Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Thomas Lauritsen by Barry Richman and Charles Weiner on 1967 February 16, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/4734
For multiple citations, "AIP" is the preferred abbreviation for the location.
Early training as a physicist; quartz-fibre electrometer; high-voltage installations, above 200 kilovolts; high-voltage accelerator, especially Van de Graaff machines; cloud chamber; fission 1939; reminiscences of Niels Bohr; leaving Europe 1939; war work at National Bureau of Standards 1940; rocket work 1940’s; post-war rehabilitation of laboratory facilities; technological improvements after the war; learning nuclear physics after the war; nuclear spin; development of shell model; rotational model 1952; gamma-ray detection; changes in research styles; research plans for the present (1967).
I’d like to start off by asking you something about your experiences as an undergraduate here. You took a B.S. in physics. In your physics work, did you have specific courses in nuclear physics?
As an undergraduate, no — I would think not likely. There was, as I recall it, a graduate course, which I probably took as a senior, which was run by Millikan on the basis of his book, The Electron, which did get us into nuclear physics. But most of our course work at that time was of a more classical, formal character. There were no specialty courses in nuclear physics.
You would have electricity and magnetism, and mechanics, subjects like that.
What about this graduate course based on Millikan’s book, The Electron? Did he talk about nuclear physics systematically?
I think not systematically, but more as a general picture of how modern physics evolved. There is a subject known to pedagogues as “Introduction to Modern Physics,” which for many people embraces the discussion of the classic experiments that occurred after 1896 that made such a dramatic change in the whole picture of physics, which led particularly to quantum mechanics, but of course also to many other subjects in physics that now comprise what we call modern physics. Nuclear physics was certainly one of those, but I would imagine that Millikan would have emphasized more such questions as the discovery of the electron originally by Thomson and by Zeeman, the early speculations about the quantization of electric charge by Stoney and later work by C.T.R. Wilson and ultimately Millikan’s own work on this. I would think also that he would have talked at some length about the photo-electric effect and possibly also about his early work in cosmic rays, which was of great significance. I think probably most of what nuclear physics students learned in those days about nuclear physics came from private reading of Rutherford, Chadwick and Ellis. I know that in my graduate time, this was the kind of book that we read and had seminars about. Also at that time were the Bethe, Bacher, Livingston articles, 1936 and 1937. And they were professionally required reading for every practitioner of nuclear physics for a long period of time. But these I think at that time were not the subject of formal courses.
You mentioned students in nuclear physics — you used that term. Did you mean physics students? Were there any students specifically studying nuclear physics as a specialty? Certainly not as undergraduates.
Not as undergraduates. I think I said graduate students.
But no undergraduates.
As undergraduates, no. We were physics students or we were engineering students or we were biology students. And as undergraduates, we had, I think, no exposure to nuclear physics — in fact, probably very little exposure to quantum mechanics even. That was regarded as a graduate course at that time.
How did you decide to build the quartz-fibre electrometer? What applications did you intend for it?
I didn’t invent that. It was my father [Charles C. Lauritsen] who devised it. My own involvement with it was purely financial. I made them and sold them as a graduate student, but I contributed nothing to the design. Possibly you have it already from him — the genesis of the idea. In retrospect, it’s a reasonably obvious idea. One knew about the gold leaf electroscope, which was a classic instrument of radiation physicists before the geiger counter. The quartz-fibre electrometer is a very much miniaturized version of this and a device which is very much more rugged and more reproducible. It had the virtue of an extremely low capacity. On the other hand, it had a rather poor voltage sensitivity. The criterion in those days — and I guess still today — for sensitivity for an electrometer for such questions as radiation measurements was the sensitivity to total charge; and this is then a function of the voltage sensitivity and the capacity. The best-known instruments of that time — very fancy electrometers — were the Dolezalek and the Hoffmann. These were the classic instruments that had been developed toward the end of the last century for very sensitive electrical measurements. They were instruments which had very high voltage sensitivity. With a mirror deflection, one could deflect a beam of light and could observe voltage changes of, I suppose, some millivolts. On the other hand, they were rather large instruments — dimensions like several centimeters — and had a relatively high capacitance. The instrument that my father designed went the other way.
It was only about 14 inches long or so.
Yes. The important element of the instrument, however, was only a few millimeters in size, which meant that it had a capacitance that was some orders of magnitude smaller than the capacitance of the more classical electrometer. It also had the virtue that it was rugged. It had the virtue that it was comparatively simple to build and easy to put directly into the ionization chamber. Again, old-fashioned radiation measurements were made by hooking an ionization chamber by means of a cable to an electrometer. This device shares a property with the gold leaf electrometer –- it has the ionization chamber as a part of the instrument. Well, what it did was to make it possible for comparative amateurs to build instruments with which they could detect extremely minute quantities of radiation at a time when the more classical instruments were pretty hard to build, pretty hard to handle and pretty cumbersome and pretty expensive — this was always a point. So Charlie built a number of these which were used for the first experiments in nuclear physics here; these were in fact made with various modifications of this quartz-fibre electroscope. You asked about my involvement in it. A number of people in the business were attracted by the simplicity of this thing and asked Charlie for copies of it. And ultimately he trained me to build them, and I sold them for what I thought at the time was a magnificent profit. I continued that for several years as a graduate student and ultimately turned the business over to a shop here in Pasadena, who, as far as I know, are still making them for sale. But it’s always been a very small business.
This was patented?
Yes, there was a patent which was assigned to the Institute on which I paid a ten percent royalty during the time that I was making these things. And I believe that the man who has been making them for the last 30 years, nearly, also pays a royalty. There are a number of similar instruments that have been developed on the basis of the same patent. As far as I know, there have never been any negotiations about royalties on that basis.
Who were some of your customers?
Almost exclusively universities. At the first, almost exclusively people who were going into the nuclear physics business or who were in the nuclear physics business. I think Merle Tuve was probably one of the earliest to get one of these. I think he got it from my father, however, before I was even in the business. This is Tuve of the Department of Terrestrial Magnetism, Carnegie Institution in Washington, who was at that time — this is now 1937-‘38 — in the nuclear physics business with a cascade generator, actually a Tesla coil type instrument. I think some also went to Berkeley to various people up there. I suppose in all I must have sold a hundred or two-hundred of them ultimately also to biologists and people of that kind who could use them for tracer studies or whatever.
Did you maintain any list of your customers? It would be interesting, if you did, to have a look at it sometime.
I doubt if I have such a list. I got badly burned on keeping my books in too good shape. At about the end of the first year or so — it may even have been two years — of my career as an industrial giant, I found out there was something called sales tax; and so to try to make a clean breast of the matter, I went to the local agency with my books, which consisted of carbon copies of all the bills. And he said, “Well, that’s fine. You keep good books. Now all you have to do is pay back the sales tax on all these sales.” It wiped out my profits for the entire year. So I don’t have those books anymore.
How much did you sell each instrument for — do you remember?
For the historical record, how much did it cost to produce?
It depends upon what you think my time was worth. In those days one regarded ten cents an hour as pretty good for a physicist.
Did this take much of your time away from other pursuits? You were doing graduate work and engaged in research while doing this, too. So how much of your time was devoted to this?
Well, it’s not the kind of thing that I would recommend as a practice, but, no, I don’t think that it was any serious loss of time, and there were many rewarding parts about it. Some of the first kinds of things that I did that might be called research I did with these instruments. For example, Charlie put me on to making depth dose measurements, using the big X-Ray installation. One of the problems for which this high-voltage installation was set up originally was to discover whether there was any virtue in going to higher voltages. At that time — this was now about ‘34 or ‘35 — the highest commercially available X-Ray installation was 200 kilovolts. One also had radium, which goes nominally to about two million, but in which the principal lines are down around 700 kilovolts. And there was some contention about whether radium was better than 200-kilovolt radiation. And there was considerable interest in having reliable sources that would go to a million volts or maybe higher. Charlie made some calculations based on the Klein Nishina formula, which were published, which suggested that from the point of view of depth dose, there was no advantage, or little advantage, in going into high voltage therapy; that particularly if one used broad beams, that the ratio of energy absorbed in the deep layers to the amount of layers absorbed superficially was essentially independent of voltage in this range. One might have thought otherwise.
After all, the higher voltage is more penetrating, and so it was natural for the medical people to feel that the more penetrating the radiation, the better. But, as Charlie showed, there is a limit to that; and what you really are interested in is not the amount of radiation that you can get to a deep-seated tumor, but rather the ratio of the amount that’s delivered there to the amount that is absorbed in the overlying layers, particularly the skin. For many situations, the skin is the limit on how much radiation can be delivered to such a deep tumor. Well, I was given the job of making measurements with phantoms which would approximate the chemical composition and shape of the human body and to measure the ratio of surface dose to depth-dose under as nearly biological conditions as possible, and to do this as a function of the energy of the machine and of the portal size. And, in fact, I found out what Charlie had expected: that for large portals, there was even some indication that the depth dose was a little higher at 400 kilovolts than it was at a million; that, in fact, 400 kilovolts would be a better choice for a commercial, generally available installation than a million. Well, this I did with these electroscopes.
When was this? Were you already in graduate school?
I think this must have been my first graduate year, which would put it about ‘37, I think.
Because the joint publication of the simple quartz-fibre electrometer was ‘37, and you went to graduate school in ‘36. You mentioned your father’s work on this around ‘34 — this paper that he had written on it.
I’m just trying to get the sequence here.
There should be a paper called “Depth Dose Measurements.”
That’s right. It follows in ‘39.
In ‘39. That work was probably done in ‘38 — I’m not sure I remember precisely. Well, you were asking me about the influence of the electroscope business on my graduate career and on my time. This was one instance. I had to make quite special devices for this purpose. There was at that time a great deal of uncertainty about how you actually measured dose to tissue media, and a lot of people were working on this from the theoretical point of view and also from the experimental point of view, trying to devise instruments that would actually measure the amount of energy delivered to tissue. Obviously, the best way to do that is to make an electroscope out of tissue. A compromise was found in making it out of plastic. So in these studies I made a number of instruments which were essentially all plastic, in which the electroscope element was contained within a very small plastic-lined ionization chamber and then were fitted with microscopes and so forth in such a way that they could be comparatively easily moved around inside of my phantom, which was made of masonite, and made in such a way that the observations could be carried out in a reasonable amount of time.
So in this sense, being in the business paid off. It meant that I was reasonably skilled in making them. I didn’t mind breaking them because I had all of this equipment that was necessary in order to build them again. As a sidelight, I remember my first training on this I suspect must have been quite arduous for my father. These fibres are two to three microns in diameter and quite hard to see. We normally worked on a 40-power binocular microscope, cemented these fibres onto the stand, which was just a small piece of bent 5-mil wire; and then there had to be a cross hair on it, of course, so that you could see it projected on the scale of the microscope. This cross hair which was cemented on the end of the fibre, consisted of another piece of the same size fibre, which then you mounted on the end of a needle and you held it over there with your shaking hand under the microscope until it touched the end of the long fibre, put a drop of shellac on, cemented with a hot iron.
I thought that was a pretty hard job. After you’re through with that, you take a pair of scissors and cut this thing so it’s about a millimeter long — all of this under a microscope. I brought my first one to Charlie, and he looked at it: “Well, it’s all right, but you should put the cross hair on the end of the fibre, not on the side.” But having that then, and having a stock of fibres, I was running these electroscopes through 10 or 15 at a time during my biggest production seasons. This helped in my work. It also made it possible for me to maintain the laboratory supply of these instruments. I suspect there are still a half a dozen lying around that have gone through a number of periods of being destroyed and rebuilt. So I think it’s fair to say that it wasn’t a great burden on me. Besides one needed the money.
What else were you doing at the time? You mentioned, in response to my question about what nuclear physics you were learning, Millikan’s course; and then you described it, and it turns out to be a general survey in a sense which led up to current work. Was there any more specialized work which you got into after a while in graduate school on nuclear physics?
Well, I would say that that came mainly, as it probably does today, from the demands of one’s research activities. I suspect that once I had the major course requirements off, I spent most of my time in the laboratory here, working under my father and with W. A. Fowler and probably five or six other graduate students in physics who had chosen this particular field. The group of us had a sort of dual responsibility. One was to maintain and operate the high-voltage X-Ray installation. The other was to work with, and to make investigations on, positive ion machines, which were in another building. Perhaps I could put this in a more useful framework. I might say that the business of artificially accelerated particles opened up in 1932 with the demonstration by Cockcroft and Walton. At that time my father and some students, including H. R. Crane, were working on a high-voltage X-Ray installation. My father got into this from studying cold emission. On hearing about the work of Cockcroft and Walton, Charlie and the others involved changed over the machine to make it a positive ion machine instead of an electron machine; and within the next year had then embarked on a program of nuclear physics involving, among other things, artificial production of neutrons, artificial production of a number of radioactive materials, a whole host of things which I think you probably have in your bibliography.
And then they worked on lithium and a number of other things.
Right, right. Well, this was a time, of course, when technique was the thing that paced what one learned; and there were several elements involved in that. One was the accelerating machine — how one produced the high-energy particles — and I’ll come back to that in a minute. The other was, of course, the detecting equipment; and in those days the emphasis was first on the electroscope, a little later on the evolution of the cloud chamber. On the accelerating machine, what the laboratory had at that time was an alternating current machine powered by the one-million volt transformer installation that had been put in here for electrical engineering work. This meant, then, that the energy of the beam fluctuated 60 times a second, so that one had far from a monochromatic beam.
One had, in all these experiments, some kind of average energy of the particles; and one could only guess with some difficulty what the influence of that energy was. Nonetheless, there were two, and probably three — at least two — such installations built mainly using old gasoline pump cylinders for insulation and steel electrodes which were fabricated for the purpose and positive ion sources of various kinds. And for a number of years then, the major part of the laboratory’s work was conducted with these alternating current instruments. At about the time when I was finishing up my work on the X-Ray studies and getting in more to the nuclear physics part, we became quite interested in the electrostatic accelerator that Herb Parkinson and Kerst had built in Wisconsin. They published a paper, I think in ‘37, describing that instrument and showing what beautiful work one could do with it. It was quite obvious from their work that the things that we were doing were crude averages over energy and that by having the energy very accurately under control — and in these early days already it was a fraction of a per cent — one could learn lots of exciting things about nuclei. So we decided that this was the thing that we had to do also, and somewhere or other Charlie got some money for a tank and various other materials, and we started to build a Van de Graaff. This was in ‘37 and ‘38.
The “we” in this case is the group that you mentioned.
Largely my father and Fowler and I. I think we were the main three at that time. During the summer of ‘38, Charlie went to Copenhagen and got involved in advising the Finsen Institute there (radium institute) on high-voltage therapy and also interested Bohr in putting in a high-voltage installation for nuclear physics at the Institute for Theoretical Physics. During that summer, Fowler and I completed the machine and got it running after a fashion. My thesis work was done on this machine with the cloud chamber that Fowler had built. Fowler and Charlie I guess had built it together. And this represented our first use of a highly monochromatic beam. It was really our first move into precision nuclear physics. It’s obvious that we didn’t build this in the dark. We visited Wisconsin, and Minnesota was also building such a machine, and we got a lot of help from those people.
Who at Minnesota — do you recall?
Oh, Johnny Williams largely. I’m sorry I can’t produce other names just immediately, but the names that I associate with the early days of useful Van de Graaff machines are largely Herb and at a slightly later stage, the group at Minnesota. Tuve had of course somewhat later a big open-air machine which is running to this day, but the crucial thing in the Van de Graaff business was to put it in a pressure tank. This was Herb’s invention. And at the same time to do all the hundred other things in order to make it a machine that you can turn on and run with.
May I ask two questions in connection with this? First, what were the problems in pressurizing the tank and making it work? Was the technology for that good at that time or were you inventing it as you went along?
The problem may have been a psychological one. The problem is you build an instrument and then you put it inside a tank where you can’t fix it. Today it’s a matter of many hours, more like a full day, lost every time you open a tank. It’s necessary to dry things out very carefully and so forth. So that it was something of a change from the old days, where if something started to spark, you’d shut off and you climbed up and poured a little sealing wax on it. It was for physicists I think at any rate a new experience that now you had to build things that would run and run without attention. So this was one of the things. And in those days I think many people, instead of comparing baby pictures or instead of comparing new physics discoveries, would compare length of time since last entry into tank as a criterion for how good one’s engineering had been. A more basic problem was, however, certainly the problem of maintenance of high-voltage gradients on commercially available materials. A lot of people used Bakelite in those days because Bakelite was a comparatively easily machined substance and was commercially available in many sizes and shapes. It turned out that there are many different kinds of Bakelite and one had to learn that. I think it was probably also Herb who applied some knowledge that the electrical engineers had had for some time that if you want to support a high voltage across an insulator, you had better divide the voltage and fix it so that it is really distributed along the insulator deliberately rather than leaving it to accident. This led to building machines like layer cakes rather than building long, slender insulators; and this of course is standard in technology now. But these things took working out. We were of course in a very good position because we had Herb’s experience to go on. On the other hand, one always makes changes and one always learns something from those changes — usually that one shouldn’t have made them in the first place.
Excuse me, on this, did you have a machinist working with you or was this work done pretty much on your own?
The physics machine shop in those days consisted probably of one or two men who mainly worked on instruments rather than such clumsy devices as accelerators. So I would guess that we got some help from them on some of the more difficult parts, but I think it’s safe to say that we drilled all the holes and put in all the bolts ourselves essentially. As a matter of fact, I recall that in order to save money on the tank, we made ourselves all of the various portals and things of this sort that had to be welded into the tank. So we spent many a long night here with our big lathe turning rungs and flanges and parts of this kind which were then incorporated into the tank at the factory. We bought this tank for $1500, I think, something like that.
Where did the money come from?
I don’t know that. Ultimately from Millikan, I’m sure, but whether it came via the medical business or some other way, I don’t know.
Well, there are ways to check on it.
What were the approaches to achieving a relatively homogenous beam at that point in time? By 1947 you have an electrostatic analyzer, but in 19…
Well, the point of the Van de Graaff is that it is essentially a dc machine. That is to say, it operates with a constant terminal voltage. The older machines, as I said, were powered by a transformer, which means that the voltage alternated 60 times a second. So one already was starting down the right street from our point of view by having a machine with a dc terminal voltage. The other thing about it is that it’s in the nature of the beast to be very stable. For example, one of the features which controls the voltage of a Van de Graaff is the rate of loss of charge by corona. This is a very steep function of voltage above a certain threshold, which means that it’s automatically very highly stabilized; so that even before the invention of automatic stabilization, one was able to do half-percent or several tenths’ percent work over long periods of time just by virtue of the inherent stability of the machine. Now, during the war, the Wisconsin machines were moved to Los Alamos, and I think it must have been at that time that the feedback system using a corona triode was devised — again by Herb’s group. This made it possible for one simply to set the voltage of the machine provided one had some way of measuring the energy of the beam. This was normally done by a small magnetic deflection. We incorporated this into our machine immediately after the war. I don’t think it was available before the war. But this was a big step forward in improving the usefulness of these machines. You mentioned the electrostatic analyzer. This was — and I think is still — rather a means of controlling the machine than a means of selecting the particles. That’s perhaps a little obscure. One can, with a beam of particles which contains a large number of energies, select out a specific energy by deflecting the whole business in a magnetic field and taking only that part that has a given radius of curvature. It turns out, however, that in a well-regulated Van de Graaff the beam itself already is sufficiently homogenous, so all you have to do is deflect it through a magnet and use the amount of deflection as a measure of the energy and as a correction.
But the electrostatic analyzer also used an electric field.
Yes. The electrostatic and the magnetic analyzer were two different ways of doing this. At the time that we were first interested in the electrostatic analyzer, we felt it had some advantages because electric fields were easier to measure than magnetic fields. That was something that was changed later. But for a long time it was a very attractive thing to us, at any rate, to simply use a voltage divider and a potentiometer. This meant that we could reproduce voltage settings on the analyzer to a tenth of a percent with no problem at all. And later — I guess Herb also was the first to do this very systematically — one built electrostatic analyzers with well-known dimensions which then could be used for absolute calibrations. Our analyzer was somewhat crude, and, in particular, we didn’t know the dimensions of it very accurately, so we always calibrated against some standard or other involving the beam. But magnetic analyzers and electrostatic analyzers now are both capable of being made with standard well-known dimensions and of being used as primary standards for voltages.
When did the pressurized Van de Graaff first become useful for actual research in nuclear physics? I’m talking about the instrument you built here. When did you first start getting results from it?
Oh, late 1938 or early 1939, I think.
What sort of work were you doing?
Let me interject here that there was another purpose in building this machine in the back of Charlie’s mind and one which we did give some investigation, and that was as a source of X-Rays. He thought that this might provide, if one wanted it anyway, a good way of making X-Rays at around a million volts — and indeed it was. Our first tests with the machine were, in fact, with the voltage reversed, and we made, for us, we thought, quite fantastic amounts of radiations of a million volts energy. And, in fact, for some time this application of Van de Graaffs was a very serious concern. As a matter of fact, a number were made — and I guess are still being made — for the purposes of providing X-Ray sources for X-Ray photography.
Was this also going on in the Cambridge area when Van de Graaff was still at Princeton? It seems that there was an operation in Cambridge.
There was indeed, and we also were in communication with the people at Cambridge. I remember visiting Trump, who was in the electrical engineering department, who was interested in making Van de Graaff machines more reliable and more stable. He also was interested in doing the basic research that was necessary but that few people were willing to do on what determined the voltage breakdown at various surfaces. Did one have to polish the electrodes? Was it best to make them of aluminum or gold or what? And certainly that group contributed over the years a great deal to the fundamental understanding of what made these machines run. I guess I said that putting the machine into a pressure tank was the first big step in making it useful, that the invention of the feedback device was the second. I would say that the third is probably the very careful engineering that the M.I.T. group did, resulting in the ultimate production of a commercial product via the High Voltage Engineering Corporation. The existence of the series of designs that this group had built up, with Van de Graaff himself and with a good many of the M.I.T. people, had made it possible for a lot of people to get into the business of using such machines without having to be in the business of building them, and I think it’s had a tremendous impact not only on people who were already in nuclear physics but on 250 small schools in various parts of the world who have been enabled by this means, by the machines, to get into business. You were asking about the first work that was done on our machine. I think that probably was my thesis work, which was an investigation of gamma rays resulting from bombardment of fluorine by protons. This was a reaction which had been studied both here and in Cambridge, England. And there was a considerable discussion about where the gamma rays came from. There were two hypotheses, one of which would imply that the energy of the gamma rays changed with bombarding energy, and the other that it would not. My job was to determine the energy of the gamma rays at two or three bombarding energies. The only instrumentation one had for this purpose in those days was the cloud chamber, which was an exceedingly tedious way of measuring gamma ray energies. One measured the curvature of positron-electron pairs, tracts in the cloud chamber, sat down with pieces of paper, and measured the radius of the circle that best fitted the track, and did this thousands of times. This operation, I suppose, must have taken the better part of the fall and the spring of 1939. I wrote my thesis in the spring of ‘39 and took my degree in the summer of ‘39. After that, Fowler and Charlie continued to use the machine for studies of nitrogen with protons and a number of other things, some of which were eventually put together and put into the published version of my thesis.
Was the cloud chamber used the one that Fowler had built?
Yes. Fowler had become a world expert on cloud chambers by that time. Cloud chambers were — and I guess still are — a little bit magical. People who run cloud chambers have their own special bottles of goo that is known to work and they won’t use anybody else’s goo.
For sealing, you mean?
For example, for gluing the black cloth down onto the piston — things of this sort. It turns out that they’re exceedingly sensitive to the most minute kinds of contamination if you get the wrong kinds of materials. And the art was in just learning these things.
Was there much collaboration here between the nuclear physics group and the cosmic ray people — say in the area of detectors?
I guess I don’t know specifically. I’m sure that Fowler must have had a lot of discussions with Carl Anderson, who had been building cloud chambers for some time; but I don’t remember any active collaborative work.
You mentioned other uses of the Van de Graaff once it was put into operation here. Was there a clear program of research defined for this group? Was it clear that your special area in nuclear physics would be of a certain type, or did the existence of the machine determine the types of experiments you could do?
I think it’s fair to say the latter. For a long time — to some extent even now — what kind of nuclear physics one does depends on what kind of facilities one has available. Charlie and Fowler had been interested in reactions involving the light nuclei. It was generally felt that because the Coulomb barriers were lower in the light nuclei than in the heavier ones, that with limited voltage available, one would be more likely to get physics out of the light nuclei. This certainly was a determining influence for a long period of time in the interest of this group in light nuclei. Another was perhaps that even with a dozen or so elements one found plenty of interesting things to keep one busy and plenty of exciting systematics. For example, one of the earliest things that was found — I think this is in Fowler’s thesis, as a matter of fact, before the Van de Graaff — had to do with the energies of various positron emitters in the light nuclei. This was, I think, the first evidence from nuclear physics that proton-proton and neutron-neutron forces were similar. This had to do with the masses of the so-called mirror nuclei, and already in that early paper, which is probably around ‘36, one can see a comparison between the masses of, for example, carbon-11 and boron-11 with an elementary computation of certain Coulomb energy corrections which had to be made and which pointed very strongly to the assumption that these two nuclei were really quite similar to one another. This is a point which is still fairly central in one’s interest in nuclear physics: to what extent are the forces between protons and neutrons and protons and protons and neutrons and neutrons really the same? So the light nuclei yielded this plum. The about the same time, yielded evidence for very strong resonances in the formation of compound nuclei. One thought that they were leading to some interesting information about helium. That developed not to be the case at that instance anyway. But there were many things that led one to feel that playing with the light nuclei would be fun, and we still think so 30 years later.
Did you have in mind then the need to test a particular model of nuclear structure?
Well, I think it may be better to say that in the very beginning one was trying different things. I mean it was a brand-new, wide-open field. Every day you went to the laboratory you found something new. So I think that if one were to say there was some system, one would probably say, “Well, there’s been a lot of fun in studying the gamma rays from this plus this. Let’s study the gamma rays from that plus that, and set up the equipment for it and do it.” And so you’ll find in those days a lot of papers on gamma rays, neutron groups, from boron plus deuterons, carbon plus deuterons, beryllium plus deuterons — things of this sort.
This is what you meant by a systematic approach.
A pursuit of one line based partly on the history that the first one had been fun, partly on the fact that one had evolved a technique for doing this job. Now if you want to study gamma rays, you sign an order for a gamma ray spectrometer and you study gamma rays. In those days some unfortunate graduate student would spend a couple of years learning how to do it and building the appropriate apparatus, so one was very much apparatus-limited. And I think possibly the reason that there was so much enthusiasm for accelerator building and for detector development at that time was that every advance in that opened completely new doors. I don’t think that it’s necessarily so that the theorist said, “Well, now, you need to build a machine for 1.3 million volts and you’ll learn something interesting.” But rather that at 1.1 or 1.2 all hell had broken loose, and so one could see obviously that by expanding in this direction on the basis of techniques that we understood, that one would learn more things. Now, this is not to say that we were not interested in the theoretical work and didn’t have a great deal of help. Robert Oppenheimer, for example, was keenly interested in the work here and was extremely helpful in making interpretations of what was going on. Like most people in the business at that time, he wasn’t right a hundred percent of the time. But it certainly was important that one had people with a background in theoretical physics who could at least warn people against making ridiculous mistakes. I would think that the thing that put the most jazz in the business probably was Niels Bohr’s description of the compound nucleus.
For experimenters this was a very attractive idea; and, in fact, he had a beautiful little model, a wooden model, a bowl with balls rolling up that you undoubtedly have in your museum, that he used to illustrate the point — the point being that if you roll a marble into an empty bowl, it rolls out again. If the bowl already has marbles in it, then the energy is distributed among the other marbles and it’ll take a long time before all that energy is again concentrated in one marble sufficient to pop it out. This was the idea of the compound-nucleus formation, and this was a thing in which we were mightily interested because some of the experiments that Charlie and Fowler had done, particularly on sharp resonances, very high up in beryllium-8, seemed to require such an interpretation. Efforts that people made to explain these very sharp resonances in other ways were simply not very rewarding. The classic case of this, I think, is the neutron resonances, which were for many years regarded as the best evidence that the nuclear reactions did in fact go through this process and had this dramatic feature in the slow neutron resonances, that they’re fantastically sharp. One simply couldn’t understand this on the basis of any direct interaction theory. So this was the guiding theory that I think motivated a good many kinds of experiments. But, again, I think it’s fair to say that many times one tried things out because one knew that something would happen, and the problem was after one had found out what happened, to try to understand it in some way.
That’s when the model came in handy afterwards in organizing these results.
How did the discovery of delayed neutrons in fission affect the understanding of the nucleus via the compound nucleus model?
It followed very naturally from the liquid drop model that was put forward by Bohr and Wheeler, I guess, in early 1939. I don’t recall whether that paper predicted the existence of delayed neutrons, but I suspect that it did. The point is, of course, in this model you divide the nucleus into a couple of pieces. These pieces have an inappropriate ratio of neutrons to protons for stability, so they must return to the stability line by converting neutrons into protons, and this means that you have a lot of quite odd beta-decaying elements, some of which will decay to states that are unstable to emission of neutrons. So the existence of delayed neutrons, I think, was no surprise. The question of how many there were was, of course, of considerable technological interest.
How did you first hear of fission? Do you remember the circumstances?
No, I’m afraid I don’t. I was probably not very much in this world at that time, finishing my thesis on this work, which had nothing to do with what was going on in the East. I recall the fission thing broke in the spring of 1939. This would have been at a time when I was reading 7000 photographs or something of this sort, trying to get a thesis put together, and studying for my oral examination, which in those days was a thing that you didn’t pass automatically. So I suspect I was pretty preoccupied, and I don’t recall a dramatic moment when I first heard of fission at that time. As I say, I was a little pressed. As a matter of fact, I left for Denmark on the same day that I took my degree here. I took my academic uniform off in the car on the way to the train, so there wasn’t this period of contemplating one’s navel and looking about one to see what was going on.
And you weren’t aware of any other reaction on campus at that time, either, then?
None that comes to mind immediately. Now, when I got to Copenhagen, of course, I found that there were people there who were working on fission problems, and I got involved in some of this work myself. But my primary job when I went to Copenhagen was supposed to be to assist in building up, improving the high-tension accelerating system there, in getting started on a Van de Graaff there, and in working with the Finsen Radium Institute in getting a high-tension system for radiation treatment. So these were the things that I was thinking about on this, my first job.
Your headquarters there were at the Institute for Theoretical physics?
Right. This was a summer when I guess everyone in Europe was away on vacation. I don’t know where they all go, but they were all away, so I was alone in the lab, and spent the summer putting together some equipment for which the parts had already been made as a part of this program. Of course this assignment changed a good deal as time went on. In the first place, the Finsen Institute thing went a good deal more slowly than had been expected. The Van de Graaff project was delayed somewhat. Other changes were more important to make in the high-tension stack. There was a great deal of interest in learning what one could about fission as opposed to these other sort of engineering efforts, and there was a war on. So, my picture changed a good deal in the course of that summer.
It’s interesting that you went over to do this. In other words, you went over as an expert, which is somewhat different from the idea of someone immediately with a doctorate going over to be a student. Is my impression correct?
Well, this is not an accident. My father had gotten to know Bohr quite well, I guess during his visit here, which must have been around ‘36 or perhaps even earlier. He may very well have known him even earlier than that. At any rate, we had rather close family connections. When I was a child we used to go to Denmark — I suppose we had done this three or four times in the course of my history up until I took my degree — and on some of those occasions we visited the Bohrs. At least I also got to know the professor very well. I was contemporaneous with his kids. I knew them very well. We used to build airplane models and go on trips and one thing and another together. In fact, I spent a good deal of time with the Bohr family — I think this must have been when we were in Denmark in the summer of 1932 just before I was to be a freshman here. At that time I spent a good deal of time with Bohr’s eldest son, Christian, who drowned in a sailing accident the following year, and I also got to know the other boys fairly well, although they were younger than I. I also know that Bohr had a very high regard for my father, and either he invited him or took advantage of his visit to Denmark in 1938. I mentioned that we were building the Van de Graaff here in ‘38. During that summer, Fowler and I were working on this, and Charlie and my mother were in Denmark. And it was at that time that Charlie pressed on Niels Bohr the idea that the Institute ought to get into the high-energy, high-voltage business, and also that he encouraged the Finsen Radium Institute, the big radiological outfit in Denmark, to get interested in somewhat more high-energy therapy than they had been doing. So it came about in some such way as this that Charlie got this thing started, and he said to Niels Bohr, “Okay, when my son will be finished next year, he can come over and help with some of the nuts and bolts part.” So Bohr invited me then to come as soon as I got my degree, and I got a Rockefeller fellowship –- I don’t know whether through his intervention or how. At any rate, I got one of those rare things and went over on that basis originally with that intent in mind.
Did you have much contact with him while you were there? Did he take a strong interest in the nuts and bolts?
Oh, yes. Bohr was a very enthusiastic experimentalist. He was interested in all of the details of any kind of experimental equipment. One of the interesting things to me was — I’ve forgotten now just how this came about — there was another long period when I was sort of alone in the lab, and I took the occasion to sit down and measure a couple of thousand cloud chamber photographs on fission fragments. And not knowing anything else to do, I classified the lengths of all the little branches. They look like little trees, you see. So what do you do, you measure the lengths of the branches and the angles and one thing and another. And I brought a big thick notebook to Bohr with all this stuff in, and I said, “Professor, what’ll I do with this?” And he rubbed his hands with glee. He must have spent a hundred hours worrying about what you could d0 with this. For him, an experimental datum was the most precious thing on earth. Although he had done experimental work himself, he nonetheless believed in experiment. He had enough background to know better, I think, but he felt that no amount of time was too much to devote to understand an experimental fact. And so he worried about this and told me more things to go back and measure and really made something out of it that I was very very happy to learn and very proud to have been a part of. But aside from that, he and the whole family were exceedingly kind to me. I spent a lot of time at their home while I was there in Denmark. Mrs. Bohr was, and remains today, one of the finest ladies in the world and one of my very warmest friends. They essentially took me over as a part of the family. The following summer, in ‘40, I spent most of the summer out at their summer home. I was very fortunate in another respect, that because of the war I was essentially the only foreigner on the place; so I had Bohr all to myself, so to speak. Bohr always had to have somebody to argue with. When he was talking, there had to be somebody there that he could shake from time to time or throw burning matches on or something. So I was that guy. So, as I say, I spent a good fraction of the summer of 1940 essentially acting as secretary, although I tried very hard to understand what he was doing, on his gigantic paper on atomic stopping power, which came out I guess nearly ten years later. It’s a classic now.
Was that what was generally on his mind, certainly that summer — you said that — but I mean in the year prior to that, was he talking about other things during that year? What seemed to be the thing that was of most interest to him?
Bohr’s main thread of interest I think at that time and for the years after was the question of how you make a world that’s fit to live in. I think that he was aware and sensitive to the implications of fission for how the war might go at a very early stage. I know that he was following quite avidly the measurements that were going on, both in Columbia and at other places in this country and in Italy, on the number of neutrons released. He would come bursting into the Institute one day and say, “It can’t be done!” The next day, “It can be done!” with equal enthusiasm. He was quite aware that if it could not be done, that would be a problem that one didn’t have to worry about. On the other hand, if it could be done, it certainly also had some very important implications. So I know that he thought very far ahead. This is already in the early fall of ‘39 when we were still able to get quite good communications and were still quite aware of what was going on. A little later that channel of communication closed up, although we didn’t become immediately aware of it, through the voluntary censorship in The Physical Review. We were for a long time, I guess, quite puzzled about why we didn’t get answers to questions and why articles that we expected to see didn’t seem to be appearing. So there was only a rather short time when Bohr had available to him the full world information on fission. But he thought of this, I’m sure, as something which might have a very significant effect on the war. I don’t know in what detail he felt this, but I know he felt it to be very important.
Had Meitner and Frisch been there and gone by the time you arrived?
Frisch was still there. He went off to England for a vacation and never came back. This must have been about in September, and I inherited his flat after that. I think Lise Meitner already had her job in Stockholm. I don’t know that, but…
She was not there when you were there.
She was not there at the time I was there. Frisch was actively working in the lab. He was our electronics expert and was involved in the experiments with fission, with the delayed fission fragments and things of this sort.
So then what originally started the visit, (where you would contribute to their development of Van de Graaffs and get them into this business) turned out, because of the discovery of fission, to be somewhat more varied, and you were thrown into the middle of some of these questions?
Oh, I don’t think that Niels Bohr ever intended to hire me as a mechanic. I think that he felt that I could do with a little learning, and he certainly worked very hard at that. I don’t think that changed in any fundamental way. What changed, of course, were some details about when we started some of these engineering operations; and, in fact, the Finsen Institute thing went on quite independently. Another thing that changed was the involvement in the fission work. Another thing that changed was that, as I say, I had Bohr to myself essentially, or he had only me to be his pupil. No, I don’t think in that regard my program there was unduly modified. Certainly I would not know any way in which I could have had a more rewarding time there aside from the not inconsiderable difficulties associated with the war.
The cloud chamber that was there at that time in which you did your work on the fission fragments — can you describe that a little bit, where they got it, how it was built, who was involved in it, and also was it any good at that time for measuring the range or other parameters on the fission fragments?
There was when I arrived a small cloud chamber that Professor Jacobsen had built. I’ve forgotten what he used it for, but I recall that he had been interested in natural radioactive decay and things of this sort. I believe that during the summer of ‘39, when I was largely working on the high-voltage installation, that I was also playing with this cloud chamber, trying to see if I could make it work. But in one way or another, I think at Bohr’s instigation, we decided that it would be fun to make some pictures of fission fragments. There had been some pictures made, I think by Curie-Joliot, of fission fragments which showed in particular that there were two; but these had not been very detailed and not more than of qualitative use. So we decided that we would try to see what could be done to make really good pictures. To what extent we used that small cloud chamber, I don’t recall. I do recall that we immediately undertook to build a big one, a nominal 12-inch diameter chamber, and that we put this together in something of a hurry and got operating with it. The man who was responsible for that was J. K. Bøggild, and he became very soon a real expert on cloud chambers — again, this witchcraft — and in particular was able to operate a cloud chamber on pure alcohol, which we all thought was excruciating because alcohol was getting hard to get. But we operated this cloud chamber at quite low pressure and were able then to get fission fragment tracks that almost literally would light up the room when they came in — tracks a good foot long. These were the tracks which I then subsequently measured. But it was largely Bøggild who brought that cloud chamber into operation.
By the end of your stay in Copenhagen, was there any greater understanding on your part and on Bohr’s part of what had begun to develop as far as the potential wartime applications?
I think not. I think that Bohr’s last information on fission was probably a report from Amaldi in late ‘39 with some kind of a number for multiplication. And I recall that he was a little puzzled — I mentioned earlier this lack of communication with this country — and I don’t recall, other than the purely scientific aspects, which were the subjects of these various papers that were written during this period on the stopping of fission fragments, that there was any discussion of reactors or bombs or things of this sort. I’m quite sure that nobody with whom I had contact knew anything about the effort that was going on. Well, there wasn’t any, as a matter of fact, at that time. You recall I left Denmark in November of 1940, and that’s long before there was any real Manhattan Project. I guess Fermi’s pile was the next year or later.
So we would not have had any of that kind of information. There must have been some of it going around, though, because in Holland the physicists got hold of all the uranium they could and all the heavy water they could and hid it away. You know the story from other sources, I’m sure. But there were things going on at that time. This was a little later than ‘40, I’m sure, but it must have been around the time of the occupation, when physicists were worrying particularly about heavy water, but also about uranium, and doing what they could so that they wouldn’t get into German hands. We had, I think, nothing of that at the Institute. There was some concern about gold.[Pause in recording] Ultra-violet radiation was used as a cure for many kinds of things, and, in fact, is still quite generally practiced. There are a lot of physiotherapists who use light. Well, Finsen was the man who first made the systematic studies on the use of light in therapy. I guess he got a Nobel Prize for it, and this Institute was then built in his name. It has more recently become a tumor institute in which one uses radiation and I suppose other things, surgical techniques and so forth, but it was a part of the hospital system of Denmark. But Bohr of course had very close connections with the people who were responsible for the medical side, the chief of the hospital and things of this sort; so that anything that could be done by physics for medicine, he was of course very interested in exploiting.
I think we ought to say at this point that we’re resuming after a brief pause.
And the last comments were in response to a question for the background of the Finsen Institute, a radiological institute in Denmark. When we interrupted for coffee, I had asked a question about the awareness in Copenhagen of the potential military applications of fission, and you had given an answer, and I think you were completing it. You had said something about the interest in heavy water and uranium elsewhere and people being conscious of the need to keep these out of German hands. You said that this didn’t affect the Institute, but there was some question of gold, and that’s where you left off.
Oh, that was a minor thing. There were gold medals and one thing and another around the place, and there was some interest in concealing those. I thought I was such a good mechanic, I offered to take it and roll it out into strips and put it into books or whatever; but it turned out that the medals that they gave me were not made of gold in the first place. They were some exceedingly hard material that I simply couldn’t manage at all. Ultimately, they were all dissolved in aqua regia, stored in bottles. Now, I don’t want to give the wrong impression about this question of the awareness in Copenhagen of the potentialities of fission. I’m only saying what I was aware of at the time, and, even qualifying that, my recollection of what I was aware of at the time. I think I should warn you that my recollection is exceedingly poor for historical purposes, and you can’t rely on it. But this was my general feeling, that Bohr was very acutely interested, was taken up by the whole question of how the world was going to come out. As you know, he was an inveterate optimist, and he would sometimes say, “Well, this is going to be long; it’s going to be hard; but when it’s all over, there will be peace sweeping the world for years and years and years and years.” And he certainly worked very hard himself to try to achieve that. It might be interesting for you to know a little about the political situation — well, about the micro-political situation which I found myself in.
I was there the first year as a Rockefeller Fellow. This was an appointment that lasted one year. At the end of that time — this was now July of 1940; Denmark had already been occupied — there were certainly very strong indications that I ought to go home. On the other hand, there were things that I was working on — particularly, the Van de Graaffs that I was interested in getting a little farther along — and I also had in the back of my mind the hope that some way I could be useful in getting Bohr out of the country. My father had already written me, as a matter of fact, as early as the summer before that he felt it was important to get Bohr and his family out of Denmark if there were any possibility of doing so. But Bohr said repeatedly no, that he felt it was important for him to stay in Denmark as long as he possibly could; and, as you know, that developed into quite a long time, and ultimately he did get out at considerable risk. As far as my own situation was concerned, Bohr had made arrangements with the American Embassy to keep in close contact, and I had an understanding that when they at the Embassy thought that the situation was too dangerous for an American, that they would let me know, so I was pretty relaxed about it.
There was, I think, in the early summer of ‘39, a big movement to get all Americans out over at Petsamo, I think it was, the port in the north of Finland; and I almost went out then. But at that moment, two days before, I decided to get married to a lovely girl I met in Denmark, and her parents didn’t think that two days was enough, so we hung around until November, when we got out by what turned out to be a very unusual and fortunate chain of circumstances. Someone — I think it must have been the precursor of Lufthansa — set up an airline between Denmark and Berlin and Barcelona; and I got a ticket on that for myself and my wife, and we flew out with no problem. The line was in operation for about two weeks, at the end of which time the Spanish border was closed to all men of military age; so had I been a little later, I wouldn’t have been able to make it. What other opportunities might have shown up, I don’t know. This is a little beside the main point. We went then to Lisbon and sat with a hundred thousand other people, hoping for a boat, and eventually that came.
How long did you have to wait?
I think it was only a couple of weeks. It was the custom at that time for the entire foreign population to meet in the chief steamship office every morning and stay there all day. It happened that I was there in this crowd when a cabin was released for this afternoon. I got it, so we managed very nicely. Well, this perhaps reflects a little, gives a little idea of the concern that Bohr had for me personally. I’m sure that he felt considerable responsibility for my safety there, and he certainly was extraordinarily helpful in making the necessary arrangements for me to get out of the country.
When you arrived in the United States, did you proceed directly back to Pasadena?
Essentially so. I learned when I got to New York that there was a possibility of a job at the Bureau of Standards, working on a war project there. I think my father came up to see me in New York or I went down to see him in Washington. You see, this is now November of 1940, and this is the beginning of the N.D.R.C., the National Defense Research Committee. From here, Richard Tolman and my father had already gone to Washington and had been spending a fair amount of time in trying to see what contributions scientists could make to the war effort. By November, Charlie had found out through some exchanges with the British that they were much interested in photo-electric cells and were purchasing very large numbers of them; and in one way or another, he deduced that this must have something to do with proximity fuses, and he started a project then at D.T.M. (Department of Terrestrial Magnetism) under Merle Tuve, whom I mentioned earlier, to start developing these. That work went ahead very quickly in terms of the number of people involved and the complexity of the operation, and a part of it was transferred to the Bureau of Standards, who had some space and set up a section for this purpose under Alexander Ellett. He is, I believe, now chief of the research division of Raytheon. He was a physicist from Iowa who had also worked in the Van de Graaff business during the preceding years. He set up a group there, of which I was one of the early members. I think I wrote my so-called examinations for the Civil Service in New York in November, and then my wife and I came to Pasadena for the Christmas season. I went back shortly after New Year’s and started my job and she followed me a few weeks later.
At the time then you had no institutional connection with Caltech?
And how long did you stay in Washington?
About ten months.
While you were in New York, did you see any of the Columbia people? Did you visit any of them?
I don’t recall. I believe that New York was a matter of two or three days. I was much involved … I had to go to the Rockefeller Foundation, for example, and deliver a report there. I guess I was one of the last victims out of northern Europe, and they were quite interested in knowing what the political situation was like; so I spent some time there. My wife, I recall, was detained on Ellis Island for some time. As soon as that was fixed up and as soon as I had got these forms out of the way for the Civil Service, then we picked up and went to Pasadena. I don’t remember any scientific contacts in New York at that time.
Then you spent, after returning from Pasadena to Washington, ten months there, and then you went back to Pasadena.
Yes. In the evolution of that project, which continued at the D.T.M. and at the Bureau, one reasonably soon had a good design for proximity fuses, particularly for bombs, and that part of it was going well. We were also interested — the British were interested — in fuses for rockets. The British had a very fine rocket at that time, a three-inch anti-aircraft rocket. It was a very high performance rocket, which my father thought would have important applications for our forces, particularly if it could be equipped with a proximity fuse. The idea, I suppose, was originally anti-aircraft, but, as it developed, there were many other things. But we were moving very slowly in this country on the rocket development. This was partly because the group which mainly had this work under N.D.R.C. (they worked down at Indian Head ammunition depot) were committed to using a type of grain which was wet extruded. One made a mush of nitroglycerine and nitrocellulose. One made a mush of this in Acetone and extruded it into soft sticks, and then allowed it to dry. Well, this is, in the first place, a very slow process and in the second place a very hard process to control because the grains change their shape so much in the course of this evaporation; and, worst of all, they tend to develop cracks. And if you develop cracks in a rocket powder grain, you’ve had it. The flame goes down into the crack and starts to operate at very high pressure. You get pressure peaks, which then blow up the rocket. So it became clear that one needed more effort on rocket development, and this was the thing that Charlie started out here at the Institute with Willie Fowler. This is now late summer of 1941. They got started on some preliminary experiments and offered me a job out here, which I think I took up in October or November of 1941. The Institute work on rockets certainly would have started in July or so of 1941. There were some test firings in a small canyon not far from here, I imagine in August and September of 1941, and I joined the project in October or November. From then on — actually already from December of 1940 until the end of the war — I was an engineer.
What happened to nuclear physics research or all physics research at Caltech during the war period?
It was essentially all closed down as far as this group was concerned. I think there was rather little research of any kind going on on the campus. Even the formal teaching program was very much modified. We had here something called the V-12 program, where we put a lot of Navy people through a quick training course; so that both the research and the teaching activities here were very strongly perturbed by the war. In 1939, just after I left here, the Van de Graaff which we had built was moved out of the old high-voltage laboratory into this building [Kellogg]. At that time the medical research program was shut down, and the space was made available then for nuclear physics; and I believe that Bill McLean, who is now the head of research at Inyokern, who followed me in my job at the Bureau of Standards on the fuse work, was one of the last to do a thesis on that machine. I think this was in 1940. At any rate, when we came to start the rocket work in the late summer of ‘41, we pushed this machine off into a corner and cleared the whole area of the building for a drafting room and machine shop and offices and one thing and another of this kind. There had already been started at that time a project for a still bigger machine, and the tank was actually in hand. This was a machine which was to have gone to 5,000,000 volts and a considerably larger pressure vessel than the old one. That had been delivered and one had just started to put things in and that stopped. That tank was later taken out and used at the torpedo-launching facility which we built up in the course of the war for testing torpedoes. And we got it back after the war somewhat distorted. So I think to give you the broad brush picture, there was no physics going on in this area, at Caltech; and I think damned little at any other place in the country. You can look at The Physical Review for those periods. It’s not thin because of censorship. There just wasn’t anybody working in the field.
Where did the people from here scatter? Were most of them absorbed by the rocket project here or did a group of them go either to the radar laboratory of M.I.T. or to Los Alamos?
Many went to the radar laboratory. Victor Neher, for example, spent the whole war at the radar laboratory. The connections with Los Alamos were not established until quite late in the war, and it essentially amounted to a bloc transfer. Charlie and Robert Oppenheimer had some kind of agreement that Los Alamos wouldn’t raid us; that when we found a time when we as a whole project could be useful, that we’d move as a whole project; and this is, in fact, the way it worked out. Broadly speaking, we developed rockets here and propellant facilities from ‘41, ‘42, ‘43. As the thing grew, it became obvious that one needed to get professionals into it, and we picked out the site for Inyokern. The Navy put a lot of construction types and extraordinarily good officers into Inyokern, developed it as a test facility for rockets, and ultimately as a production facility for rocket propellants; so that by some late time here — I don’t remember whether it was late ‘44 or early ‘45 — a lot of this could be turned over to Inyokern and also to a production engineering organization, which Trevor Gardner had set up. This is the Gardner who later became Assistant Secretary for Air, who was very instrumental in getting the missile business on a sound footing. He was at that time a production engineer with Plumb Tool, and came to us with the idea of organizing our production problems. You see, for a small research group, we had production problems. We had at one time something over 200 small shops, small and big shops, building parts for us. I think we ultimately delivered more than half a million rockets to the Navy. Until the last year of the war, every rocket that was shot by the Navy was made in Pasadena through our organization. This had to be built up, obviously, and by late ‘44, I suspect it must have been, this was in pretty good hands, and we were able then to look at the Los Alamos thing and to be of what help we could, and it turned out that there were some blocs of work — again, of engineering production procurement character — that we could handle for them very conveniently.
You could handle from here, though, without moving.
Right. No, we kept our organization operating here; and, as a matter of fact, some parts of it remained in the rocket business, working primarily together with the Inyokern group, continuing new developments, and things of this sort. But by this time it was pretty obvious to us that the rocket development — any more rocket development that we did — was for the post-war period and it didn’t interest us. So my father then became liaison — he spent actually most of his time in Los Alamos during that period — and others of us would travel back and forth from time to time. But principally we had some procurement and some development jobs that had nothing to do with physics at any point, which I hope were useful.
Now, how did things resume again back here? How did you get back into peacetime nuclear physics research? Had you been making some plans for it when you realized the war was approaching an end?
I had not certainly. This engineering job that I found myself in was of such a character that it didn’t leave much time for rumination. My first thought when I did turn my attention to the post-war period was how to relearn all of the physics that I had lost. Without wishing to cry crocodile tears or to feel unduly sorry for myself, I will point out that this came at just the catastrophic moment in my career. I had had one year, a little over a year, as a research fellow-granted, in the best of all possible places — but then to be pulled out of physics for five years meant that I really had to go back and start from scratch. Everything I had learned in that time I had forgotten. I guess I never was a sufficiently good student so that the stuff really stuck very well anyway. So this was certainly my principal ambition. And in line with that ambition, I got a job teaching freshmen and gradually worked myself up through sophomores and juniors and seniors and got a second college education. I must say I’m working on the third now. With respect to the laboratory, I think you probably know a good deal of this story already. I’m sure you must have studied the origins of the O.N.R. and the situation that confronted the country and us in particular at the end of the war. As you know, it was obvious that the Institute lay in ruins as far as undertaking any physics research was concerned.
The Institute was beginning to gather its staff back again, but laboratory facilities — meters and what-have-you — one really was starting from scratch. Our Van de Graaff, for example, I said it had been pushed over into a corner. [Pause in recording] As an indication of the status of our equipment, our Van de Graaff had been used as a decompression chamber for divers on at least two occasions. And this involved essentially eviscerating the machine. Also, all of the other little things that one needs around a laboratory — meters, batteries, short lengths of wire — these simply don’t survive five years of the kind of occupation that we had imposed on ourselves. So there was a very serious problem about equipment. There was also a serious problem about money and a serious problem about jobs, where to get them. I’m not quite sure when O.N.R. first started, how early the money started coming from O.N.R., but it was quite prompt. That, plus kind of a useful arrangement that we had in connection with the contract, served to give us the financial boost that we needed. The arrangement was, I think, that all of the changes that had been made in order to accommodate the Navy project (I should have said this rocket project was paid for by the Navy, sponsored by the Navy, to the extent of several hundred million dollars), all of the changes that were made in the physical plant, all of the equipment and instruments that had been used, were replaced by the Navy or a suitable cash settlement was made. On top of that, equipment that could be declared surplus was made available to us so that things that we had bought for the project — lathes, for example; we’re still using lathes today that we got from that project — we were able to enter bids on and to buy for the Institute. So these were financial things that helped a great deal in getting started.
Now, you asked about the program that was laid out. The program that was laid out was to get that Van de Graaff going again. We knew as a result of the experience of the years just before the war that there were many things that could be improved on the Van de Graaff and we undertook to do that. We even had a little money at that time so that we were able to buy things instead of going down to the local power company and picking up broken insulators. We actually had some budget to spend. We even bought belts instead of making them. So in some respects, even the rather small amount of money that we had at that time by modern standards made it possible for us to move ahead rather quickly and to get the machine back into operation.
What about the plans that had been underway for building the larger machine?
Those were carried forward, although the design was now changed a great deal; and that machine was a little bit delayed because when we got the tank back it was rather badly distorted. It was designed to operate at 150 pounds per square inch, and they had operated it routinely at 300 for several years. This had resulted in the thing becoming slightly out of round, among other things. So it took a while to get that put back, but, again, the Navy paid for rehabilitating that tank and for re-installing it here in the laboratory. That machine was built at a much more relaxed pace than the older machine, which was built up and then put into operation rather quickly. I guess I don’t remember quite the sequence. Maybe if I look at the list of papers I can tell you a little more about what we started to do.
The first paper was in ‘47 with Christy, Cohen, Fowler…
Yes. Well, that was an experiment on the neutrino. The neutrino was invented quite a long time before the war and was a sort of a poor joke in many respects in that it had the property of being that particle which was put in in order to satisfy the conservation of momentum and of spin and such things, but nobody had ever detected one; and so one always said, “Well, the neutrino is a device which has been invented to conceal our ignorance.” This was a question in which we were quite interested. Fowler had worked on beta decay, and of course it’s in beta decay that the neutrino rears its ugly head most prominently. So a question was: "Does a neutrino also take away momentum?” From any theoretical point of view, one would say, “Yes, of course. It’s a particle which has a zero mass. It takes away a certain amount of energy. The amount of momentum that it takes away is 1/c times this energy.” But we thought, nonetheless, that it was something that it was worth doing an experiment on, and Fowler in particular had had a love for a particular kind of decay that occurs in Beryllium-8, where Beryllium-8 flies into two alpha particles and provides a very sensitive measure of transverse momentum. This was an experiment that could be done with the accelerator in not necessarily its best and final shape. It could be done without the need for an elaborate magnetic analysis. It could be done with a cloud chamber. The cloud chamber was something that we had some technique and background on here. So we set out to operate this experiment. As I say, it was a cloud chamber experiment. I guess it’s the last one that was done around here. It involved taking a very large number of pictures and measuring the angles of tracks and one thing and another. We proved to our satisfaction that the neutrino did in fact have the correct amount of momentum, that that feature of the neutrino’s properties was consistent with what one would expect.
You had some specific questions on this period, didn’t you, Barry?
Well, really on the cloud chamber. I was going to ask a question that perhaps you’ve answered in part — how useful was the cloud chamber at the end of the war for nuclear physics?
Well, you see, the development during the war, particularly at M.I.T. and at Los Alamos, had very much changed the whole picture of the relation between physics and electronics. Before the war, one knew a little about dc amplifiers. People had pet kinds of tubes — F6P9 - 54 I remember was one especially pet type — which had very low noise and were very sensitive electrometer tubes and could be used for measuring very small currents. Competition now you see with the quartz-fibre electroscope; competition with the Geiger counter, which by that time was at the point where almost anybody could make them. But dc amplifiers were very hard to handle. One was always having great trouble with noise problems and instability, oscillation — loop oscillation in the amplifier — and variation in gain. What one learned during the war was how to handle fast electronics, and this made a major difference in the whole attitude toward detecting systems. Instead of trying to measure accumulation of charge on a condenser or something of this sort, one measured pulses from a fast ionization chamber or a Geiger counter, and one used fast electronics. I suppose the most important invention there was the feedback loop which made such devices very precise, linear, and reproducible in gain, and the fact that one could run at such frequency bands that the ordinary sources of noise were not particularly important. A 60-cycle noise, for example: you used to build an amplifier in a sewer pipe in order to get rid of the 60-cycle pick-up. Now you build an amplifier that’s unresponsive to anything below 60 kilocycles or so. So this was an important contribution from the war. And one of the things that we did very assiduously was to go to Los Alamos and get copies of all of the drawings of amplifiers. For a long time, all the amplifiers that were used in physics in this country after the war were copied from the Higinbotham circuits that had been developed at Los Alamos.
Did, in turn, the knowledge for this come from the group that was at the radar laboratory at M.I.T.? Some of these people seemed to have first worked there and then gone to Los Alamos.
Yes. Well, I’m not an expert in this field by any means, but I know that Matt Sands, for example, was heavily involved in the putting into practice of one’s improved understanding of high-frequency electronics; high-pulse, short-pulse electronics. He and Elmore wrote the classic book on the subject, I guess just after the war, on the basis of their work at M.I.T. And Sands brought this also to Los Alamos. A number of people were involved in it. I remember the names of Sands and Higinbotham particularly, but there were a dozen other very sharp people working with fast electronics. At the same time there was a very important development on fast ionization chambers. In the days before the war, one used air in ionization chambers and collected the ions. Now, in air, which contains oxygen, the ions move very slowly. The electrons, which are formed in the process, in the absorption of energy via the gas, move very quickly; but they attach immediately to oxygen.
One learned at Los Alamos that one had to have very pure gases completely free of electro-negative substances. And this was a big step forward. This meant that one could operate ionization chambers instead of the millisecond range, in the microsecond range; three orders of magnitude change in the characteristic times of ion collection. The characteristic times that one thought about in terms of count rates, in terms of amplifier bandpass were drastically changed. So there was an important injection of technology from the Los Alamos project and from the radar laboratory also into physics at just this time. Now, it took us a while to get to it. We continued to work with electroscopes and we learned how to build Geiger counters and we started building a few amplifiers. But our major work I guess was still with counter techniques for some time thereafter. We started — I suppose it’s more reasonable to say Charlie and Willie started — just after the war to try to do something about the detection problem. I think I said earlier that there are two aspects to an experimental situation. One is the accelerator and the other is what you have available in the way of detecting equipment, and Charlie had already interested himself in the second phase in connection with the development of the quartz-fibre electrometer, Fowler with his work on the cloud chamber, and all of us together in making this electrostatic accelerator.
Well, recognizing the importance of these things, we started to work on such things again after the war. One of the developments was the electrostatic analyzer, which turned out to have a precision that was something like an order of magnitude better than the kinds of things that we and others had been using before the war. Another was the evolution of a double-focusing, magnetic spectrometer in which we followed some pioneering work that had been done in Sweden by Svartholm and K.MG. Siegbahn, I believe. One hadn’t really thought — or I, at any rate, hadn’t really thought, that one would come to the point where one would want to do more than just count how many particles come out of a reaction. Ultimately, to have so many that one could divide them up into momentum bands and sweep them around a magnet and focus them on a counter: these were hard things for me to understand and to conceive. But we did develop this spectrometer — largely my father and Fowler and a number of graduate students — and it turned out to be an enormously fruitful development. Just as the Van de Graaff improved the precision of accelerated beams by two orders of magnitude, so did the magnetic spectrometer improve the precision with which one could do experiments on the particles resulting from a reaction — improve the situation by a couple of orders of magnitude. I got interested myself in another type of magnetic spectrometer which had been developed at M.I.T., I guess immediately after the war. I think Martin Deutsch probably was the person who first published this particular version — a lens spectrometer which operated without iron, which had the virtue then of a highly reproducible field, the possibility of high precision…
You mentioned Martin Deutsch.
Yes, I was referring to the general interest here and of course simultaneously in many other laboratories in the world, who were getting back into nuclear physics, in making the step forward in the capabilities that one could command in terms of instrumentation. I mentioned the improvement in the accelerator, in the control of the accelerator, through the self-stabilizing arrangement, in the ability to measure the energy of the accelerated beam by means of either electrostatic or magnetic analysis, in the ability to measure the energies of the charged particles which came from reactions by means of focusing magnetic spectrometers. One problem which at that time was crying for attention was the matter of measuring gamma ray energies. I guess this was not a complete accident, since I had spent a lot of very frustrating nights in doing cloud chamber work on gamma rays with precisions of the order of five to ten per cent, which was simply not adequate to the purpose when one got into a little more detail. The development of the magnetic lens spectrometer — I mentioned the article by Deutsch in which he showed that this was a device with quite acceptably large solid angle, luminosity, and still a quite acceptably high resolution. He showed that this device could be made practically and could be used for beta decay studies and one thing and another of this sort.
It seemed an easy thing to duplicate, so we built one here, and tied it into our accelerator in such a way that we could produce targets immediately at the focus of the lens itself. In beta ray work, one customarily makes a beta source by bombarding at some place, and one carries it to the spectrometer, puts it into place, pumps down and starts making records. This is fine if the life is seconds or more, but if it’s a short life, then this is not practical. In particular, we hoped to be able to do gamma ray work this way by bombarding the target immediately at the focus of the lens and converting the gamma rays by means of heavy foils — again, distant only a millimeter or less — so that we could on line, so to speak, use this beta spectrometer as a spectrometer for the secondary electrons that result from gamma ray conversion. This enterprise turned out to be very rewarding, and we for several years then exploited this instrument for gamma ray measurements, and also undertook some beta ray problems that were not accessible for other people. In particular Lithium-8 and Boron-12 are two beta emitters which have lifetimes of less than a second in the one case and less than 30 milliseconds in the other, and energies up to 16 million volts, which were considerably beyond the capability of other beta spectrometers in the world. We had built ours deliberately to go to 16 million volts with a lot of copper and a lot of power. So we were able to exploit this spectrometer in the investigation of a number of beta spectra and in connection with a number of gamma spectra.
What was known about nuclear level structure at that point in time? What was the general knowledge that was available to you when you were forming your problems?
I’m glad you asked that question because it brings me to the next point. We had all of us to re-educate ourselves on nuclear physics. Obviously, what we all did was to break out the old Bethe, Bacher, and Livingston articles and start reading them again. But, more than that, or perhaps I should say equally important and directly connected with that, was the question of recalling, recovering what had already been done. After all, this field had been exploited by a lot of people with a lot of facilities from 1932 to 1940, and there was a lot of information available in the literature. So one of the things we decided we should do here was to go back to the literature and see what had been done. Up to ‘36 or ‘37, Bethe and Livingston had made a systematic survey of all of the nuclear physics work that had been done up to that point. And so some of us undertook to see what had been done since and to put it into some kind of systematic shape.
This led us quite soon to concentrate on the light nuclei, just on the ground that it was already a pretty big literature, and to evolve a kind of system of categorizing the various kinds of information according to their bearing on specific nuclei. One already knew that some reactions give you information about the compound nucleus which is formed in the process. Others give you information about the final nucleus. And we tried to unscramble this and essentially to put 20 boxes around in the room and to put in each box the information that related to that particular nucleus. I think I took perhaps the main responsibility for that together with a then graduate student, W. F. Hornyak and I wrote the first one of these together. Other students were involved in the literature survey. This must be 1948.
“Energy Levels of Light Nuclei.”
Yes. Yes, this was published. Well, the project was started, as I said, in order to get the laboratory some supply of information. We published it in a volume of Reviews of Modern Physics dedicated to R. A. Millikan. In that same issue Fowler and my father worked up a rather complete development of the development of the theory of gamma ray measurements, interpretation of Geiger counter and ionization chamber measurements, and also a good deal on what one could then say about the theory of nuclear reactions. So these two pieces of work were essentially our report on our home work in that period of ‘46 and ‘47. I’m perhaps not being quite responsive to your question because I think you want to know what theoretical developments were pressing us at that time. I think it’s fair to say that as far as I myself was concerned, I was more working on the basis1here are some experiments that have been done; they lead to such-and-such conclusions; is there another way of firming up those conclusions or of seeing that these apply to other places? I think my preoccupation then and now with keeping up with the experimental techniques was such that I did not have as deep a theoretical understanding of these problems as one should in order to do such a thing intelligently.
Where would you have been able to obtain it at the time? How could you have gotten it—from some particular article or book?
Well, there was certainly beginning already in ‘46 and ‘47 to be published a good deal of work in the literature, theoretical work. We had at that time quite a close connection with Robert Oppenheimer, although his interests were more in the high-energy field, in the meson problems; and, in particular, Christy joined our group at that time, and he was sort of our mainstay. Instead of learning physics, we would just ask Christy questions. This lasted for a long time, that he was the person who kept us informed about new theoretical developments, who asked us questions that led to experiments that turned out to be interesting, or to whom we went with questions that came from the experiments that turned out to be interesting theoretically. As one example of that… Well, I should say on this energy level summary business, because it has since been necessary for the laboratory to have this sort of thing, we’ve continued this; and I guess you have seen that that series has been rejuvenated after dying eight times or something like that. Every time we do it, we swear we won’t do it again, but we’re in the middle of it now. But that particular process — literature search — has been quite important to us in keeping up with what is going on in the world and what are new and interesting experiments.
Occasionally, when one puts this all together, one begins to see some light. One example of this is in connection with the mirror nuclei. I don’t know at what date people started to put together experimental evidence that not only were the ground stakes of mirror nuclei in agreement with the hypothesis of the equality of forces that I mentioned — this was something that had been shown by Fowler in ‘36 — but also that excited states agreed; and in the course of this survey work, I was able to accumulate a good deal of experimental information from the literature and start juxtaposing nuclei in such a way that one could see where the agreements were and one could begin to look at where there were obvious gaps in our knowledge. For a long time life was just such that one member of a mirror pair would be very well known and the other would be very hard to work on. But once one got interested in this, one found ways of dealing with the hard-to-work-on nuclei. So this was one place where the literature work served to motivate the experimental work both here and other places. Another in a similar connection was observation at a certain level in Nitrogen-l4, which was very well established in certain reactions and just didn’t appear at all in another reaction which had been studied over and over and over again. And this seemed to me very mysterious.
It happened to be the reaction to Carbon-12, alpha, leading to deuterons plus Nitrogen-14, in the first excited state. And I was in the business at that time of trying to inter-compare various reactions to see whether one really had good evidence for such and such a level or not. One saw it in all the reactions where it should appear and one had good evidence, but the missing reactions always bothered one. So I mentioned this to Christy — this I think must have been in about 1950 — and he looked at it and immediately saw that this was an example of the operation of an Isobaric spin selection rule, a rule which followed quite readily from the mathematics (now 15 years later, I can even do it myself) which if neutrons and protons were as alike as one thought — and this was already in a formalism generated in 1932, I guess; that one would expect these selection rules to appear, and this was an example of such a selection rule. Within the next several months, of course it broke out all over the country and it now is one of the facts that have sort of been put to bed in this business. It is a fact that these selection rules are obeyed, and now we’re more interested in knowing to what extent are they disobeyed — one per cent, one-tenth of a per cent, one hundredth of a per cent? So this is one of the ways in which the experimental information — and from our point of view it was important to include all the experimental information that we could lay hands on, whether generated here or elsewhere — served as a backbone for further experiments here and also elsewhere.
In 1948 or perhaps ‘49 the shell model, a very old model, was rejuvenated, and I wonder if you could describe the circumstances under which that happened.
And the relationship to the work you were doing here. That’s more to the point.
Well, first, more generally, of course this became known through the first paper of Maria Mayer and of Jensen, Suess and Haxel and I guess this is ‘49. It was certainly an exceedingly exciting development for us, obviously, because it put together a lot of things about heavy nuclei that had needed understanding. Now, in the light nuclei, the situation is not quite the same as it was for the heavier ones in that we’d always had a shell model. A shell model was in fact first proposed by, I believe, Bartlett (I think this is discussed in Bethe and Bacher’s article) at a very early stage; but it broke down because the numbers didn’t fit. And, as you perhaps know, the great triumph of the Mayer-Jensen point of view was in taking into account a large spin-orbit coupling which made the shell closures occur at different numbers. However, these closures still occur at the same numbers for light nuclei. In the light nuclei the spin-orbit coupling is not as fully developed as it is in the heavier nuclei. So people had all the time been talking about the light nuclei in terms of a shell model.
There had already in 1937 been a classic paper of Wigner and Feenberg, I believe, and later Phillips and Feenberg, in which all of the levels that were to be expected from a shell model description of the nuclei in which we were immediately interested were calculated in an extreme L-S coupling. I should maybe interject that in atomic physics one had the kind of coupling which is called L-S or Russell-Saunders coupling, which characterizes the way in which the angular momenta are coupled, and which worked quite well for light atoms. For heavier atoms one finds strong indications of another kind of coupling which is called spin-orbit or j-j coupling. And the same thing occurs apparently in nuclei in the p shell — and we’ve always called it the p shell — from helium through oxygen. One had these calculations of Wigner and Phillips and Feenberg which said explicitly in terms of only two parameters, I believe, where one ought to expect all of the excited states of the light nuclei.
However, even in ‘49 and ‘50 and ‘52, I think it fair to say that the experimental information on the light nuclei was sufficiently poor, there was enough bad information, that one couldn’t make a comparison with any confidence. I guess the shocker came when boron-10 was discovered to have a spin of 3. It was one of a number of things that occurred at about the same time. But it suddenly became obvious — the j-j coupling was rearing its ugly head in the light nuclei — that we’d have to learn about these matters. Dave Inglis wrote an article in 1953 in which he summarized the theoretical situation with respect to the shell model or the classic L-S coupling and the light nuclei. In this article he made comparisons between what one would calculate on the basis of this earlier paper that I mentioned — the 1937 paper — and what observations one had that one could rely upon. He concluded from this study that the simple L-S coupling was not adequate. He then made calculations in the other extreme of coupling, and by interpolation predicted where levels ought to be expected to be found in this group of nuclei of which I’m talking now, the p shell, with various degrees of coupling between. At the same time he put — I think Kurath was his student — Kurath to work in making these calculations on a more refined basis. So by ‘54 or ‘55 — Kurath’s paper was ‘56 — one already had quite good calculations (by this time they had to be machine calculations) of shell model predictions on where were the energy levels of the light nuclei. My own feeling during all of that time was a little embarrassed in that I felt I would like to go to people, like our friend Christy, or people like Inglis, and shake them and say, “Look, why don’t you make some predictions? You’re paid to make predictions. You’re paid to analyze these beautiful levels that we experimentalists discover and explain them to us.” But I was embarrassed about it, because I realized that a very large fraction of the information that we were giving these people was wrong; and it’s always a little embarrassing…
How did you realize that at that very time?
Partly from the rapidity with which it was changing, partly from the fact that in trying to make our assessment of the situation, we had very frequently to rely on experiments that we could see were poor in the absence of any other information at all. It’s like the maps in the old days. Somebody comes back having seen an island. Well, it turns out he’s only seen one side of it. What’s your best information? It’s an island. So you draw it as an island. There was an enormous amount of work which by modern standards would be absolutely unpublishable today. But it was all one had. And this work was done by our revered pioneers in the field. One should of course not go shaking sticks at them because they did poor work under conditions where poor work was the only thing one could do. But when one examined some of this work and stacked it up against what was available in other situations, one could see that a lot of it was clearly wrong, but it was the best guess. And I think in the course of the summaries that we have made of this, our confidence in the correctness of the information that we put on our diagrams and of our ability to assess new information — our confidence in these two things has increased a great deal.
I suppose it’s fair to say that the fact that these things agree more and more and more with Kurath’s calculations also gives us some hope. On the question of nuclear models, we’ve talked a little about the shell model, I certainly don’t want to leave the impression that it did not have a great impact in our understanding even of the light nuclei, and certainly the existence of the shell model, its usefulness in heavier elements, the fact that it worked in a large variety of situations, gave people like Inglis courage to make the kind of calculations that were needed in order to give us some hooks to hang our coats on, to give us some leads for experiments. And a good deal of our work (I would say a particularly central thought in our literature survey work) has been making inter-comparisons between what is in the literature in the experimental field and what the calculations of Kurath and later the British group — Elliot, Flowers, Tony Lane; a lot of people are in the act now — Bruce French I should also mention as having illuminated a great deal of this field. So the shell model — and particularly the question of the couplings that should be taken into account in this model — have been a central issue in the whole development.
At the same time, though, I should explain perhaps that our attitude in experimental work here has been of a slightly odd character from some points of view in that although in the literature survey work, our meat and drink is catalogues of levels, we do very little of that ourselves. We have, for example, not very suitable instrumentation for studying the whole energy level spectrum of a given nucleus — certainly a very important job, and if other people weren’t doing it excellently, we would feel very pressed to do it. But we have rather designed our detecting equipment and chosen our projects for more special motivations. The neutrino experiment is an example of this. Later on, the same understanding and taking advantage of the same properties of Beryllium-8, which we discussed earlier, led us to an experiment having to do with the kind of coupling that occurs in beta decay. This was after the thing with parity broke. And we took advantage… The thing I’m trying to say is that we took advantage of special kinds of nuclear reactions with special levels, special nuclear phenomena, to study special effects, which were in some sense broader than what we would call nuclear spectroscopy, the massing of information in a systematic way about levels.
Some of these special efforts have been in the field of trying to understand more about beta decay. Many of them have been in the astrophysical vein; and, as you know, a great deal of our special experiments have been motivated by astrophysical considerations, by the need for certain kinds of astrophysical data. One other thing which I should perhaps say about the model thing. I can say without any doubt that for the heavy nuclei, the next most dramatic event was the discovery of the collective model; and I think myself that this also will turn out to have been a dramatic event for the understanding of the light nuclei. This happened in 1952 when I was fortunate enough to be with my family in Denmark on a Fulbright appointment at the time when Bohr and Mottelson were developing their ideas on the rotational model. I was working with Torben Huus at that time on some obscure level in Lithium-6 when Huus suddenly made the connection between the things that Mottelson and Bohr were saying about Coulomb excitation and some muddy objectionable background that he had run into in an experiment that he had, incidentally, done here the previous year. But in a matter of weeks, literally, he was able to make this idea grow. It had to do with the bombardment of tantalum by protons, and he put a great deal of effort into studying just this odd background radiation that we had found so objectionable in the previous experiments. And this, then, was the thing that represented the first experimental verification of the Coulomb excitation process at Copenhagen and led to the further work that has been carried on so successfully there ever since.
When was it that you were there?
In 1952, 1953.
When they came out with it.
Yes. I had arranged the seminar in which it was announced. I was supposed to be giving some seminars, and I got tired of just being the only one to give seminars, so I asked Ben if he could talk about something, and he said, well, he could talk about this. But I think that that development of the understanding that there can well be other degrees of freedom in the nucleus that are more important than individual particle degrees of freedom, as I say, is known to have led to a complete change in our whole understanding of the heavy nuclei, and I think also, even in the s-d shell nuclei, where particularly Chalk River has shown that this model is very important there, we’re much interested here presently in trying to see to what extent this model also has utility in a system with 9 particles or 10 particles or 12 particles. From the pedagogical point of view — by which I mean from the point of view of making a poor, uneducated experimentalist understand what those guys are saying — it’s a godsend, just a beautiful little model r1ost as good as the billiard ball. You can make calculations with it without understanding anything, and you just have yourself a ball of fun. You make experiments, you make predictions, all by yourself; so it’s a beautiful thing in this respect.
I stress this because I think there’s a tendency in a lot of people thinking about this problem to say, “The interesting problems in nuclear physics are almost certainly now solved.” The success of the intermediate coupling shell model is so dramatic that one can safely say, “If we merely put in enough input information into a machine, the machine can calculate an answer to all of these questions.” From the point of view of the theorist, I think one can understand some revulsion to a field which has gotten to that point, and many theorists have left the field, I think precisely for that reason. They’re not interested in competing with computers. On the other hand, I think an experimentalist and a teacher take a slightly different point of view about the importance of model work or of approximations, to put it in another sense. I think that discovering new fundamental facts of nature, new basic principles, is obviously the most rewarding business that there can be. But discovering new ways of explaining what we already understand in a vague sort of way is, I think, also a very important enterprise. I think Niels Bohr probably was the man who in my mind most epitomized this point of view. Niels Bohr was in the forefront of a great many dramatic developments in physics.
He was certainly the first to understand in any deep way many of the new things — the quantum mechanics, fission, the liquid drop model, many many things about physics he had the insight to understand. But more than that, he devoted his whole life to making these things understandable. It was not enough to Bohr that he understood something. He also must make it understandable. He was led sometimes to almost ludicrous situations. I was very much a “yes boy” when Bohr had me under his wing. I think he distrusted me. He would try to convince me of something, and I would say, “Yes, I’m convinced,” and he would take me and shake me and say, “No, but are you really convinced?” This is wandering a little, but I think it’s an important point. One lives in nuclear physics to some extent in the hope that there is an ultimate simplicity about it. It may be too much to hope that it’s really simple, but maybe some facets of the problem can be understood in a simple way. The shell model was really the first fruition of that hope. When this shell model went under the waves the first time, it was because it was too simple and because it just didn’t fit enough facts to be interesting. But when it rejuvenated in 1950, it fitted so many facts that one had to take it seriously. Yet in some ways it’s ludicrous to think of the nucleus as being described in terms of central forces. You see what this means: this means that the particles which are acting on one another can be represented as being acted on mainly by a central particle, a central source. In the atom you have such a force. In the atom the electrons are all tied to this post in the middle, which is the nucleus, and their interaction with that is strong compared to their interactions with one another.
Take the post out and you have a nucleus. But the fact that you can treat this nucleus with the same mathematics, the same point of view, as if the post were there, has been one of the most baffling facts in the whole history of the subject; and Victor Weisskopf, bless his name for making physics available to experimentalists, has explained this in a very neat, plausible, satisfying way. It’s not a complete way, and it’s not enough yet so that one knows where the theory is going to break down. But think this is where the fun lies — in the hope that by refining the theory, by improving our intuition — obviously, the mind of man has to go part way in this: one can’t say, “I refuse to understand physics that I can’t understand on the basis of Newtonian mechanics.” One has to educate one’s intuition. One has to be willing to accept the uncertainty principle and fuzzy electrons and relativistic transformations and things of this sort. But as one gets used to these things, one calls those “intuitive concepts.” And as the theory gets simplified a little bit, then one has a gradual merging between human intuition and theoretical physics. And I think that is the point that we’re striving for. It’s not enough to me that a machine can make a nucleus.
Well said. You answered a number of concluding questions. I don’t think we can say anything beyond this point really unless there’s something very specifically.
Well, I hate to waste your tape. I have a tape-recorder of my own, and I know what this stuff costs.
That really summed things up historically and philosophically.
Let me ask one thing. You have a lecture that you want to get to. You must leave at a quarter to five.
it starts at a quarter of five.
Yes, I must leave at 20 of or I won’t get in at all.
Well, would there be any virtue in discussing the problems of detection in gamma rays? There was one point, for example, at which you mentioned that detection depended upon secondary electrons. There were several other methods. At the point that you started doing this, it must have been a very tricky business; and it seemed to me that was a fairly important phase of the experimental work. You’ve got about 15 minutes…
Really only ten minutes.
Well, I certainly would not say this was a technique, this conversion of electrons. It was invented here. In fact, Ellis used quite similar techniques in the very early days of the study of radioactivity. The point that I was trying to make was that we were very much taken up here in that period with the evolution within the limits of our ability of detecting equipment. We felt very strongly that it was not enough to have a high-precision accelerator, that one needed these other devices. And each of these involved some problems. None involved any fundamental new discoveries. The idea of the conversion of gamma rays in heavy foils was very old. There was no problem. The idea of using a magnet to bend particles was very old. The idea of building a quartz-fibre magnetometer that could be built by a graduate student and would have a precision of one part in l03 did not require new principles, but it required a point of view. “Here’s something that needs to get done. Let’s do it.” On the gamma ray detection business, I think you know that not very long after we were involved in this, using this technique, other techniques were developed. The magnetic lens spectrometer is still used for this purpose. Ours is no more because it burned up, but there is, for example, a very fine spectrometer in Brookhaven that does this kind of work and it does even very sophisticated work in measuring multi-polarities. But the dramatic event in gamma ray detection was the invention of the sodium iodide-photomultiplier combination which I think Ruby Sherr had some part in and McIntyre, presently of Texas, had some part in; and Hofstadter. Yes, it was Hofstadter and McIntyre, I believe, who did the first work, but Ruby Sherr was also in that field. When one could buy sodium iodide crystals one could do tremendous things with gamma rays. With this magnetic lens spectrometer, we were working at efficiencies like 10-5. You can now work with efficiencies like one. So the level at which you can detect and measure gamma rays has changed. I mentioned my thesis work was on gamma rays from fluorine plus protons. That gamma-ray spectrum, which I announced was a single line, after eight months work, turned out later to have three different lines in it, and it’s ordinarily used now in a 20-minute run to check out a counter.
What other characteristics can you identify relating to the change in style of research in this field over the period from the time you did your thesis work to the developments of the last ten years?
I would think from an experimentalist’s point of view that — again, the broad brush characterization — 15 years ago, 20 years ago, 30 years ago, what one did in nuclear physics was very much determined by what instrumentation one had available. One built a cloud chamber and one ran it. Whatever experiment could be done with a cloud chamber was put on the list. We made, in fact, a considerable point of just this when we decided to ask for funds for a tandem Van de Graaff here. We had by this time — this is ‘36, I guess — built three machines for about one, two and three million volts respectively. And although we were very happy with these, we were continually being frustrated. You see, by this time we were beginning to understand a little of the theories. By this time we were beginning to see what kind of questions Kurath would like to have answered, or Elliot, or Lane. “What are they talking about? What measurements can one make that bear on the theory?” Well, out of the set of measurements that bear on the theory, we had to choose the sub-set that were intersected by our facilities. So we did one kind of experiment or two or three different kinds of experiments, always limited to three million volts bombarding energy and always limited to the kinds of detection techniques that we had in the house. For example, we never got to be very good at neutrons in those days. With the design of the new facility, which went to 10 million volts, which was enormously more flexible in terms of what kind of ions you use, we moved into a different phase in a sense — not black and not white. But more and more we were able after the installation of this machine, to choose experiments on the basis of their interest and to put together the equipment to do the experiment. We now have the ability to go to 18 million volts to particle energies some kinds. We can bombard this or that with any ion up to carbon and oxygen. We have enormous variety, enormous flexibility in our accelerating equipment. At the same time with new developments — now passed sodium iodide for the lithium drifted germanium detector for gamma rays, fast counters for particles which you buy off the shelf, magnetic spectrometers — we have a rather full arsenal for treating the kinds of problems that we’re interested in. So our attitude nowadays is more: what would it be nice to know, rather than what can we find out?
And you indicated just before this what you would like to know, what would be nice to know.
Yes. Well, we have embarked on a number of programs of this kind which have sort of broad outlines. It would be nice to know a lot about A=6. When we started the machine, we decided that six particles was a nice number, and we set up several different groups of people working on various problems. Well, after a couple of years we learned a hell of a lot about A=6, but in the course of that we ran into all kinds of other interesting things, so that that simple beginning has mushroomed out into an enormous forest of experiments. So serendipity still has its place and complete chance will have its place. I would hate to get into the position where one used a laboratory like this only for measurements and never did experiments in the sense that one did things in which one was looking for a definite result. These I call measurements. I think an experiment is something in which one runs into something maybe at midnight, some effect one doesn’t understand, and one follows that lead down. Most of the time it leads to a broken wire someplace, but once in a while it becomes an experiment; it becomes a new effect.
It’s time to quit, but let me cheat by telling you the type of one or two bits of information I’d like, and you can decide whether we can answer that within a minute. One of them is: what other groups were doing the same work that you’ve described that Caltech was doing in the post-war period and now? And another one is, the source of research support. You mentioned ONR’s role in rebuilding the laboratory.
ONR has supported us ever since with budgets that started at around $100,000 and are currently around $1,000,000. There are in this country at least a dozen front-rank laboratories doing precisely the same kind of work that we are. I think of course first of all of Wisconsin. Wisconsin has led in many new developments of instruments and also have been responsible for many new very important steps forward in understanding the nuclei. Rice Institute has also been a friendly competitor from way back; and, as I say, a dozen others I won’t try to tick off because I’ll forget the most important one.
Well, some time we’ll try to put together a preliminary list and ask for your comments. What about the final thing, about attracting people to the field? Has the new post-war generation of physicists replenished the ranks in this field? Is it growing and attracting many of them?
Well, the best thing I can do there is to refer you to the statistics. I don’t think that a personal, local impression would be very useful. Over the past 10 or 15 years, about a quarter of the Ph.D.’s in physics in this country have been in low-energy nuclear physics. In the field in which we have been conducting these surveys, the literature has grown exponentially, I suppose, with a half life of, like, ten years. Our current bibliography contains, I would guess, nearly 10,000 accessions since 1959.
So these symptoms would support these other statements you made about the things yet to do in this field, the motivation of the people in it, and the ability now to conduct the type of work that one would like to do.
Yes. Well, I think many people would also place a great deal of emphasis on the usefulness of the field in teaching. I think it’s an excellent place to train graduate students. A lot of our students have gone out into quite different fields and have done very well. I think the kind of training that they get in nuclear physics is a very broad one and a very useful one; but I don’t think that that’s the real excuse for being in the business or for encouraging students to go into the business. If it isn’t fun, then it isn’t worth doing.
T. Lauritsen, "Depth Dose Measurements with a 1000KV Roentgen Tube," Am. J. Roentgenology and Radium Therapy, 41, 1003-1006 (1939).
Later: This was a lapse of memory: I now recall quite distinctly a good deal of excitement here when the first reports of 100-200 MeV ionizing events were published. The idea of secondary neutrons and chain reactions, however, had not come under discussion at that time.
W. F. Hornyak and T. Lauritsen, Rev. Mod. Phys. 20, 91 (1948).
M. G. Mayer, Phys. Rev. 74, 235 (1948).
O. Haxel, J. H. D. Jensen, and H. E. Seuss, Phys. Rev. 75, 1766 (1949).