís infamous head of the NKVD Lavrenti Pavlovich] Beria had been shot I thought, and still think, that there were much better opportunities to really wind down the nuclear arms race, instead of just letting it drift up ahead on its own inertia, which is basically what we and the Russians did.
Barth:Was your position shared by others in the theory group?
Rosenbluth:Oh, yes, by quite a lot of people. Carson Mark, for example.
Barth:But he stayed on in Los Alamos?
Rosenbluth:He stayed on, yes.
Barth:After BRAVO you worked two more years at Los Alamos.
Rosenbluth:He stayed on, yes.
Barth:After BRAVO you worked two more years at Los Alamos. What did you focus on during that period from 1954 to 1956?
Rosenbluth:Well, one thing was statistical mechanics. This actually goes back to the time after the MIKE shot in 1953. I did this work on statistical mechanics with the Monte Carlo, which I collaborated with Teller on. Iíve actually written up a talk on that, which I donít know who has the copyright. Probably, I think, Los Alamos.
Barth:Was it at this point a very new approach to doing physics?
Rosenbluth:Yes. Yes, it was. Largely because computers were just becoming available.
Barth:So basically the Monte Carlo we use today in high energy physics and so on is based on your early work in the Ď50s?
Rosenbluth:Well, itís related to it, certainly. I mean, even before I got into it people were using Monte Carlo to trace neutron tracks in complicated reactors and so on. And photon transport using these statistical randomized methods, which was obviously the way to do complicated multi-dimensional calculations. And then Teller had the idea of maybe one should apply these techniques to statistical ensembles, so then I worked out this appropriate algorithm for doing that. We did the first papers on that. That was almost totally classical physics. Itís been a very widely growing field ever since. This meeting at Los Alamos in June had, maybe 200 or 300 people from all over the world.
Barth:Your collaborators for these papers were Edward Teller and Nick Metropolis and your former wife?
Rosenbluth:Yes. She actually did all the coding, which at that time was a new art for these new machines. You know, no compilers or anything like that.
Barth:And itís also listing A.H. Teller.
Rosenbluth:That was Tellerís wife, who during the war had been one of these computer [women]— he wanted her to get back into the work, but she never showed up. So she was basically—
Barth:Put on the paper for it?
Rosenbluth:Yes. As was Metropolis, I should say. Metropolis was boss of the computer laboratory. We never had a single scientific discussion with him.
Barth:This was then basically your work in 1955 and 1956?
Rosenbluth:No, earlier than that. It sort of started during the period after we had done the MIKE calculations on the SEAC, the Bureau of Standards machine, and it seemed to me there wasnít much more to be done on those so I was looking to switch fields. Of course, I knew there must be something that they could do on computers that they couldnít do before. This was what I hit on, or Teller hit on, Iím not sure.
Barth:So this was essentially a good way to put the weapons design work a little bit on the side and try something different?
Rosenbluth:Yes, thatís right.
Barth:Is there anything else you would like to mention about your time at Los Alamos?
Rosenbluth:Well, I mean, there was a silly episode about the MIKE shot, which I think may have been in Rhodesí book also. I had been sort of in charge of doing all the calculations on the secondary and the general principles everybody followed was to over-design everything to make sure it would work, but I had never looked into the primary. And so once before the test I went down and talked to the primary design group and I found to my horror that in order to make sure it worked they had packed in so much plutonium it was sure to pre-detonate. So in fact a separate primary was flown out to the Marshall Islands a week or two before the shot.
Barth:This sounds like a straightforward physics error. I mean, pre-detonation of plutonium was well known.
Rosenbluth:Oh, yes. It was just an engineering error. They didnít think about that. If somebody had pointed it out they would have calculated it in a minute.
Barth:Somebody had overlooked that you cannot put critical mass of plutonium together before it detonates?
Rosenbluth:Well, they knew this in principle. But somehow when applying it to this particular application they were so obsessed with making sure that it would detonate and give a big yield that they forgot about this.
Barth:How was the work environment at Los Alamos? Was it very collegiate, was it very academic, was it stimulating?
Rosenbluth:Yes, I would say all was very collegiate. All these people you mentioned, people like Conrad [Longmire], Ted Taylor, Harris Mayer, and a number of others, were all good friends and we worked together well. So we worked long hoursóit was a six day week, and usually overtime on top of that. But it was stimulating. We all thought it was important.
Barth:How did political events influence your work? I think about the Oppenheimer case, and JOE-4, the first Russian thermonuclear device in 1953., which I think was actually only a booster.
Rosenbluth:Yes. Well, I think we knew fairly quickly after the radiochemical samples that it was just a booster, it wasnít really an H-bomb, which didnít really come until í56. It bolstered the feeling which most of the scientists had anyway, that the Russians were no dummies and were going to catch up much quicker than the politicians were thinking. The Oppenheimer trial we were all pretty dismayed about. I remember there was a group of us in Los Alamos, I guess prior to the Federation of American Scientists, who protested against it. I know I had a several hour long talk with Teller about it before his proceeding, going to Washington, telling him that it was ridiculous, that certainly nobody thought that Oppenheimer was disloyal. And Teller said as he said at the trial, ďNo, he didnít think he was disloyal either, but he thought he [Oppenheimer] was making very bad decisions.Ē I was pointing out that it didnít seem these loyalty hearings were the proper forum for just firing somebody because you didnít like his point of view. But I warned Teller, which he was already well aware of, that he was going to really lose a very large fraction of his old friends if he did this. But he said he felt it was his duty to the government, that they had supported him on the H-bomb, and while he wasnít going to lie about anything, he wasnít going to say that he thought Oppenheimer was disloyal, he felt he had to give the testimony as he saw it. You must have read his testimony. If you read it literally, itís not so damning.
Barth:But the perception of him going to Washington and testifying against Oppenheimer, of course, had consequences for him?
Barth:Did this affect you as his former student as well? Did you feel that others would say, ďRosenbluth is also a Teller man, therefore be careful what you tell himĒ?
Rosenbluth:I donít think I felt that. Certainly at the lab it was known that I had been opposed to his going to testify at the trial. Probably later on in my career some people associated me with Tellerís ideas, which, you know, I increasingly diverged from.
Barth:Could you fill us in about your participation, your role, in the development of the Ulam-Teller invention [which made modern hydrogen weapons possible]?
Rosenbluth:Yes. I was a junior member and I had been doing all the detailed physics calculations on the booster and the GEORGE shot, which were similar, and I wasnít even aware at the time until months later of this joint Ulam-Teller paper. So I can only give my own impression, which is that Teller really understood all the physics of what was going onóhe was really a physicistóand that Ulam didnít. And what I had seen what Ulam suggested was very vague and it seemed to me didnít constitute any kind of a real invention, whereas what Teller came up with, with the help of Freddie de Hoffmann, me, and Dick Garwin, and others, was a pretty complete design before even the very detailed calculations had been done. So from my point of view, it really should be called ďthe Teller invention.Ē Ulam was a mathematician, not a physicist. A very smart and charming guy, but he was one of these people when he wandered into your office you had a feeling you were going to have a pleasant conversation for the next half hour, but you wouldnít learn anything and nothing new would get done. On the other hand, when Teller came by you knew you were going to have sort of an intensive grilling and some new and relevant ideas thrown at you. So from this perspective, I couldnít understand really why Ulam was receiving equal billing with Teller. But I donít know what went on in the higher level discussions.
Barth:As far as I understand, from Rhodesí book, Ulam had the idea of radiation pressure first. But Iím not certain about that.
Rosenbluth:No, I donít think he did. He had the idea, which was, of course, an old one of— well, the necessary one, having one bomb explode and implode and heat another one, but he was going to use first hydrodynamics and just the shockwaves and then neutron heating, which would have been a disaster. It would have blown it up before it got going. It was Teller who came up with the radiation.
Barth:It was x-ray heating essentially?
Rosenbluth:Yes, radiation implosion. And with the general design of having radiation implosion and a fission sparkplug in the middle. The virtues of high compression were something that surprised me a little bit in Tellerís memoirs, which he claimed not to remember it, but it was certainly something which I had gone over with him in great detail in connection with the booster and the GREENHOUSE shots.
Barth:So if I understand you correctly, it was part of your work to emphasize these points and Teller didn't mention your participation in these issues in his memoirs?
Rosenbluth:Well, I think he did. I had never had any conception of a full design or anything. Looking back on it, I knew all the pieces, but I was operating as a physicist trying to calculate the physics.
Barth:Richard Garwin has received quite a bit of press in the last couple of years for his nuclear weapons work. Could you put this into perspective? How do you see this story?
Rosenbluth:Yes. Dick was, of course, also a fellow graduate student at Chicago. He was sort of unique among the younger generation in that he was both an experimentalist who understood engineering concerns, and a theorist. And Teller (I didnít even know this until later), apparently once he had sort of vented his ideas with Freddie de Hoffmann and me, had asked Garwin to actually make a design that could embody these things that would be practical to use, which Dick did, and I wasnít even aware he was making such a design, but he did come in and have discussions on the physics points with me. And Dick, of course, remembers this all very well.
Barth:If you would try to evaluate his contribution compared to your role and your contribution to the super, how would you weigh these things? Thatís a hard question.
Rosenbluth:A hard question. Modesty forbids me.
Barth:Anything else you would like to mention concerning the Teller-Ulam, or as you mentioned, the Teller principle?
Rosenbluth:No, thatís really the—
Barth:You mentioned Freddie de Hoffmann. What was his role?
Rosenbluth:He was sort of an acolyte to Edward. He wondered around with him all the time. They were fellow Hungarians, I guess. He was very quick and he was very efficient. Also a good friend. Itís by-the-by, but I have a story about Freddie de Hoffmann, as he was notorious for spending all day on the telephone calling people. Itís a great pity he died before the age of cell phones. But anyway, at one point when I was considering moving out here, which was later in the Ď80s, I got a call at 5:00 oíclock in the morning from Freddie, who I hadnít heard from for many years, and he greeted me and he said he had now moved on to the Salk Institute but he still wanted very much and hoped I would come back to UCSD and General Atomic and if there was anything he could do, ďLet me know.Ē The conversation went on for a minute or two and then he said, ďOh, excuse me. Theyíre coming to get me to wheel me into the operating room.Ē In fact, he was one of the few unfortunate people who died from an AIDS infected transfusion, which people didnít really know about then.
Barth:He is also somebody, of course, who worked with you later at General Atomics.
Rosenbluth:Thatís right. He was the boss there.
Barth:We will come back to that. Well, I think we covered quite a bit on Los Alamos. Just a few questions. Did you have any contact with Hans Bethe at this time?
Rosenbluth:Yes. I mean, Hans would come through. I think he spent a year or certainly his summers there. I had great admiration for him. He was a brilliant physicist and knew how to calculate everything. Maybe this isnít politically correct to say, but he always struck me as the sort of prototype, stereotype almost, of the good German. A man of total integrity and belief in his principles. He still had some of the, at that time, traditional pressure and stiffness. But Iím sure he never did an unethical thing in his life. A brilliant scientist.
Barth:Maybe we can switch to Sherwood [the Los Alamos program of controlled nuclear fusion]. How did this look like? Did Sherwood run at Los Alamos while other groups focused on nuclear testing.?
Rosenbluth:Well, it was a separate project. Iím not sure at how many sites it existed in the US. The first one was almost certainly the one I was acquainted with at Los Alamos with Jim Tuck, who had this pinch, which he took me down to see one time. I guess this probably was after all the work on the weapons tests and probably after the Monte Carlo work. I just found the physics fascinating. He would apply high voltage across and break down the gas and it would produce high currents in the magnetic field, but he didnít know really how to estimate how rapidly it would implode or what temperature it would come to and what kind of shocks would develop. So I worked all that out; Dick Garwin also collaborated on that. That was sort of my first interest. Iím trying to remember who else was involved in that. Because there was a nationwide project, Sherwood, but that didnít really get started for a couple of years. Livermore, once it was established, which was about that time, became probably the heaviest player.
Barth:With Richard Post?
Rosenbluth:Richard Post, yes and Herb York.
Barth:You started to work on these issues in 1953 or 1954?
Rosenbluth:Yes, I think 1953.
Barth:At this point Sherwood was already in full swing?
Rosenbluth:Well, a very partial swing. It was totally classified. There was a group at Princeton also, the Matterhorn group. Edward Frieman was deeply involved in that. John Wheeler, of course. And Lyman Spitzer was the guru of that.
Barth:So their work was highly classified until í58. And at Los Alamos it was basically in two different buildings? The theory group which worked on DT burning and the theory group which worked on fusion, were two different groups, or was there a significant overlap?
Rosenbluth:Oh, a significant overlap.
Barth:There was Jim Tuck. Who else was involved at this point on the Los Alamos side, with whom you might have had contact with at this time?
Rosenbluth:Iím not sure I even remember at this point. I mean, it was a small group. There was Harry Dreiser, who was a good theorist, but there wasnít much in the way of theory.
Barth:What was Jim Tuckís motivation to start this line of work?
Rosenbluth:He actually had been working, I think, in England before he went to the Manhattan Project.
Barth:He brought some of these ideas [about fusion] with him? With the long-term prospect that this might be important for energy production or basic physics?
Rosenbluth:Yes. Well, most of the people had a dual motivation. The reason for it being classified was that it was supposed to be so simple that you could produce huge amounts of neutrons and huge amounts of weapons fuel with it. But I think the people working on it didnít really have such illusions and they were more interested in the science.
Barth:When you started to work with the Sherwood Project, how much did you and your co-workers know about the Russian work in this area? Did anybody have an inkling that something like this would be going on?
Barth:So it was really that there might be something but nobody knew anything for certain?
Barth:Even with all the clearances the physicists had?
Barth:[That is, American physicists learned about the Soviet work on controlled fusion] only when Kurchatov visited England in 1956?
Barth:What was the reaction of the American Sherwood workers in your group when they heard about this?
Rosenbluth:Well, we were of course quite excited; and we all, I think, looked forward to it becoming international, declassified projects.
Barth:So there was a wish to work in the open, to be able to publish in the open literature?
Barth:Was Kurchatov known at this time as the father of the Russian A-bomb?
Rosenbluth:Not to me. I donít know.
Barth:Was he then regarded as a leading Russian physicist?
Rosenbluth:Yes. Iím not sure whether he was known as the head of the Russian Atomic Energy Program. Thatís quite a remarkable initiative to have done that.
Barth:And then the push was to declassify Sherwood. Was this a positive motivation for you to stay with fusion, to work in a non-classified area?
Rosenbluth:Yes. It certainly made it easier to move on to other places.
Barth:When I looked into your involvement with Sherwood I found a paper with Richard Garwin on the motor theory.
Rosenbluth:Thatís the one I was telling you about.
Barth:Okay. So this is part of your work for Sherwood. Anything else that comes to mind that you did for Sherwood during this time?
Rosenbluth:Yes, quite a bit. There was work on the theory of Coulomb scattering. At that time Livermore, as soon as they got going, had this strong mirror program.
Rosenbluth:With Post. Calculating the end losses from that involved a lot of detailed mathematical scrutiny of putting together all the collisions and seeing how they would add up to end losses and the role of electrons and ions. That work I actually did in 1955, I think, when I was on sabbatical from Los Alamos to Berkeley. This work which was still all classified, of course. But we did publish an unclassified version of that paper, which didnít mention mirror regimes but just mentioned the Coulomb scattering and how the distribution functions evolved.
Rosenbluth:Yes. [Roald] Sagdeev later told me, he was a student of Landauís at that point, that Landau had gotten very excited about this paper and assigned people to read it and comment on it, and my name became well known in the Russian Sherwood, or whatever it was, community at that point. Landau had actually done similar work, which we didnít know about.
Barth:You mentioned Berkeley. I have not found references to your time at Berkeley. This was basically on sabbatical from Los Alamos to work on Sherwood at Berkeley?
Rosenbluth:Thatís right. Well, I guess I could have worked on what I wanted, but I worked on Sherwood.
Barth:And who was at Berkeley who could support your work?
Rosenbluth:Dave Judd, and William MacDonald. He is still at Maryland, I believe. There was a lot of participation from Livermore also.
Barth:But this was basically for a semester or for a whole year?
Rosenbluth:A whole year.
Barth:From 1955 to 1956?
Rosenbluth:Well, it was eight months or something like that.
Barth:Did you have a chance to go to the Geneva ďAtoms for PeaceĒ conference in 1958?
Barth:Also in 1955?
Rosenbluth:No, in 1958.
Barth:Can you tell us about this experience?
Rosenbluth:Well, I should say, a year or so before that, 1955 or 1956, we built a collaborative program with the British, which had been terminated at the end of the war. So we got to meet some of our British colleagues, particularly Bill Thompson, who was the head of their theory program. He became one of my very best friends. He, unfortunately, died a few years ago. So we had that much international contact. Then we of course were very excited about meeting with the Russians. At that time there werenít many other groups, I guess, that were involved. I donít think the Europeans had really [a program] — the French had some small program. So it was very exciting largely, again, to some extent on a historical-political basis. Both sides felt as Sagdeev put it that we were ďmeeting people from Mars.Ē That was certainly true of our reactions. On the physics side it was probably more relief than anything else. We were sort of nervous that maybe theyíd be way ahead of us. In fact, I would say it turned out they were, at least in theory, maybe a little bit behind us, but basically doing the same things.
Barth:So essentially an American and a Soviet delegation went to Geneva, built their exhibits, and the scientists would then be available to answer questions from the general public to ask questions?
Rosenbluth:Yes, and to talk. A certain amount of talking with the Russians. There were of course obvious KGB types hanging around.
Barth:Do any names come to mind still?
Rosenbluth:Iíve forgotten the KGB types. But they were sort of obvious. Well, there were one or two hard line communists among the scientists, but most of the NKVD men, or KGB men didnít know the science and they kind of stood out. The Russian scientists and we were from the same part of Mars and they were from Jupiter or something. I asked Sagdeev at one point ďWho are all these guys hanging around,Ē when I had him in the corridor, and he said, ďYou know, weíve shut down an awful lot of Gulags and we have to do something to keep these people employed.Ē
Barth:Sagdeev has a good sense of humor?.
Rosenbluth:Yes. Very good. We enjoyed each other. We were the jokesters probably of our delegations. You know, it was astounding to me that here was this guy who had grown up and indoctrinated from birth, totally indoctrinated from birth until the death of Stalin, at least, with communist propaganda, and yet in talking with him there in Geneva he had a humanistic outlook that was not notably different from my own. That was true with a number of the Russians.
Barth:This must have been an important experience.
Rosenbluth:Oh, yes. It was. It was a thrilling experience.
Barth:So Sagdeev was at the Atoms for Peace conference already in 1958?
Barth:Who else comes to mind? Lev A. Artsimovich, was he there?
Rosenbluth:Artsimovich was there. [B. B.] Kadomtsev was there.
Barth:[Alex A.] Galeev?
Rosenbluth:No, I donít think Galeev. Iím not quite sure. I donít think so.
Barth:He was too young for that. And Andrei Sakharov, of course, was not there?
Barth:Any other Russian that comes to mind?
Rosenbluth:[Mikhail A.] Leontovich was there.
Barth:Was there time to sit together informally over a beer and chat about physics?
Rosenbluth:Yes, if you could get away from the handlers. Leontovich was sort of my picture of a Tolstoy-ian saint. We had some nice walks through the gardens and talks. I didnít speak any Russian and he didnít speak any English, but we got along in French. [Vitaly D.] Shafranov was there. Heís been at the Kurchatov Institute for a long time.
Barth:You mentioned the ďRussian handlers.Ē I think itís a fair enough question to see what role the State Department playedódid they give you any advice, ďCheck out these Russians. Tell us how far theyíve gotten.Ē Was there any ďAmerican handlersĒ who tried to keep the physicists in line?
Rosenbluth:Well, they gave us the usual lectures about, ďBeware of Russian blondes that you run into,Ē and so on, which we were never fortunate enough to do. And of course, they gave us the usual lectures about classified information. I never detected any attempt to turn conversations toward classified topics or to pick up information. At least not from the scientists.
Barth:Was there ever a situation where you felt that a Russian counterpart had betrayed your trust when you talked about various issues, that you had the impression that he or she would be working for the KGB and that you had to be careful?
Rosenbluth:Yes. Probably so. Except, as I said, both sides were careful not to get into anything which I would consider classified information. Unlike the Russians, I didnít feel particularly constrained about discussing my political opinions. What difference that would make if they told the KGB, Iím not sure. It must have been much trickier for them, I would think.
Barth:Did the Geneva Atoms for Peace conference in1958 start your interest in collaborating with the Soviets? Or was there interest before that?
Rosenbluth:Well, there was certainly curiosity before that. I think in general Iíve always been sort of an internationalist and felt that certainly— I think most scientists are, and feel that good science gets done everywhere, wherever the conditions are right for it. And certainly having seen the quality both of the science and the people at Geneva enormously strengthened this feeling.
Barth:How long did you stay in Geneva? A couple of days?
Rosenbluth:No, it was longer than that. A couple of weeks.
Barth:In other words, a whole delegation of leading American physicists would drop their work for a number of weeks to participate in these exhibits and be available to answer questions from the public?
Barth:The general public had access to these exhibits and there was basically a shift system that every now and then Dr. Rosenbluth would be on duty to answer questions?
Rosenbluth:Yes. Of course, it was mostly the experimentalists who would be talking at the exhibits. But weíd give theoretical lectures occasionally.
Barth:And the overall sense when you came from Geneva was very positive that this collaboration looked promising and that this was now declassified?
Barth:Moving from Geneva now to the scientific developments of this period, one would think about the theory of instability of the pinch, for example.
Rosenbluth:Yes, sure. It was obvious from experiments that these pinch discharges went into wriggling instabilities and so-called hose instabilities where they would bulge up in the middle and developing all kinds of turbulence. This had to be understood. Some of that had been done by the Russians, some of it had been done by the Princeton group, and I did some of it. I think my main interest during this period was trying to understand in what sense— I mean, so far people had just used hydrodynamic equations, which really didnít apply because as you know the statistical basis of hydrodynamics is really collisions between molecules. When the collisions are very frequent, then they have to stay near the local Maxwellians. And that wasnít really true at all in the process. I guess actually the early work on MHD was done by the SwedesĖ [Hannes] Alfvťn and [Stig] Lundquist. But everybody used that, including we and the Russians. I think I was the first to actually show how you would go from the kinetic picture to a somewhat modified form of hydrodynamics, where you were helped by the fact that in two dimensions the particles were nearly bound to a strong magnetic field so they werenít free to wander in two dimensions. In the third dimension, they could slide up and down the field lines.
Barth:Gyrating around the field lines?
Rosenbluth:Yes. So they were confined in two dimensions but not three. And, of course, collisions played a role eventually in finding a third direction. I think the work I enjoyed the most in plasma physics was in that period of probably 1955 to1959, after I had moved here, to really understanding how you would make this connection. A very important collaborator there was Norman Rostoker.
Barth:Can you tell us about this collaboration? Why was it important? What was the significance of this collaboration?
Rosenbluth:Well, from my point of view there just wasnít really a justification for using these magnetic hydrodynamic equations. The stability of the results you got seemed to be qualitatively right, but to understand where they came from and what the limitations were and how you go beyond them was to me crucially important with establishing the basis of the field, the connection between plasma physics and more conventional fluid dynamics.
Barth:So when we try to understand this move from a kinetic picture to the picture of interacting particles, how did you make this transition? Why did it seem a good step to go? You said there was qualitative agreement between theory and—
Rosenbluth:Well, qualitative in some situations. But you didnít know quite where it would apply. To me itís been sort of unsatisfactory, though I guess many people in many fields of physics, like solid state physics, take some sort of reasonable set of equations and assume theyíre valid and then calculate consequences. But in this case I wanted to really understand why, and under what circumstances, these equations would apply. And thatís what this work I did primarily with Rostoker accomplished.
Barth:What was his strengths? What did he bring to the table? Was he a mathematician, more mathematically inclined?
Rosenbluth:No. He was a very good theoretical physicist. He had been a solid state physicist. He understood statistical mechanics and was quite good at formalism. Of course, I started collaborating with him after we moved to General Atomics. He was one of the early people here too. I believe Norman was Ed Creutzís student. Or maybe he was Walter Cohenís student, who was Ed Creutzís student. But anyway, Ed persuaded him to come early to General Atomics. Norman is quite an amusing fellow. We enjoyed each other. Heís very smart.
Barth:Maybe this is a good opportunity to move towards your departure from Los Alamos to General Atomics in 1956. You mentioned earlier that part of your motivation was that you felt the Super problem had been solved to a large extent.
Barth:And that you were somewhat tired, if I understand you correctly, of the clearance and classification issues.
Rosenbluth:Well, that didnít really bother me too much. Maybe it should have. In fact in some ways it was a positive advantage, because being a lazy person, there was only a limited group of people you had to share your information with, which you would sort of do by word of mouth at various meetings. There wasnít such a necessity for careful lengthy write-ups, which I never much liked. It was more of a club, but it was unhealthy for the science, obviously. Which I could see. But as far as personally bothering me, no.
Barth:Unhealthy in what respect?
Rosenbluth:Well, you were restricting it to a small group of people, whereas there were many other people out there who could have contributed, like Rostoker, for example.
Barth:What motivated you to go to General Atomic? Why not go to, letís say, Princeton?
Rosenbluth:General Atomic was founded as a very exciting adventure. You know, it was John Jay Hopkins in the days when industry thought that science and technology could do everything. That depended on a first-rate scientific laboratory with close ties to industry. He appointed Freddie de Hoffmann, who was a friend, very persuasive, and who I also knew was a dedicated scientist. And Freddie, of course, had the backing of Edward Teller. And during the war I had been stationed in California. I really liked it. I thought it would be a place Iíd like to go. My wife was very eager to leave Los Alamos. We had had a child and she wasnít eager to— So she quit work. It really is a place, I think, which is good if both halves of the marriage are scientists and are working, but pretty dull if the wife isnít. She was quite desperate to get away. Probably if the choice had been left totally to me I would have stayed on. But it sounded as if General Atomic would offer an exciting new venue, which it did. I came to meet Ed Creutz, who I liked a lot. So I decided.
Barth:Ed Creutz recruited a number of physicists from the labs to work at General Atomics?
Rosenbluth:And Freddie de Hoffmann, yes. They were contacts. And from the University. I mean, like, Norman was at Carnegie Tech.
Barth:Did you ever consider going to General Electric that point?
Rosenbluth:No. I mean, there were a few people who went to GE, like Henry Hurwitz. They became much more, I would say, militarized with the Navy submarine program and industrialized with the reactor program. There have always been sort of interesting fringe programs going on around here in additional to the main line themes, which are quite interesting. The original Triga concept [for a small research reactor design], which Ted Taylor I guess was really responsible for. It was an exciting new venture, and it was I guess quite a successful one. I thing theyíve sold a couple of hundred around the world. I got involved in the beginning with the physics of that, which had some interesting problems. Then there was this project, Orion, which you no doubt have heard of that.
Barth:Yes. I have not read the book, so I look forward to that.
Rosenbluth:A lot of these projects (some were more far out than others) however had the virtue that since they were covering new ground there would be sort of new physics to understand before you could get down to a real design or embodiment phase, and I always enjoyed that sort of work.
Barth:Project Orion, just to mention this briefly, you worked on this while at GA?
Rosenbluth:While at GA, yes.
Barth:So this was with Creutz and others?
Rosenbluth:Yes. Ted Taylor was the boss. Freeman Dyson was out for a year or so. I only worked on it for six months or so. After I thought I had pretty well skimmed the cream off, the physics cream, it became more and more an engineering project and a funding problem; I wasnít too interested in that. By that time we had recruited a good collection of I think what were the smartest young plasma theorists in the world.
Barth:It is quite remarkable that an industrial enterprise was able to organize an academically minded theory group in plasma physics.
Barth:How could this happen?
Rosenbluth:The times were right with industry. The thought that investing in long-range science was much more a profitable thing than we probably do today. Freddie de Hoffmann was most encouraging, and I think the attractions of San Diego suited everybody. I think there were a lot of people from the laboratories who were eager to change venues, like Al Simon, from Oak Ridge.
Barth:At that time Freddie de Hoffmann was General Atomicsí President and Ed Creutz was Director of Research?
Barth:So Creutz did the day-to-day guidance for the theory group on various projects?
Barth:So he probably headed the fusion reactors group?
Barth:And you headed the plasma physics theory group?
Rosenbluth:The plasma theory group.
Barth:You stayed at GA for a number of years. What were the major projects that come to mind?
Rosenbluth:We didnít have much of an experimental program. There was some going on in the rest of the world, and we collaborated certainly closely with Princeton and even with the foreign programs. But that was the period in which I think most of the basic ideas of fusion plasma physics were being formulated. Not just here but around the world. The first of them was this reconciling of the kinetic picture with conventional hydrodynamics, which I mentioned. Then actually the next obvious step was what happened when you went beyond thatówhere there were new types of activity which were simply not covered by conventional hydrodynamics, things on the scale of ion gyroradius, very small scale events. As the years went by and measurements got good enough, it became clear that the plasma losses due to this sort of thing were really very key to the problem. Actually the Russians probably started the interest in that Kadomtsev and Roald [Sagdeev], and we did a lot of calculation along those lines of what would be likely to ensue. This small scale turbulence, as you know, is a fairly ubiquitous problem in physics, which is still evolving to this day, and we made a number of contributions to that.
Barth:Is this is a connection back to your Los Alamos time, thinking about instabilities, thinking about turbulences?
Rosenbluth:Yes. I mean, itís generic. But these were the turbulences we were able to detect and some of them at Los Alamos where the MHD type of turbulence, the fluid, which we by this time had learned pretty well what the boundaries of that were, how you could avoid them. But the smaller scale so-called micro turbulence you couldnít really avoid. So the question became then how dangerous were they? What could you do to suppress them? How dangerous were they? How could you diagnose them? So it was many, many years until we were building up a big theoretical framework. Of course, itís all highly non-linear turbulence physics. So the framework, a lot of it was like doing bomb design on Marchants, but more complicated. But it wasnít until sort of good experiments started coming in, in the late Ď60s, that this got onto a sound scientific basis comparing theory with experiment. Thatís been the ongoing theme ever since. I guess the new feature is that in the last 10 or 15 years the computers have started to be big enough that you can really tackle these turbulence problems with some degree of realism.
Barth:The turbulences you focused on during this time, were these turbulences for one particular machine or for all kinds of machines?
Rosenbluth:A number of machines. But I would say itís mostly for the Tokamaks. But certainly early in the Ď50s and Ď60s, a lot of my work was done with mirror machines.
Barth:At this time you were still confident or you hoped that this mirror machine might be a path towards a nuclear fusion reactor?
Rosenbluth:Yes. They all looked difficult, but that looked particularly difficult.
Barth:We will definitely come back in later sessions to the evaluation of the ups and down of mirror machines versus the Tokamak. You mentioned the Tokamak. The Tokamak revolution was only in í68, so when did toroidal machines of the Tokamak kind become prevalent for theoreticians in this country?
Rosenbluth:Of course the big toroidal machine in the US prior to that was the Stellarator at Princeton. There is an amusing story about that which, I think, Sagdeev has probably told you. Well, when I got to Geneva I was not terribly surprised to see that the Russians had thought of pinches, Tokamaks (which we really hadnít pursued but we were aware of), and mirrors. But in particular they had a Stellarator, and the Stellarator always seemed to me like something I never quite understood how Spitzer was ever able to envision it. His geometrical intuition was better than mine. Sure enough, the Russians showed up with a Stellarator. And Sagdeev later told me that that was just a fake. Artsimovich had heard about our Stellarator and told them we couldnít claim that we had something they hadnít thought of, so they just added it on.
Barth:A good piece of foreign policy.
Rosenbluth:Yes. Thatís one of the few examples I know that that sort of chicanery was going on in this business. But anyway, as you know, the US Stellarator results were very depressing in terms of confinement. The Russians were claiming much better results with their Tokamak.
Barth:Does ďMuch better resultsĒ mean better confinement times?
Barth:And temperatures and density?
Barth:All three of them? Or was one more prevalent than others?
Rosenbluth:Well, they all depend on the others. If you have better confinement time you can heat it for longer. There were well understood limitations from these MHD instabilities. Though the confinement time was better than the Stellarators, they didnít really understand what the problems were, though they suspected it was due to the small scale turbulence. Anyway, the people of this country were very skeptical of— I think there was somewhat of a chauvinistic attitude that nobody could really do anything better than we were doing and didnít believe the Tokomak results. And youíve probably heard the story that a group from Culham [Laboratory in England] went to Moscow—
Barth:To do diagnostic measurements?
Rosenbluth:Yes, with the Thompson scattering diagnostics and basically verify the Soviet results, whereupon the biggest Stellarator at Princeton was hastily reconverted to a Tokamak. Harold Furth was one of my closest collaborators. During this period I remember having strong arguments with him where he was saying their results canít possibly be right, and this could be wrong with them and that could be wrong with them, and so on. I had no real counter-argument, except I said, ďThese are very smart people who understand all these issues, and Iím sure theyíve looked into them, or at least approximate them.Ē
Barth:And they did?
Barth:Itís interesting to see this attitude towards Soviet science at this point after Sputnik. Sputnik had demonstrated that the Russians were capable, and still we see this reaction. And this was in which year?
Rosenbluth:í68, í69. Well, the Russians were certainly never as good at— they were good at massive projects like rockets. Their diagnostics and delicate measurements were never up to the par that we had, but they were also aware of their limitations and knew when they were good enough, which was what counted.
Barth:Which brings me back to one issue, the interaction between you as a theoretician and the theory group on one hand and the diagnostics people and the engineers on the other. When you said that you thought about diagnostics for the Tokamak, how close was the interaction with those who actually designed the measurement or the machine at this point? Would you go to design meetings where engineers would say, ďHow do we get a diagnostics window into the Tokamak?Ē
Rosenbluth:Not too much. I would talk to a number of experimentalists like Harold Furth who did talk to both groups. I would talk to the experimentalists, but probably not the diagnostics people, unless— Sometimes the physics of what laid behind the diagnostics was still open, and then I would talk with them. But questions of how you would miniaturize it or what kind of electronics you would use, I never got into that. There was a spread where each group had a partial overlap with its neighbors. Theorists, of course, on the top. Or the bottom, depending on who youíre talking to.
Barth:Who was part of the GA Fusion Group, besides yourself, Bill Drummond, and Norman Rostoker? Was Bruno Coppi working for GA at this time?
Rosenbluth:No, though he would visit from time to time.
Barth:So he was only there occasionally.
Rosenbluth:Al Simon. Ken Fowler.
Barth:David Courant as a mathematician?
Rosenbluth:He may have been out here for reactors or something.
Barth:So the theory group was essentially five to ten people?
Rosenbluth:Something like that, yes.
Barth:Is there anything else you would like to see covered about the GA period? We talked about your work on the instabilities. Anything about the funding aspects? I found a couple of references to the Texas Atomic Energy Research Association.
Rosenbluth:That, I guess, was— Bill Drummond, Iím not sure whether he originated this or not; or maybe de Hoffmann did. But he was able to procure funding from a group of Texas utilities. Which, as I said, probably thought prospects were on a much shorter range then they really were, and they supported us nicely for close to ten years. And then they decided that the funding would discontinue unless the group moved to Texas, which was of course, unreasonable. And Drummond moved to Texas, but nobody else was willing to. So the group split up and went its separate ways, and of course had considerable impact elsewhere. Ken Fowler became the head of fusion at Livermore. Norman Rostoker went up to be a Professor at Cornell and the head of the Applied Physics Department. Al Simon got to be head of the Applied Physics Department at Rochester. And generally people, I think when they split up, became seed groups for a number of places. I had an offer from the Institute for Advanced Study at Princeton, which seemed to be sort of an ideal circumstance because I would be quite close to the Princeton Plasma Lab, and also itís a very prestigious place.
Barth:Before we move to your Princeton time, did you have any contacts to AEC Commissioners while at GA? Did you have to testify [in Congress]?
Rosenbluth:No. I had met several of them on various occasions, but not in their Commissioner capacity.
Barth:So [AEC chairman Lewis] Strauss didnít come to the lab?
Well, Strauss was a peculiar connection actually. Because during World War I he had been part of the Hoover Relief Administration in Russia. And my father actually worked under Strauss. So my father was sort of a friend, a distant friend of Strauss. And Strauss had a certain interest in my career. He never did anything — well, maybe he did something for me, I donít know. But we would be quite friendly on that basis. On one occasion he was handing out medals for work, and he gave me a medal for my work, and my father was there and they had a long reunion and that was quite nice. And of course, he was the great pusher for the H-bomb in the AEC. I never talked to him about that, but I presume he knew about it from Teller.
Edward Teller, ďThe Work of Many People,Ē Science 121 (25 February 1955), 267–275.
Ibid.; quote on p. 271.
Richard Rhodes, The Making of the Atomic Bomb (New York: Touchstone, 1986).
George Dyson, Project Orion: The True Story of the Atomic Spaceship (New York: Henry Holt, 2002)
Session I | Session II