Oral History Transcript — Dr. Carson Mark
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. Carson Mark
Carson Mark; February 24, 1995
ABSTRACT: Discussion of the development of the H bomb at Los Alamos, assisted by John Wheeler’s Project Matterhorn, in the period 1950-1952. Precise recollections of the Ulam-Teller idea. Observations on the contributions of John Wheeler, Stan Ulam, Edward Teller, Norris Bradbury, Marshall Holloway, Enrico Fermi, Conrad Longmire, Ted Taylor, John Reitz, Freddie de Hoffmann, and Cornelius Everett. Comments on President Truman and Lewis Strauss (Atomic Energy Commissioner). Discussion of the physics involved. Remarks on tests in the Pacific in 1951 and 1952.
Mark:Okay. Now you told me your daughter [Nina Tannenwald] is graduated from the training Holloway [David Holloway of Stanford University]could give her and is now launching students at Boulder in the general sea of international affairs, perpetuating misconceptions Holloway may have implanted in her, or you may have. It sounds pretty chancy, doesn't it? [Foregoing with tongue-in-cheek.] And out of that is now to come the view adopted by the next generation. I feel quite sure of her earnestness and her intelligence, but her exposure to things past isn't any greater than anyone else's. Maybe it's a little less than some other people's. You're looking for recollections of H bomb development, 1950 to '51, when you were here.
Ford:And also in the subsequent year of Matterhorn cooperation.
Mark:51-'52. Well, you were at Matterhorn. Is that how long you were there?
Ford:Well, yes. And some part-time into '53.
Mark:Hadn't it pretty much evaporated by the end of '52?
Ford:K. After the Mike shot it ratcheted down quite a lot, just writing a final report and cleaning up odds and ends.
Mark:Okay. There was a chap who may have gone with you to Princeton who had been a captain in the army. I think he went with Matterhorn, and later he certainly went with Livermore. I've been trying to remember his name, which I can't manage. I remember him fairly well. He was not a tall man, he was short and solid and forthright, good in his perception of technical things, strong in his convictions. You don't remember such a guy?
Ford:Not clearly. [Added later: perhaps Carl Hausman.]
He was with the little group that was here, of whom I don't remember many. I think he was still referred to as Captain, although he was out of the Armed Forces. It doesn't matter. I remember him at Livermore, where he still had a strong impression and I made an effort to interest him in coming here. I didn't succeed in that. Now, the first sentence [in KWF's letter of 1/16/95 to Mark] bothers me already considerably. My recollections of the H bomb development as it relates to John Wheeler, '50-'51. "I know," say you, "that he was first recruited to come to Los Alamos in the fall of '49 by Henry de Wolf Smyth and Teller." Now, the idea of recruiting to come here didn't really have any content until the end of 1950. You said the fall of '49. In October '49 the word of the first Russian test swept the country and it shocked a number of people into their conclusions as to what we should do. But the decision to push hard on the hydrogen bomb wasn't reached during that fall. It was reached in the last day of January of 1950 when Truman announced — a very mild announcement, incidentally that he would instruct the Atomic Energy Commission to work on nuclear weapons, including work on the so-called hydrogen bomb. That wasn't a "drop everything, take first priority" kind of statement. It was an afterthought almost.
He wanted the commission to continue working on all kinds of nuclear explosives, including the hydrogen bomb. Edward [Teller] and a few people took that statement and converted it into a presidential order to give first priority to work on the hydrogen bomb. If you go back and read it, it didn't have that quality at all. However, in all references to it from that time on it had that quality of being a directive, and as if it were specific. Truman didn't know what he was talking about, that much was quite clear, and the way he talked about it didn't make it a directive. It included the hydrogen bomb, it didn't exclude it, but it didn't say to give it priority or anything like that. But Teller took it that way right from the beginning. And so he was thereby in a position to do active recruiting, and it was in that frame of mind that he approached Johnny. But you have said that process went on in the fall of '49. I think beginning January '50 is a better position in time.
Ford:Well actually, Smyth did telephone Wheeler in Paris in, I believe, November, and Teller also talked to Wheeler that fall, so that they were jumping the gun certainly relative to the Truman announcement, but they were trying to talk Wheeler into agreeing to come to Los Alamos.
Mark:That might well be, and I'm not denying it of course. I'm surprised because the basis of a presidential directive wasn't available to them until the last day of January 1950. And even then it was a rather qualitative encouragement rather than a directive anyway. Then I had never thought of Henry Smyth as wholly enthusiastic about the proposal. I thought of him much more as being on the fence along with the GAC [General Advisory Committee], not being sure if this is what we really want, or not being sure if there is enough known about it to tag it as a definite objective. I had thought that, but I've never talked with him about it. I didn't know that he had felt positive enough on the subject either from the point of view of feasibility or desirability to call Johnny in Paris. But you're telling me that he got on the phone with Johnny, as did Edward. And Johnny let that cook until February 1950. Certainly he made a decision sometime in that time period, because he showed up here in March 1950 and it had to take a few days to get plane reservations and cancel his hotel space and things like that.
Ford:Actually ship and train at that time.
Mark:Whatever he did. He appeared here with commendable, admirable, surprising promptness. And I always thought he was sort of a nut for doing that, on the strength of the evidence that was available. But you have said he was first recruited by them, made the decision in February, left Paris soon after that. Then an event you mentioned that I didn't know anything whatever about, that he met Freddie de Hoffman in Nice for a briefing on the status of the work.
Ford:Was Freddie at that time already Edward's chief lieutenant?
Mark:I think so. Freddie never really joined the group here that constituted the T Division. He [stood] on the side, and he ran errands for Edward right from the beginning. And then you say they met in Nice for a briefing which was certainly a violation of security regulations. For one thing, he shouldn't have been told about it at that stage, for another thing he shouldn't have talked to him in any of the facilities available to them in Nice. He should have been discussing things like this only inside a controlled area. And only to a person whose clearance for the subject he was certain, which couldn't have happened.
Ford:John may have already had Q clearance, because he was on the Reactor Safeguard Committee.
He was on the Reactor Safeguard Committee, which made some kind of a clearance available, or even posed(?) it. It didn't provide clearance for discussing weapons at all. I was on that committee for twelve years and cleared while on the committee, but only for matters that came up involving GE's plans to build something or other, or troubles with something or other, and so forth. Never for topics which had their basis here [in Los Alamos]. And Johnny was never here, and so his ACRS clearance couldn't have covered discussion of the H bomb which Truman had specifically said wasn't to occur. There was to be no publicity even to the bogeyman that the H bomb might represent. Members of the GAC were prohibited from addressing meetings or writing articles on the status of multi-megaton explosions. That was a sad mistake on Truman's part. It sort of imposed a restriction from public discussion of a very important national decision, which shouldn't have been conducted under the restricted area that he left open. Well, the rest of your sentence I have no complaint about, that he arrived here in March. You say "I believe." I believe it too. And I thought it was fantastically sudden, given the initiation of such a possibility, that he should have made it all the way from Paris to here and scrapped the plans that he probably had fairly positively laid out for what he would do in Europe in favor of the plans which even Freddie in Nice couldn't have conveyed in a very comprehensible fashion.
The whole proposal for the H bomb was discussed really only by Edward and people he had instructed on the subject, in violation of Truman's clear instructions not to discuss it. And only in the form that Edward was in the position to convey at that time, which was the damage radius — or radii — of a 40-megaton explosion detonated at whatever altitude he chose for the lecture, 10,000 feet or something like that, and comparing that with the effect of 20 kilotons detonated at 2,000 feet, as in Hiroshima and Nagasaki. And he took that account to many people in the military establishment, Secretaries of Services, advisory committees of Service chiefs. Freddie accompanied him on some of those expeditions. They put out the estimates of the effects of 40 megatons. They used the number 40 for some reason. I'm sure not only [that number], but that's the one that gets noted by people like Hewlett who wrote the official history of the AEC. It was easy to picture exciting guys like Finletter — he was Secretary of the Air Force, I believe. I've forgotten the name of who was [Secretary of the] Navy at that time. Johnson was Secretary of Defense. So Teller took this account to ignorant characters like that who were certainly in a position to understand what it meant to have a few extra miles of 5 psi overpressure.
Most of them didn't understand what it meant to have a few extra factors on the radiation field, although they could appreciate the blast and thermal radii and effects, because those are the things that were so evidently well known already from Hiroshima. And that allowed them to conclude that if the Russians got this bomb first our position would be intolerable and therefore they joined the class of people who said work on this is imperative, as fast as possible. And since you're not working as fast as possible, you're not working fast enough. Who was in that crew? Well, the guys I mentioned: Finletter, Secretary of the Air Force, I think he was. That's easy to find out. He didn't understand the mechanics of a nuclear explosion at all, nor have any basis for guessing whether the hydrogen bomb was likely or would have difficulty in performing. Edward informed them that it would give 40 megatons. From there on the arithmetic was easy. I think he gave the same story to Wheeler. Because it's the only story there was to give. Freddie didn't know anything beyond what Edward was in a position to say, but he could certainly carry that message around since he had sat through the talk often enough. Johnny was not in any position to assess the problems connected with an explosion, it's likelihood or difficulty with any degree of assurance. But he could — I'm sure he did — receive the assurance that a lot of people whose opinion commanded respect were very strongly pro-super and were already taking offense at the fact that Edward also conveyed that people here weren't paying any attention to the work on the prospect — which they immediately saw was of such importance. They didn't know whether we were or weren't putting attention on the problem, but they heard from Edward that we weren't; therefore they knew that. And that was the basis of a lot of criticism and objection to Los Alamos, reaction to the idea of the super.
Ford:Carson, when did Edward come back full time to Los Alamos? Was that in the fall of '49?
No, it was July. He came back at the end of the term in '49 (the end of the term in Chicago) for "full time" — put that in quotes. He was going to be here for a year. He was taking a year's leave from Chicago. He was going to be working on the super while he was here, and that session started in July of '49. By the time of the Russian test, he extended his planned stay. It didn't take any effort to extend the plan to stay; it took a little activity to extend an absence from Chicago. But he extended his leave from Chicago, in view of the Russian bomb, at the end of September. So he was here full time, but it was on a basis like that. He had come in the beginning of the summer because of a feeling that the work on the super wasn't proceeding as fast as he thought it should or felt it could. He spent the summer here when the Russians set off their first bomb, and that entered a lot of people's thoughts as stepping up the urgency of work on weapons. The Russians were so much closer than so many people had been assuming.
Lewis Strauss was a very active person in this discussion and it was he who opined that the American response should be a quantum jump in the efficacy of weapons. And by that he meant going into the megaton range. The biggest explosion up to that point was 20 kilotons. And while we knew we could build things bigger than that, it was alleged that the Russians might be building megatons, and what would we do then? The argument was easy: "The Russians are proceeding so much faster than people had said they might in 1945. For all we know they might be on the fringe of a super already." There was no basis for saying that, but it was an impossible point to debate. In fact that's how this whole story went. There was no particular basis for saying that the Russians were doing a hydrogen bomb or were near to doing it. There was a good basis for … What would the effect of 40 megatons be [some missing words here]?. That should be fairly clear. Whether the Russians were anywhere close to thinking of that [wasn't known]. Their first test was only in the 20 kiloton range. That's all that was known at that time, and it certified progress, except that it was faster than many people believed or had said. It wasn't really faster than Seitz and Bethe had said, it was faster than General Groves had said, and as lots of politicians opined. It was, after all, four years, and it had taken us two-and-a-half. A lot of that [poor forecasting] was misconception, misreading what building an atom bomb really meant. Since it had come to the American public and to the Congress as a totally unexpected secret, it was swathed in this notion of impenetrable secrecy right from the beginning. And a secret is something which either has to be discovered or which can easily be conveyed.
Whereas the atom bomb was just a subject on which there was no real secrecy involved early on. It was clear from the beginning of Los Alamos that a nuclear explosion could be pictured, and what do we need to know to flesh out detailed plans to engineer such a thing? That was known at the time Los Alamos had its first week of coming together in a conference. And that didn't prove that everything had been thought of, but it did lay out a track you could follow with an explosion probably emerging at the end of it. Lots of things had to be ascertained. Many of them were mentioned by Serber in his introductory lectures. We needed to know neutron cross sections, which we didn't know. It was perfectly clear how to go about measuring them, and people did. Fission cross section, number of neutrons per fission, the length of time before the neutrons were released from fission. It was thought that that was 10-13 seconds, like the transit time of a nucleon through the nucleus [10-21 s?]. That time was never measured, of course, but it was measured here to be less than 10-8 s, whereas the other length of time was 10-13 and you couldn't picture measuring that. So it's harder to know if the neutrons appeared instantaneously. At the very start that would have been a possible holdup. If it had been as long as 10-8 seconds, the nuclear explosion wouldn't have proceeded as it did. Anyway, it only took less than three years from the stage of Serber's introductory lectures to the appearance of a fission bomb. Why it should have been thought that it was going to take the Russians 20 years, there is no way of explaining that. General Groves said that's how long it would take.
There is no reason for supposing that that wasn't a well based estimate, so people were surprised when it only took the Russians four years. But there was no reason for that except the overlay of secrecy — really the overlaying of the idea of secrecy being involved and having to be worked through. It was never pictured as it should have been, as just a job which a strong enough effort well enough equipped with funds and manpower and implementation could work through in a few years. So the fact of their coming to an explosion was a shock to everyone, and Lewis Strauss says we have to make a quantum jump, otherwise we'll be smothered by the Russians with their hydrogen bombs. No doubt they're getting close, and so on, and so on, building up, amplifying. Truman personally looked to impose that irrelevant line of thought, being such a enthusiast for secrecy. Now nobody had any doubt that all of this was secret to him and that the fission process came as a great surprise to him. Fair enough, no reason why it shouldn't. But to say therefore it's secret and therefore we're in trouble if we don't keep tight-lipped on the subject is part of the problem of the times. And the idea of discussing the possibility of larger explosions, vague as they were — he put a blanket injunction against doing that.
That meant that the thing was discussed in a way that things should never be discussed in a democracy, and that is to say the lack of realism that it would be favorable to have as much of it as one could. So Freddie talking in Nice to the extent necessary to excite Johnny's interest was obviously in violation of security as it would have been read at that time — and as it might still be read. The kind of person who works now on classification has a lower level of competence than the people who worked on classification in 1945. He understands less and insists on more today than he did then. And so the business of classification has gone downhill in the direction of amplifying an unnecessary sense of mystery. That's why we shouldn't be talking here. I had suggested to Kay [Mrs. Mark] that if Ken doesn't have a clearance, maybe we should talk in the lab, but then if he doesn't have a clearance he can't get in the lab.
Mark:That would have been futile. Well, in writing what you will come down to you will necessarily be on rather uneasy ice. Some of the things that I haven't read here make that particularly certain. What was the state of understanding at the time Johnny came here? That in my view would be a criticism of Johnny. He might not see it that way, but I would be inclined to say how could a sensible man have responded to such garbage, disrupting his own plans, which had presumably received a lot of thought and on which he set some store in advance. The only way was that the state of affairs was misrepresented, or improperly represented to him, to give him the feeling that this thing is closer in than I thought, it really does need all the muscle that one can round up for it. And that if we round up that muscle it will come through. None of those things [was] true, as you now know. It wasn't a matter of adding even very capable, bright people. There may not be very many like Johnny, but Johnny would be an example. Add a great number of them, will you speed up this process? And the answer is, we don't know enough about it to answer. Edward knew enough about it to answer, because he never had any doubt. It was an article of religious faith he was carrying around. I hadn't realized that that had infected Henry de Wolf Smyth, with whom I became not well acquainted but definitely fond [of] in the little encounters I had with him. Where have we got to?
Ford:Coming to the questions in my letter. The first question: What was the principal problem with the "classical super" as envisaged in 1950? I should remember but don't. I remember the concept of the inverse Compton effect. Was that one of the problems.
Mark:Okay. That was a problem. It had been solved, and a lot of work had gone into it, and one was in the position to make quantitative estimates of this process. Well, look, the classical super — I think one can allow himself to say what the classical super looked like at that time. A place where one has to work a little harder, what did various people think of that, and what was clearly pinned down and what wasn't. The classical super was the idea that a detonation wave started in a stick of deuterium — liquid deuterium — would propagate, from the end at which you started it through the length of the stick that you had assembled. And if it gave a uniform burning as you reached one spot or a further spot along the cylinder, then the yield that would be produced didn't have a natural limit. You could assemble as much liquid deuterium as you wanted because it didn't have a critical mass. You could just go on adding to the length of the dewar that you were building, and that would add to the total output if and when a propagation wave was initiated and burned through it. It was not known whether a burning wave could be established and would represent a steady state in this cylinder and would proceed on, engulfing more and more of material, nor was it well established what it would take to start such a propagation wave. There were two questions: What would it take to start it? If you supplied those conditions, would it propagate? Neither question was answered.
Ford:So the definition of the classical super is the hypothetical propagation along an unlimited quantity of deuterium?
Mark:After you have initiated a burn wave at one end. Unknown was what would it take to initiate, and if you provided one or another set of conditions would a wave propagate or not. Neither of those questions had an answer at the time we're talking of here. Now Edward was the source of the assertion that there would be a steady-state propagation, that a detonation wave would proceed. And the super therefore had an open-ended yield. That frightened the GAC. Why they believed that, I have never understood. That group had enough intelligence in it to realize that this statement of propagation to a definite, a large outcome wasn't established as a known fact. But they didn't choose to reflect any particular complaint about the fact that this just wasn't known, although some of the guys in that group would have realized that if they'd come down to talk about it. Fermi, Oppenheimer, Rabi, good God, you couldn't pull the wool over the eyes of that crowd, if their eyes were open and they were thinking about what you were saying. But they seem to have accepted the myth that the propagation would occur, that the bomb therefore had no limit to its yield, and consequently was immoral because it was such a horrifying object. That they seemed to accept without proof.
Ford:My recollection is that at least into the fall of 1950, the answer to the question "Will the classical super work?" was "Probably not," that as more calculations were done there was less and less reason for optimism about it.
Mark:That is correct. By the fall of '50 there was strong reason not to be optimistic. We'll come to that bit. I said there were two major problems. One was: What would it take to initiate? And the second: If initiated, would a propagation follow? Neither was answered — or no answer was known to either — in the early part of 1950, at which time we're already committed to produce a thing with those properties. We were instructed to, not committed to. We never said we could. We said we didn't know. That isn't a very acceptable answer. What people wanted was, "We want to hear that you're working on it and that it will soon be done." And we kept saying, "No, we don't really know." Edward kept saying, "There's no doubt." All right. We are on your question, "What was the principal problem [with the classical super] as envisaged in 1950?" I told you there were two.
Mark:They were quite different in their physics. Each was significant. The question of what it would take to start the condition had its own significance. Edward was quite confident on that point: It could be done with no more than 200 grams of tritium. He had no basis for that. There never was a basis for that. But that was an important point. John Manley, working here, took that sort of assertion seriously and wrote a long paper in which he compared the cost-benefit ratio, if you like, of having a super which took 200 grams of tritium or using the 200 grams in other ways that we could think of. Which was the more effective? The answer was of course straightforward. It was simply obvious.
Ford:The "other ways that we could think of" meaning as a booster to a fission weapon?
To boost fission weapons and use a large number of them, instead of using this much tritium in one swipe. Now, there's no doubt that if you could run the 40-megaton super with 200 grams of tritium, it would do a lot of damage. It would take a lot of effort to make that much damage with weapons of the size we could picture, like some tens of kilotons. You have to build a large number, pay for a large number, and deliver a large number, none of those being trivial items. John Manley wrote a paper which raised a question as to which would be more cost effective. Without coming out with a confident answer. His efforts weren't given much attention because the kind of person reading them wasn't well equipped to think about the reality. Well, after all, we see the blast damage is three miles. What could compare with that? All you have is one mile blast damage from 20 kilotons. It was a very reasonable question John tried to deal with, but he didn't really have input which was needed to give a complete discussion. He did write a complete paper, he did raise a real question. He pointed out that if it took more than a certain amount of tritium, it wouldn't be worthwhile. Only if it took less would you be making money. Well, the only basis for fixing any amount was the assertion Edward had made that 200 grams is certainly enough, and [he asserted], since that's based on a simple-minded disposition of the tritium, it's quite certain that we can achieve the effect with less by being more thoughtful on how we dispose of the tritium we make use of, and if we can get that number down then the answer to John's question would be clearer than it is today. That was a state of mind in the fall of 1949.
Manley wrote his paper about the time the GAC made its decision not to enter into a crash program — to advise against a crash program. They already had John's [John Manley's] paper in front of them. They already knew, if they chose to think about it, that this 200 grams was not a demonstrated number, but just Edward's supposition. He said it often, and when Edward speaks he speaks very persuasively, and so the number had some reality in conversational terms, but no reality in terms of actual physics. Ulam, starting in about December 1949, took that problem to heart and worked at it for several months. And there's the famous Ulam-Everett calculation which was precisely on that problem. Assuming that you have gotten burning in a mixture of deuterium and tritium of a certain size, let us calculate how that excites some adjacent deuterium. Does it put it in a state where self-sustaining burning is established? And if not, we'll add some more tritium over here to increase the energy that can be delivered to the deuterium and maybe by pushing on that point it will come to the point where this much tritium will initiate a self-sustaining state whereas smaller amounts will not. That was the subject of the Ulam-Everett effort, which is very often referred to. And, as I've tried to say it, it made the nature of the unknown question clearer and the significance of the answer clearer. Do those run into classification problems? The answer is they certainly should not. And any classification officer who … I'll never say it again.
Ford:[laughs] Your final remark [on the last tape] was that any classification officer who thinks your description of the classical super problems should be classified isn't thinking straight, or something to that effect.
Is not assessing the true significance of the information conveyed. If we were going to come out with an answer that 125 grams would start things going, that might be classified. But since we're going to come out with an answer that we haven't thought of enough to achieve this state, that can't be classified. We're not giving a number a person can use to repeat the process. And no one answer to that question could be classified. Since the question is asked and we never got an answer, it shouldn't be classified, even though discussion of it later might catch somebody's attention. That's what I'm saying. Even that's clear. Ulam started, then, with a small amount of tritium, like the amount that Edward had suggested would be enough, and he had this amount of tritium mixed with deuterium burning, and taking the neutrons' long-range effects from that burning, applying them to an adjacent mass of deuterium to see what state it put the deuterium in, and then prepared to ask if I had the deuterium in that state, whether it would sustain itself. So he started with an amount which was equal to that which had been bruited about, not that many people would recognize, and came to the conclusion that depositing all the energy that burning that much stuff over here could transport into a mass of deuterium over there would not raise the physical state of the deuterium to a self-sustaining level.
So he increased the amount, and he increased it several times, and he came to the conclusion that it would take a great deal more than had been suggested, and that still wouldn't work. That didn't prove that the state couldn't be established; it just proved this much tritium didn't establish it. And what those numbers were that he used, I don't really know and I don't think it matters. He upped the amount several times over the amount which had been qualitatively referred to. As he increased that amount this annoyed Edward more. He thought it was being done malevolently. He suspected Stan of adopting arguments against the super without even trying to establish arguments for it. That really wasn't the case. Stan didn't mind in the least that the super didn't work. That pleased him. The fact that Edward became furious pleased him too. So it's a pretty reprehensible situation, Edward accusing Stan of doing this malevolently, whereas all he was doing was saying, "If this much doesn't do it, I'll add a little," like doubling it. He doubled it several times. Deposited the energy in deuterium as well as he could with somewhat qualitative sorts of estimates. He was using Monte Carlo processes to deposit his energy and distribute it, and that's not analytic. Some guessing is involved, some errors are possible, but the Monte Carlo technique, even though it's not exact, if done with any thought is not likely to be misleading. Stan confessed that the method used was not exact, so it wasn't a proof. But Edward thought that the fact that he came to—and I'm not going to say how many grams he came to, which was a great deal higher than Edward appeared to offer — it was getting into the region that Manley was prepared to argue would make the whole proposition nonsensical.
A quite different argument of course, based on different types of input. Anyway, Edward and Ulam had a contretemps over the outcome of this calculation in which Edward either accused Stan or claimed that Stan was acting in a negative motivation, whereas Stan argued that he wasn't really led by motivation at all. Finding that a certain amount of tritium didn't suffice, he naturally went on to try some more. Now this calculation is pretty persuasive, but not totally demonstrating that situation, because you could think of deuterium and tritium in a different geometry than the one you happen to use. It's going to have a different consequence in the arbitrarily large deuterium mass sitting next to it. Make it thick here and thin at the edges. That'll lead to a lot more tritium than if I make it spherical or a different amount. So you can't say it proves that a given amount cannot be disposed in such a way as to have the effect. It doesn't really do that. It gives the indication if disposed in this way that comes to mind, it doesn't have that effect. Then when you take a look at the deuterium and ask whether it's been brought to a sustainable level, you now again have a calculation which you don't really have the makings of — you don't have all the input you need to make the calculation analytic. You've got to again use a probabilistic Monte Carlo style discussion to ask if having brought this much of the deuterium to a nuclear temperature of 30 kilovolts, what's going to happen? Well, you can release some energy from that deuterium, try to distribute it, and you have to admit that you haven't necessarily covered the whole case, the whole possible case.
Stan was inclined to think and say that he proved that the amount of tritium needed was beyond any reasonable quantity to invest in any one explosion. I don't think he was really entitled to state that flatly, but that didn't mean that he didn't state it flatly. Edward wasn't entitled to say that Stan was doing this out of underhanded motives. He wasn't entitled to say that, but he did say that. So that led, as you could picture, to nasty feelings. Where Johnny wound up in that impasse, I don't know.
Ford:Let me ask a practical question. Were Everett's and Ulam's calculations carried out entirely by hand with the young ladies with the Marchant calculators, or was the IBM CPC brought in?
Oh, this was done with estimates. It was really done by Everett, who recorded the numbers and started the next round to introduce some new energy. Done by Everett on a slide rule. With guesses as to the deposition ranges of 14-MeV neutrons in the deuterium medium. Now those guesses aren't very bad. Again, they won't be accurate, but they're not likely to be very misleading. However, how do you decide when you've got a certain mass of heated deuterium — and you certainly got that — whether it is hot enough to give enough energy to spread itself or not. That's a tougher question and a different question. And your way of going about that might be open to criticism if you're using a probabilistic style argument to plead your case. We'll come back to that sort of question in a minute. There was the other question, would a burning propagate in deuterium of liquid density? This calculation that I'm describing was wound up by February of 1951. Truman's announcement was at the end of January, 1950. This calculation carried over to a little more than a year later than that, proving to most people that the amount of tritium was way off the estimates Edward had made and that it was an unconscionable amount to think of investing in one shot.
It would be hard to produce many shots with that much tritium because Edward had excited Ernest Lawrence and Luis Alvarez on the business of making tritium for its use in a super and had led them to aim their efforts at a gram a day. Well, if you need many thousand grams, you need some thousand days. The tritium decays in ten years. You have to work pretty hard to keep up. You come to that point anyway. They weren't getting close to a gram a day, and they were working their foolish heads off. They were spending money like water on the big accelerator they were designing, with the purpose of possibly making a gram of tritium a day. So that wasn't going anywhere, and Ulam's numbers were making even that number questionable. Ulam's numbers were themselves questionable, but that was all one had to look at. Now, having taken the Ulam-Everett calculation, which has often been discussed — discussed in Ulam's book, The Adventures of a Mathematician, discussed in a number of places — the significance of that Ulam-Everett calculation to find out how much tritium might put the deuterium in a sustainable state. And the answer — I don't know what numbers are given, but the answer was that they tried larger and larger amounts without succeeding. The fact that Edward was very annoyed about this is also written by Hewlett in the history of the Atomic Energy Commission covering those years. I think he gives enough on that calculation — it's unclassified — that one could say all that one would want to say about it anyway. Pretty good authority. Ulam also gives enough on the subject. So between the two you can keep yourself clear of legitimate classification concerns. It's also said—again Hewlett is my reference at this point. You are familiar with that?
In August of 1950 Ulam and Fermi decided to make a much more difficult calculation on the question of propagation. Are there conditions that if established in the certain section of the cylinder will bring a subsequent section up to that same condition? Assume a nuclear temperature of 50 kilovolts. Well, if that doesn't do it, we'll assume 60. Can we get a piece of the cylinder being cold liquid adjacent to that up to the same state. So there's a simple-minded notion as to how propagation would in fact occur. Fermi and Ulam tackled that question in a sort of brute-force way, again using heavily Monte Carlo style estimates concerning the deposition of the energy developed in some of the hot region into the cool region and the exchange of energy back and forth to find out if their cold region was brought up to the level of this one or not. And the cold one receives heat from the hot one, there's 14-MeV neutrons formed over here, deposits energy over there, and 2-MeV neutrons the same thing. There are radiation quanta depositing energy also. They have a certain range in uncompressed deuterium, and deposit more in the electrons than in the nuclei, but they heat things up. So they tried to isolate those effects and asked the question; Would it seem that a section that hasn't been affected yet can be brought to the same level as one that is at an arbitrary level which we'll choose? They just went up to that … propagating thing.
They spent several weeks on this. And they concluded that the new section would not come up to the level of the old one, even when they made the old level as hot as they could see an excuse for. As I say, if they made it 50 kilovolts and it didn't propagate, they could make it 60 and see if that propagated. The answer was they couldn't get indication of propagation with the dd cross sections which were believed at the time to apply. They tried it over again with a different assumption about the dd cross sections, because that's what determines the amount of energy generated in the cold section, whether it’s going to go on further or not. They start with d + d, giving 3 MeV + 2 MeV neutrons, and we'll put that 2 MeV neutron in the range of that much, and keep track of the products that are down here, the proton [slip of tongue — he meant neutron] and helium 3, and proton and the triton, and keep track of how much of those we have, and let them react with the rate at which they're inclined to or believed to and put out their products and spread them and see whether or not we get an indication of maintaining or transmitting to this section the starting conditions of that section. I think this is Fermi helping to use the Monte Carlo. He was a great friend of Edward, and even Edward I don't think ever considered accusing Fermi of subversive intention, steering the results the wrong way. He could easily take that view of Ulam, but not at all, you see, of Enrico. Anyway they came out with the conclusion that in order to see signs of propagation the cross sections would have to be at least 60 percent bigger than they were at the time believed to be for the dd reaction. Well those cross sections had received a lot of attention.
There is no reason to suppose that they weren't fairly well pinned down. It's a pretty negative outcome, except that they confessed that a lot of their work had had to be done on the basis of guesses and estimates since they weren't in a position to do anything analytical with that process. You guys were doing that to the extent that it was feasible. You weren't finding much encouraging along that line. They had to estimate how much energy the nuclei transmitted to the quanta, the energy-loss term, how much energy the nuclei transmitted through other nuclei, keep track of the energy level of the electrons, the quanta, and the nuclei. There was a lot of work on those rates of transfer, and during the war written up in compilations by Konopinski and others, the Super Handbook. So one had pretty good data on handling those processes. Provided one fed them in in a defensible fashion. You were just about to say something.
Ford:A question. That August 1950 work, was that an Ulam-Fermi collaboration?
Ford:Because I remember an occasion — perhaps it was the following summer — when Ulam and Fermi were competitively carrying out a calculation. I think that was when Ulam and Everett were using the MANIAC, and Fermi was using a very attractive young lady with a Marchant calculator, and there was sort of a friendly competition. But I don't remember what it was they were calculating.
Mark:I don't remember that scene at all.
Ford:That must have been the following summer.
Mark:Well, that's possible. No, this was a joint paper. The output of this work led to the conclusion that the cross section would have to be bigger if there was to be an outcome of propagation. And they made the cross section bigger and showed that with the means they were using they got propagation. Whereas with cross sections of the sort that were recommended, they didn't. Pretty damn persuasive, especially when you add Enrico's skepticism into the picture. Now that calculation was taken to the GAC in the fall of 1950.
Ford:Yes. I recall a very fat report that Wheeler and his group helped to prepare for the GAC, one that we were informally calling the telephone book, which tried to pull together — to summarize — all of these past calculations on the classical super.
Mark:Yes. I remember — not so well as you — that such a report was prepared, and carried off to the GAC by Johnny and Edward and presented to the GAC. It's my recollection, but not very specific, that the effect was not badly described by Hewlett, who said something of the sort that the presentation to the GAC persuaded them that nothing much had changed, except that Edward was still as enthusiastic as ever. He provided the main continuing support for the idea that the super was there and could be reached if one only kept trying.
Ford:In Hewlett's account of that GAC meeting in the fall of '50, he quotes Oppenheimer as offering his "frustrated gratitude" to Teller and Wheeler for their report, indicating that the GAC members appreciated all the work but they were still left in a quandary and equally confused about what the prospects were.
Mark:Perhaps Hewlett puts that in, but in addition, he put in something to the effect that Edward's enthusiasm was undiminished but that that was still the main support to the idea that the super was there, and that that hadn't been wholeheartedly persuasive. I may have screwed that up a little, so don't quote me on that subject, but there is a statement in there, as I believe it comes out, that the GAC was impressed with Edward's continuing conviction rather than with any new facts brought forth. That would be all you'd want to say about this telephone book thing. And he didn't change their minds. He didn't make the super seem desirable, which they already said it wasn't, nor did it make them say it was more feasible than they thought it was before, which was the 50:50 chance that something like it might work. Well, that brings us to the end of 1950. Just about the end of Johnny's heavy interaction. I think that report to the GAC was maybe the last big event in his work here.
Ford:He was involved also in preparation for a test — was it the Ivy test? — that occurred in the spring of '51 or the summer of '51?
Mark:Oh. The whole Matterhorn group at Los Alamos was heavily involved in that spring '51 test, the Greenhouse test.
Ford:Greenhouse, that's right.
There were two thermonuclear shots in there, and those booster tests which Edward … [missing words]. I think that was the first induced thermonuclear reaction. A small scale, but still macroscopic. It was hailed as a great accomplishment, and it was. It was an accomplishment in my view in two different respects. It would have been easy to have worked out that design and had some flaw in it so it didn't come through, behave as your calculations indicated it should. [missing words] You could have made a mistake somewhere in not having too sharp an angle one place or another or having the wrong opacity for the walls of the pipe, [missing words] instead of [missing words]. Consequently we could have made simple mistakes, which would have changed the result at the little hohlraum at the end, and you could have mis-estimated the behavior of the material in that hohlraum. It's been brought up to a temperature of [missing words] some [missing words] of all that, and the nuclei are supposed to get to the temperature that is very much higher than the temperatures imposed on the region by [missing words] kilovolt-like temperature that led off from there.
So it was a great success in that it showed that people were in the position to calculate these things and get a reasonable picture of what happens. It was a tremendous fact. And the calculation involved approximations, guesses at many points. These guesses had been reasonable enough not to vitiate the outcome. It meant that you could calculate things involving kilovolt-like temperatures — containing them, steering them, manipulating them, and having them have the effects which you argued for them. And in particular having the nuclear temperature run away from the ambient temperature. So there was success in those respects, this was a big success. It wouldn't have been certain in advance that we could handle all those terms correctly, and I don't think that it was certain up to the moment of the test if they had handled it correctly. Had they been mishandled a little bit, the results would have strayed seriously away from the calculated results, which instead were the ones observed. And theory's done a lot. One can make these estimates and get them right. The way you people had done it was adequate to that purpose. There is demonstration that you can do it. If we go back to the classical super, apart from this encouragement, this sort of arithmetic covers the case — and quite a complicated case. It says nothing about the behavior of the classical super. It doesn't prove propagation, it doesn't prove initiation, it doesn't prove anything about deuterium burning. Those things weren't wide open questions. So the behavior answered that question for you. We were treating things fairly reasonably — reasonably enough to get a picture of what happens at the end of a complicated chain which we'd never seen before.
But it doesn't answer the outstanding question: Will a classical super have a propagating burn feature or not? It doesn't touch that question. So this was an unnecessary exercise in a way. There wasn't a helluva lot of question about whether or not things might happen in this way, in general. That didn't close the question. Our arithmetic might be off. It was enough energy to make you confident that burning would proceed with the temperatures one would achieve at the cross sections one had in hand for the purpose. If the cross sections had been wildly off, that could have upset the whole thing. But they weren't, obviously. So it didn't bear on the existence questions of the super. It bore on questions of which one might have said in advance: We don't really doubt that things might behave in this way; we doubt whether we will calculate it correctly, and this business of exchanging radiation quanta and material energy is useless, and the hydrodynamics of the material heated into this plasma-type condition. It is new to us, we might do that wrong. We are not really greatly in doubt about that; we assume we do it about right, or right enough, and the experiment proves that we did. Edward made the claim that this test sort of authenticated the qualitative statement made about the super, but it didn't. Because it didn't really bear on that. He was very encouraged. He has written many times that without experiments, we wouldn't have had the courage to go on working on the super. That's a very reasonable statement, but the experiment as it was laid out didn't have much acting back on the things that were open on the super. Say I. These are all my biased statements. Where else do we need to go on your questions?
Ford:Well, if we're in the spring of '51, your comments on, of course, the famous Teller-Ulam idea would be good to have, as to their relative roles — without saying anything that's still classified.
Mark:I'll say some things which I don't especially want you to recover from your damn tape.
I don't think Ulam ever gave a hoot about the super. It's true he didn't do the calculations, which Edward found disappointing, in a malevolent way. I think he did them honestly, and he worked with Fermi to discuss propagation. He was not trying to prove propagation doesn't occur, he just didn't find any. So I don't think he stayed awake nights trying to figure out how to make the super run. His contribution was great, very significant, but in a manner fortuitous.
He came into my office one afternoon, and I didn't really want to see him, because we were running an operation in Nevada right at that time. He put a lot more attention into the tests that you are interested in, and he interrupted some time that I had hoped to have available for something else. I've forgotten what it was. I wish it didn't matter. All steamed up, wanting to talk about some thinking of his recently, on which he put the tag. "It would be wonderful to see the effect of a bomb in a box." He used that phrase a number of times. And it would be a lot more interesting to be reading something on that line than what we were doing in Nevada, where we had our hands full with what we were doing. "We should throw that whole program in the wastebasket and do something interesting," said Stan. He didn't really care much about what we were doing in Nevada. I guess I don't either — didn't either — because I can't remember what it was. It was a more or less straightforward set of experiments with bombs of different physical dimensions than we'd shot before, and different arrangements and proportions of fissile material. We wondered if we could calculate them correctly, and the results all came out very agreeably.
No big surprises, I believe. And they were important in a trivial workaday sense. They helped us make better use of the stockpile that existed of plutonium and oralloy [term for U235]. We could mix those things in better proportions than we were previously mixing and get the same damage effect with the same amount of fissile material, or getting a better damage effect, so that we were doing worthwhile things in that trivial sense. And I was very interested in how we were making out with those. Stan didn't know anything about what we were doing in that department, came in and dismissed the whole thing, and said that it'd be a lot more interesting to explore the effects you could achieve with a bomb in a box. Put a bomb here and wrap a heavy case around it. Energy will come out and down here in the corner we can put a very small amount of plutonium and compress it to a very high degree and have a reaction from an amount of plutonium that we'd never previously thought we could do anything with. All very correct statements, not having any relation to the super at all. As I spoke then, that's how they were put. So why don't we just change the program and do something interesting like that? Here we've got months getting the program approved in Washington, months getting [missing words] to build a tower(?), but he never had the vision to build a device [missing words].
All of this Stan would have thrown over the fence, happily. Starting on a program that didn't answer any of those questions that we'd impressed the AEC with, and the military with, ourselves with, which didn't have anything to do with the super. Having given me this hour-long exposure to the differences — qualitative differences — which we might expect to find if we put a bomb in a box and used the temperature, high-energy density, that would result to achieve hydrodynamic effects on a new scale, completely off scale from anything we'd done before. He gave me this long lecture. Then the next day he went and spent an hour with Edward in Edward's office, and Edward immediately applied the ideas he was talking about to the effect they might have on a thermonuclear assembly. Instead of squeezing a little plutonium very hard and getting a reaction from a smaller amount than we'd ever considered before, Edward moved that over to taking a mass of lithium deuteride or deuterium and squeezing it to a degree that we'd never imagined before. So the bomb in the box was immediately translated into an approach to a thermonuclear reaction, and Edward translated what I'd heard from Stan to a means of handling appreciable amounts of thermonuclear fuel. Stan had been going to heat a small pellet of plutonium or compress it by unprecedented pressures transmitted hydrodynamically inside the materials placed in the box. A box of beryllium, let's say, or graphite, whatever you like. Set the bomb off over in the far corner, let the shocks proceed through the material that we filled the box with and come to bear on this little bit of fissile material.
We'll subject it to pressures beyond anything we previously imagined, and that would lead to new effects. That was the bomb in the box. When it came out of the conversation with Edward, it had not been a little small fission pellet that we were trying to influence, it would have been a package of thermonuclear fuel. And we won't try to transmit the energy by controlling the shocks moving through a box full of material. We'll have transmitted the energy by the flow of radiation within the box, applying to the package pressures that the radiation would apply. So that was indeed a pattern for a super — for an appreciable thermonuclear reaction in which a package had this unprecedented pressure applied to it, going to unprecedented density and perhaps put in a state to burn. Now Edward's idea of having radiation flow, having just come from discussions with the — what the hell was the name of that experiment?
Ford:The one at Greenhouse?
Ford:I don't remember.
Mark:George, wasn't it?
Ford:Yes, that sounds right.
Mark:It was Greenhouse George. Well, here one had been transmitting energy by radiation down a channel and imploding the thing at the end of the channel. And Edward thought it would be better in a box to have the energy flow by radiation. Well, it was natural for him to think in those terms because of his involvement with George. It was not too natural, because the smallest fission bomb we had at that time was the Trinity bomb, 20 kilotons and 10,000 pounds, 5 feet diameter high explosive. That was the Trinity test, that was the Nagasaki bomb. That was the smallest fission bomb we'd ever fired. Well the radiation doesn't flow from a bomb like that very much, and so Edward thought of having a radiation flow slip through neatly wasn't terribly well based. We didn't have a bomb that would have that effect. We quickly got one though, through the work that Edward disapproved of, which was pushed by [Marshall] Holloway and company, who were trying to build smaller fission bombs — fission bombs in which the energy per unit weight is a great deal higher. Those opened up the prospect immediately of having temperatures which you could tap from outside the bomb, and which flow through space. [missing words]. So the Trinity bomb was not the source that Edward called on to distribute the energy in the box. Neither was Ulam's pattern a good one. Because that would have been one helluva shambles to work with, trying to steer shocks with their unknown, variable, pernicious velocities through different materials, bringing them to bear on something over here in the way that it sits(?). The radiation did that beautifully. The shocks would have been a whole new year's work to find out if you believed yourself at all, and then to find out if it did work that way. It would have been a nightmare, but a manageable nightmare. The energy densities were there. High explosive has a density of about six-tenths of a gram per cubic centimeter, so a kiloton of high explosives occupies 600 cubic meters. But we were on the verge of building bombs which put a kiloton of energy into one cubic meter — easily. And we got the bombs smaller than Trinity. They did it very easily. So triggers for thermonuclear arms occupies spaces about like that [Carson held up his hands here, indicating dimensions of about 1 foot by 1 foot], for several kilotons. They give several kilotons per cubic meter instead of one kiloton per 600 cubic meters, so off the bat the energy levels are new and unprecedented and totally unfamiliar, and we could do things with them if we put our mind to it. All right, that's the generation of that idea, the bomb in the box Ulam brought in, taken to Edward, who began to picture thermonuclear [missing word] subjected to these kinds of pressures. Even he didn't think in terms of the fact that instead of one kiloton per 600 cubic meters we were getting several kilotons per one cubic meter. That would have simplified things for everybody [words missing] that length of time. We had to flounder around wondering how to transport energy by radiation through space. We didn't have a bomb that had that property at all. It was a guessing business, since we knew we could do with a lot less material and [missing word] than we'd been used to. And hence that characteristic of letting radiation flow copiously. That followed like [missing words].
Ford:Was Ted Taylor a principal in the design of the smaller, more transparent trigger?
Ford:Who were the main characters?
Well, there were a number of people. Ted was certainly interested in that subject and effective in it. I wouldn't want to name anybody in particular. The work on fission bombs at that time was going on in W-4, under Holloway in W Division. Holloway was practical and effective and would just see that the military really didn't want 10,000 [missing words] in their planes. They'd really rather have a quicker and smaller plane than the B-29, like the B-36 [missing words]. And that didn't want to carry Trinity style objects at all, because they couldn't be fitted in. Why wasn't the military more interested than [missing words] in bombs being smaller? [Missing words] immediately came through the thing in which the diameter was half the previous diameter. The yield was about the same. With the smaller bomb, there was less material that had to be heated by the fission explosion, less by a factor of 8 or 10 right away, and one could see it radiate; the temperature would engulf all of the materials in the inner layers inside the high explosion. We tried to build Edward's super with the fission devices we had, and we had a tremendous pressure over here at this corner. At this corner, cool as a cucumber. Until a shock wave came by and [missing words]. Unless we shaped the shock wave successfully to do what we were trying to do. [Missing words] and that's what Ulam brought in, and I'm glad it didn't get carried further, because it would have been in a miserable, horrifying job to get any feeling of [missing words] what we were doing. Now of course had we started to do that job, we would have come immediately to Edward's way of going about it and say, "Well this radiation flows so much more easily than a shock, and we'll use that." That's what we would have wound up doing. So it didn't take an invention to get the radiation implosion. We would have wound up making an implosion with radiation, no matter how hard you tried to do something else, because it was so much easier to picture — once you had a bomb small enough to make it feasible. We were at the edge of doing that — but not because of the super.
So there's the Ulam-Teller invention. Now who did what about it? Those two guys talked for an hour — nobody else was there — they wrote a joint paper in which they discussed the feasibility of having a large package of stuff that you turn the energy from a fission bomb loose on and compress it like crazy, and maybe it's thermonuclear fuel, maybe it isn't. That led to the idea, which was taken up immediately, of the equilibrium super. You guys had been working all the time on the classical super, which could also have been called the runaway super. In order for it to have a chance to work at all, the temperature of the material had to break loose from equilibrium. If you take liquid deuterium and figure out how much energy is released per cubic centimeter of liquid deuterium if you go through all the nuclear reactions it is capable of going through, all the dd reactions and all the dt reactions, you get so many kilotons per kilogram — not a terribly mysterious number. Well, we know how to do that. We even have some confidence in the arithmetic we got from the George shot experience that tells us we're at home in working with these temperatures and energy densities, no one [missing words] you're talking about. Well, how do you describe that breakthrough, to whom do you attribute the credit? I told you all I know about it, which was a conversation one afternoon with Ulam and again the next morning with Ulam, after he talked with Edward, who brought the results of their joint conversation. How their joint conversation went I do not know. I know as far as Ulam is concerned what he was prepared to take with him. I know that Edward was able to bring the radiation transport picture with him. Ulam was not thinking in those terms, and Edward wasn't thinking in the terms of the fission bombs we had. Ulam's scheme …
Ford:Was it you yourself who added that element of clear insight, the need to change the nature of the fission bomb trigger from what was then available?
Mark:Well, I've only done that in thinking about this question. I don't know that it was ever clearly brought out in the open in the lab in that way. It's just clear when you go back and remember what kind of problems we had, but you couldn't have started to talk about … Well, you could talk about radiation coming off the surface, and we did talk about it. I think we got most of the bomb material up to the temperature of a volt [kilovolt meant?], something like that.
Ford:I was trying to pin down in my mind who the principal people were in T Division who were working on H bomb matters after the Teller-Ulam idea. Marshall Rosenbluth was there, was he not?
Ford:He was a key figure?
Ford:And Conrad Longmire?
Mark:Also. And Burt Freeman, who was with you, was another key figure. John Reitz was there throughout the whole story, before you came and after you left. He worked with Marshall [Rosenbluth] on the equation of state of thermonuclear fuel materials like deuterium to begin with, like hydrogen gas, and like lithium deuteride.
Ford:Once Wheeler's group moved to Princeton, did the division of labor kind of fall out very naturally, or was there some semi-formal understanding of who would do what?
Oh, semi-formal. There were lots of questions and there were different people standing in different relations to the various questions. John Reitz was in a good position to handle the atomic physics that went into the equation of state. And of course Rosenbluth and Reitz were together on one aspect of that. Burt Freeman, who had been working with you guys, came across and did some very nice work on the relation to the transport of radiation and the box with the secondary in place, getting an effective mean free path for radiation, its geometrical mean free path. If you take this and the geometry, how far can the radiation go without meeting any obstruction — the geometrical mean free path? John Reitz was of real importance on the equations of state which we needed now, entirely new temperature and pressure ranges. Rosenbluth and Longmire were important, and they always will be in such a collection, standing out. Edward had stirred up the need of a second lab to handle the vast array of new questions which he didn't suppose Los Alamos could handle fast enough, so he would need a new lab and new people to keep the new work moving in such a way to give any chance of staying ahead of the Russians, and hence came the Livermore Lab. They took shape starting in July 1952 in response to Edward's fomenting, from Commission and the GAC and everywhere else he could reach. It took a lot of money, a lot of effort, and then they had to get into business, cut their teeth for business.
They made their first serious experiment on the new approach to thermonuclear gadgets. They were formed in '52 in order to make sure we kept ahead or apace with the Russians. They made their first experiment in 1956, and it was a complete flop. It gave 100 kilotons when it was advertised to be in the megaton class. Edward had had a large hand in the direction of the work of that lab, and Herb York was the nominal director. — half a day's thought, the reason for the disastrous failure of their [Livermore's] shot. I think he brought this to the surface at about the same time as a very [word missing], very bright guy [words missing] one of their people put his finger on the cause of the trouble. I believe we isolated the trouble first before they did. But I know there was Walter [Goad] who isolated it for us, and it was in their having omitted to pay proper attention to the way the neutrons would behave in the miserable geometry they had(?). You were asking who did important work here. Well, that was fairly important work that Walter did after their shot failed and before they [words missing], and it has to be seen to in every design: Well, that was certainly an important component of the work that had to be done here. Walter was not the only person in a good position to reach that question, but he was in a position to reach it and he did. George Bell also could, of course. [Words missing] neutronics as he was. Really our success, whatever success that we had, bore on all the people I've mentioned. Not only, but identifiable. What else do you want on that subject? You can't make any use of it anyway.
Ford:A question about the final yield of the Mike shot. The final predicted yield from the Matterhorn calculations was 7 megatons. The measured yield was a little over 10. I had always thought that this was merely that we had been somewhat conservative in our approach and that 30 percent was within a very reasonable margin of error, considering it was a totally new concept.
Mark:Oh, that I would have to agree with.
Ford:But Wheeler believes that we left something out, and part of the reason for that 30 percent difference is that we failed to take into account an additional source of energy release in the lithium, and I don't know what that is. Do you have any recollection?
Mark:There wasn't any lithium in the Mike shot.
Ford:Oh, you're right. I'm sorry. But he said we left out some additional source of energy, he believes, in the calculations we did.
Mark:Now if you're talking of the Bravo, it might have to be lithium. Bravo was the first shot we made with solid fuel. Mike was liquid fuel.
Mark:Bravo was — what? — 6 percent lithium 6, something like that. Something like that.
Ford:Was that in 1953?
Mark:No, that was '54. [Said '52 on tape, a slip of the tongue.] It was the first one after Mike, and it went off quite a bit higher than our calculations called for. More than three megatons of ten. It was a 15-megaton deal, and we estimated 7. And the cause was the source of energy that we didn't take into account in the lithium.
Ford:OK. That must be what Wheeler is remembering.
Mark:Yeah, I would think so. It really was a factor in that case. That doesn't relate to the Mike shot, which gave about 10, and I think it had been estimated to be like that, but nobody put any weight on the number came out for that. There wasn't any lithium in it; it was pure deuterium. There was lots of room for a few megatons displacement in it. You had to pretend that you had the geometry, and had everything in its right place. However, the Bravo shot, which was in March of '54, gave twice the energy that the arithmetic had called for, fairly nearly. It gave 15 and we'd called for 7. Why was that? It was a complete mis-estimate of the amount of fast neutron interaction with lithium producing tritium. We knew that 14-MeV neutrons on either lithium 6 or lithium 7 go through an n-2n reaction, and that gives you a triton plus whatever else you get. That gives it to you very fast. Thermalized neutrons give you a triton from lithium 6, but that takes time because they have to be thermalized, their energy's reduced, and they diffuse rather slowly. 14-MeV neutron give you a triton from lithium 6 with no time for slowing down or diffusion. They do an n-2n reaction. The slow neutrons do it by n-2n, and every capture leads to a triton. The fast neutrons collide with lithium but not every reaction gives you a triton. You don't get an n-2n reaction every time. There's a cross section for the n-2n reaction which is quite a bit bigger than we had been prepared to use, and that speeds up the production of tritons, both increases the amount and advances the time when they appear in a way we hadn't allowed for and that made an appreciable difference in the estimate of energy. I don't have that absolutely, definitely in mind, but that was the nature of the picture.
Ford:That's interesting, because I think that clarifies what Wheeler was thinking about.
Mark:Yeah. I want to say a little more. [Among the questions posed in Kenneth W. Ford letter of 1/16/95 to Carson Mark: When was the Family Committee formed? Who were its principal members? What was its charge?] The Family Committee was an ad hoc committee formed here, I would have to guess, in the fall of 1949 which contained just about essentially the division leaders of those divisions which were likely to have or did have a lot of work to do in connection with the super. We didn't know what the super was at that time, but we knew that T Division would be much involved. We knew that the chemistry division would have a lot of new work to do in handling the materials. We knew that there would be new work in physics, experimental physics. Those are mostly it, so there was myself and Edward, Joe Gamow, Stan Ulam, and I've pretty much forgotten who else, except Marshall Holloway, who was in charge of the engineering, and Eric Jette, in charge of the chemical work. I think Jerry Kellogg had to be there, or else maybe Taschek for any experimental physics that was called for. It was a catch as catch can collection of such people for such reasons if it was thought that the super work would impinge on their division's efforts. They met every week or so and talked about what they saw coming.
Ford:Did you chair the committee?
Mark:No. I don't know who did. Darol Froman took part in it. He'd be more likely to have chaired it than I, as he represented [Norris] Bradbury directly. Those division leaders, plus McDougal from high explosives and a few other people, more or less implied by the ones I mentioned, talked about the anticipated problems introduced by work on the super. Well, in that way the work on the George shot got attention in the Family Committee, and the work on the Item shot, which was the booster, got work in that committee. These both involved handling thermonuclear fuels and involved a rearranging of the fissile material for a normal bomb. The Item shot was a spherical bomb, but it had to have a large bore in the middle to contain dt.
Ford:Why don't we just get a few more remarks about the Family Committee. I'll tell you why I was so interested in that subject. I attended as a guest occasionally during the year I was there when I or Wheeler's group had something to present, and I was so impressed with the way in which this was working, bringing all these different people from different parts of the lab together. The modern jargon would be "matrix management." It was a wonderful example of focusing many different people from different areas on a single problem.
Okay. I was never so impressed with it [laughs] as part of it. It just seemed so natural that it didn't require comment. I've told you all there was to say. The people who were expected to be involved were there, and it was known why they were there, and they added that part into the discussion, telling the rest of the group that they were going to take longer to finish something than other people had been expecting, and things like that. Very natural. The George shot required a lot of straightforward fission weapon engineering. It had an unfamiliar shape. I was presumably there in part to discuss the likely efficiency of this implosion which led then in turn to the temperature that we would have available at the end of the pipe. And so on and so on, not very well planned but understandably [word missing]. That was one of the jobs of the Family Committee. It had similar jobs in connection with the booster shot, which again involved integrating liquid dt into a fission bomb assembly, which meant a different assembly than anything we'd ever seen before, and other differences. And the attempts to figure out how the implosion would proceed through a different set of materials than we were used to, and the neutron history of the explosion, totally different from anything we were used to, and the interaction of that with the dt was T Division business, so I was naturally there for that. All right. Somewhere in the preparation of the George shot, which involved really a lot of rather sticky engineering problems.
All new problems, all trivial, mechanical problems which couldn't interest anybody except the guy who had to do them — and they could involve him in going mad — were handled administratively by Marshall Holloway and people who worked for him. They were the ones who had to know how thick the wall was and how long it was, the other details of the metals — what was their radii, what was their thickness, how did they tie in with the rest of the material? Marshall had to see through getting all of those things attended to. He's a slightly conservative guy — or was when he was alive — so if he thought it would take a week from start to finish he was inclined to say he needed two weeks, or he needed more than one week, because he was aware that they could make a casting and if it got cracked they had to start over again, so he needed something close to twice the likely time to get a time that they could guarantee to work in. He insisted on that kind of approach to his problems. That's the way he was. He didn't always need that spare week or days, but he always had to allow for it, for his own personal need. So that would lead him into fights with Edward in particular. Edward didn't want to let go of the dimensions of the walls or of the gas until he had to. Marshall wanted to know the dimensions of those walls as soon as he absolutely needed to, so he said he needed those dimensions because otherwise he couldn't finish, couldn't prepare for the shot. Edward bridled at this, thought that if he put it off until next Monday instead of tomorrow what difference did it make.
Marshall said it made a lot of difference, because that would cut into the three days he was allowing himself to really do the testing if it was necessary. He tried to get comfortable to work, a time he'd laid out, and he felt committed to deliver in that kind of time—but not to be told, when he said he needed it on Friday, that he couldn't have it yet, you have to wait for a little longer, which was Edward's inclination. So he and Edward got into disgusting arguments for these reasons, just because of their difference in personality, approach. I had to side with Marshall, because I knew that Marshall was trying to make sure that if he said the thing, he needed the stuff on a Friday so that he could produce the results on Monday, and that he believed that, because he felt committed to doing that, whereas Edward didn't feel committed. If he'd said last week that he would have the results on Friday, he didn't feel committed to have it on Friday; he felt Monday would be just as well. And so he would have spits and spites and spats with Marshall Holloway in the middle of this committee on points like that — which made Edward say and feel that Marshall wasn't enthusiastic with the project, was dragging his feet, he was really concerned with other things and not trying to help. None of that was true, but in Edward's mind it was all true in spades. Whereas Duncan McDougal never got into such intimate wrangling with Edward and Edward concluded that McDougal was a white-haired boy that really wanted to cooperate with everything as much as he could, Marshall didn't really want to cooperate, was holding up everything he could and being a difficulty.
Then when, later, Bradbury had to decide what to do about the mechanical arrangements for the full scale test, Marshall Holloway said he would be willing to undertake that job and felt he was in a good position to do so. Bradbury was uneasy. "Can you work with Edward?" says he. Marshall said, "Well, I think so. It will be difficult, but I think I can." Edward of course refused to work with him. He [Teller] would not stay under those conditions, and that [he said] was not putting the lab's best effort into getting the job done. He [Holloway] was offensive and impossible to work with — a clown named Marshall Holloway, who was in fact a very responsible [words missing]. Edward didn't think he could work with somebody who had been challenging him [words missing]. And of course they would have …