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 Charles Slichter by Lillian Hoddeson on 1977 April 29,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
For multiple citations, "AIP" is the preferred abbreviation for the location.
Topics discussed include: family background; Slichter's time as a student at Harvard University; John Van Vleck; his time at the University of Illinois; solid state physics; Fred Seitz; discovery of superconductivity; Albert Overhauser; John Bardeen; Ginsberg-Landau theory; Bardeen-Cooper-Schreiffer (BCS) theory.
The plan for today is to discuss the discovery of superconductivity but before we get immersed in all that, I would like to ask you a few questions about your early background. You were born in Ithaca, N.Y. in 1924.
I also read that you grew up in Cambridge?
Can you please tell me —?
Yes. My dad was a professor of economics at Cornell. We were there, except for two years in Washington, D.C., when he was working at the Brookings Institute, and then he was hired by the Harvard Business School, and we moved to Cambridge in 1930. Later on, he became a professor in the economics department, and then was one of the first University Professors, so called, at Harvard. So I grew up in Cambridge, really.
I see. Your mother, what did she do?
Well, she was — as they say — a housewife. But she was active in a variety of different things. She taught school for a while, taught Latin and English, on graduation from college. Both my parents went to the University of Wisconsin and came from Madison. In fact, we had a summer home there which I have inherited, and we used to spend our summers in Wisconsin, three months every summer. I’m partly Midwestern.
I don’t see any scientific background.
Well, my mother’s father was a civil engineer, and my father’s father was a mathematician, whose field was really applied math. He was head of the math department at the University of Wisconsin. In fact, he brought the mathematician Van Vleck, who is the father of J.H. Van Vleck you know, the famous physics Van Vleck, to the University of Wisconsin, recruited him. My grandfather was later Dean of the graduate school at Wisconsin, and one of the people who helped found the famous Wisconsin Alumni Research Foundation, you know, from which they get a whole batch of research support these days.
And I have an uncle — my father was the oldest of four, and one of his brothers is a geophysicist. In fact, he was head of the Institute of Geophysics of UCLA.
Would he visit the family occasionally?
Yes. They lived in Cambridge. He was a professor at MIT for — in the thirties — and they lived about three blocks from where we lived, so I knew them. My parents always, I think, more or less assumed I was going to go into science, because math was always my best subject. And they were great ones for —
— you’re saying now, in high school math, or junior high school math?
Oh, from earliest age, math my best subject. You know, I suppose, arithmetic, you’d call it, at an early age. The fact that both grandfathers were in technical fields meant that they always really encouraged me a great deal. My brother’s a chemist.
You have one brother?
Yes, one brother. He’s actually — well, he’s really a chemical physicist.
Younger or older?
He’s two years older. He’s executive director of, I guess it’s called Executive Director of Chemistry and Material Sciences at Bell laboratories.
Oh, of course.
He has the job that Bruce Hanney used to have before Bruce became the Vice President for Research.
Yes, of course. I’ve seen his name and I’ve often wondered if he was related to you. Sure.
He’s two years older than I am.
Let’s see, junior high school, I suppose you went to in Cambridge and elementary school in Washington and Ithaca?
Well, I started first grade in Washington, and was in it for a few months, and then was taken out of school because I couldn’t see the blackboard. In the summer, I got glasses, out in our summer home in Wisconsin. I re-entered the first grade to try again when we moved to Cambridge. I went to a little private school called the Buckingham School up through the fifth grade, and then went over to a little school called Brown and Nichols, where I stayed until I graduated.
Graduated high school?
Yes, a small school, something like 30 people in the senior class.
I see. You must have some special recollections of events that were particularly significant in that small school that helped to direct you.
Well, I guess a couple of things. One is that it was — they had good science teachers. They had good teachers in general. It was a very good school, and the other students were quite good. But I guess the thing which I liked about it best of all was that being a small school, as it was, — and compulsory athletics, you know, you didn’t have, sort of a division between the students who were athletes, and the students who weren’t. And everyone — all the teams were made up of all of us. You know, you had to have that, to have a football team and a baseball team and things of this type. And I guess that, in many ways, strikes me as the most important aspect of that school to me, because it got me out of just being you know, a bookworm. I can imagine that had I gone to a great big high school, you know, I wouldn’t have played any sports. That has a big effect on a boy, to have a chance to do something other than just study.
Did they have a good laboratory?
No, they didn’t, it was a reasonably poorly equipped school, sort of run-down thing, mostly faculty children. But they — the teachers were good, and they had high standards. So I got a really good education. I had to write a lot, and we had a theme a week, you know, starting in the 8th grade, and the themes were graded — not like schools today, where they never grade any papers. But you’d write a paper and the thing would be graded and returned to you, — just a very good old fashioned education.
You probably developed rather close relationships to the teachers? There were so few of them.
Yes, although, of course, you know, you never saw them outside at school.
The teachers were — well, except for the fact that, you know, the football coach was the English teacher, and the baseball coach was the French teacher, you know that sort of thing. So you’d see them that way. But of course, when I hit Harvard, the teaching was nowhere near as good as it was at Brown and Nichols.
Especially the contrast with the English teacher was the most dramatic. I had a graduate assistant, something like that; maybe he was an instructor, teaching elementary English in college. I don’t think the language instruction was as good, by and large. We had a really rigorous education in Brown and Nichols, and the teachers were just absolutely on top of everything.
What brought you to Harvard? Was it clear all along that you were going to go there?
No. The big decision was, Harvard versus Wisconsin. I didn’t have an interest in any place other than Harvard or Wisconsin. Actually, the thing that decided me was my French teacher who said to me, “Well, you know, you can go to Harvard or Wisconsin, but if you go to Wisconsin you’ll spend the rest of your life explaining to people that you could have gone to Harvard.” I reflected on that and I realized, by God, he’s right. Why go to Wisconsin? Just because I had this family connection, you know, sort of sentimental feeling about it. And I decided that really wasn’t a good way to decide. And I certainly don’t regret it. Harvard’s a fabulous place. I was the right sort of person to go there.
What do you mean?
Well, I wasn’t scared of the place. I was used to being independent. Harvard’s a good place for students who are quite independent, and don’t need help from a teacher. You know,+ if you’re a self-starter. If you need someone to monitor you and sort of tell you when you’re getting in trouble or something of that kind, it’s not a good place. But it’s a very good place for students who can make their own way. Another thing about it which was terrific is that there wasn’t any one student group that was sort of dominant. You know, lots of places, there’s maybe a fraternity crowd or something like this, which sort of sets all the tone. At Harvard, there was a whole bunch of different sorts of people. There were the club men, you know, and the Midwest high school kids and the athletes — I don’t know, all different bunches, and you could easily, you could always find some group with whom you’re congenial, no matter whom you were. I think that’s really great, because it meant that you were really in a much richer environment.
You were there during the war?
Well, yes. I was there till 1943. I entered, see, 1941, and my freshman year, we had Pearl Harbor. And then Harvard had a two year elementary physics course, and I was in the first year of that. Well with the war coming along, they decided that they wanted to speed up the education, so in the middle of the year, they switched from a two year to a one year, and so we had a hurry-up treatment of certain topics in the spring semester, which were jammed in. Then I started going to summer school. I went to the summer school and took some more physics courses.
At Harvard, yes. I took second year calculus, and I took the E and M course which followed the elementary —
— who were some of your teachers?
Well, let’s see. In freshman physics Wendell Furry, as the lecturer. And then my section man was Critchfield, you know, who’s a well-known physicist. Peter Weiss was one of the section men in physics, although I didn’t happen to have him. You knew him at Rutgers, of course. And then, I’m trying to think who taught the E and M. I don’t remember who taught the E and M course in that summer school, but to show you what the situation was, all the graduate students had left, so when the fall came, why, I was hired as a grader of lab reports, in the E and M course which I had just taken, and Van Vleck was teaching that, so this was my first contact with him — although my family had known him. He’s from Wisconsin, and he lived down the street from where my father lived. He was somewhat younger, but my patents had known the Van Vlecks for years.
Did you get to know him quite well?
Yes, I got to know him very well. He is just a terribly nice man, and has always taken a very deep personal interest in me and I feel a very special tie to him.
Did he talk about physics with you in that period?
Yes. Yes. Well, at that point I was really just his grader, you know. With the war on, I wanted to leave, and I told him that I was going to enlist in the Signal Corps. I thought that this made sense, because this is what you know, they had all this radar going on, and the students were going to radar school and stuff like this, and I thought that’s probably what I ought to do. And he told me — I told him this, I guess, in January of 1943, so this is my second year of college that I wanted to do that. I was taking courses, the electronics courses that Chaffey taught vacuum tubes and circuit courses, things of this sort. And he said, “Well, there’s this chemist, E. Bright Wilson, Jr. who has a war project down on Cape Cod, and they‘re looking for some electronics people.” And he wanted to know if I would — he said I really ought to go talk with him, and see if I could get a job. So through Van I made an appointment to see Wilson, and went over to talk to him in Gibbs’s Lab, which had classified work going on and it was all sealed off and you had to get special — you know, identify yourself when you entered and all this. And I talked to them, and heard nothing from him. So as we were getting well into the spring, Van Vleck asked me again what had happened. I said, well, I’d talked to Wilson and I’d never heard from him again, so I assumed he probably figured I was — you know, I’d had no practical experience, I’d just had these courses, and what was I? I guess I was 19 years old. So I figured that he probably thought he could do much better than that. But about a day or so later I got a call from him. Obviously what happened is that Van called him up. And they hired me. So I went down to Woods Hole and worked in the lab doing electronics there during the war. I was there till January of 1946, and I did circuit design, circuit maintenance and circuit repair. We were measuring pressure time curves, things like that, for underwater explosives.
You wrote your first paper then on migration of gas bubbles.
That’s right. Yes, right. I was involved with running the electronics, with which we did these experiments.
A terrific experience for an undergraduate.
Well, it was. One thing that came along which was fabulous — we were going to make a special trip; we had an arrangement to work with the US Navy and with a Canadian ship, which had certain equipment on board, and we wanted to make measurements of the gas bubb1e and the pressure pulse from it. And we needed a very deep part of the ocean, where you wouldn’t get bottom reflections, which would louse things up. So it was decided that we’d go down to the Bahamas, to a place called the [???] of the Ocean, and there was this sailboat 160 foot catch that the Oceanographic Institute had called THE ATLANTIS and we were going to outfit that, and I was put in charge of the electronics. I was supposed to be able to maintain it and do any repair stuff that was needed. Well, we decided that we needed four oscilloscopes, which would have very good frequency response from low frequencies on up, and so we designed from scratch and built these oscilloscopes. Of course, these days you wouldn’t do that, you’d go buy it, but the commercially available things weren’t good enough for what we needed, and so we built these things, and we built some spares. Well, it was a fantastic experience, because here you had an objective, you know, a practical objective, and a deadline. We had to get the design de-bugged and everything put together within a matter of several months and then take it down to the Bahamas. And we were down there for maybe a month or six weeks, something like that, running these experiments. It was really quite exciting.
So the war had a tremendous impact on your career —
Oh, it was fabulous — oh yes, it did. And when I got back, I still hadn’t graduated, of course. I got some credit towards graduation for the time I spent down there. I came back at midyears. One of the most important things that happened to me, frankly, was that I went to talk to Van about coming back and what I should take, and he said, “Well, you can take this mechanics course I’m teaching.” “But,” he said, “You will have to work up the first half on your own.” And so he told me what the text was, and I got it, and he had a bunch of mimeographed —
— what was the text?
Synge and Griffith. Marvelous book, a really good book.
Yes, I know it.
And I really learned how to study out of that. He gave me a set of what he called review questions, which he handed out to his class, and he said, “Work through the problems, go through these chapters.” So I spent a couple of months working nights, studying that book and working through it. And the thing that was so neat, was that it was full of these problems, “Show that this is so or what have you,” so you really knew whether you got it right or not. But you’re absolutely all on your own there, you know and if you can’t get the thing, you have to go back and figure out what it is you’re missing. It just absolutely changed my study habits. When I got back, my grades just changed completely. I mean, I’d been a so-so student, half A’s and half B’s, something like that, and after I got back, I guess the majority of exams I wrote, you know, hour exams and final exams, were in essence perfect papers. You know just all out of having really learned. Before I went down there, I’d go into an exam and they would ask me a question on the exam, and gee whiz, I never thought of that. It was full of surprises. And that just didn’t happen to me afterwards. Of course, I was older. But I think the real thing was that I’d learned so much how to study, just by having to be on my own. And of course, I’d learned working in that lab, I’d learned so much about how to learn something on my own. You know, about circuit design, because we were doing things which I hadn’t had in courses. We got books.
There are certainly lessons in there for people who are teaching physics.
It’s really good to have some work experience, you know if it is really pertinent. This was so pertinent. The other thing is, you know, I was really fired up, to get back there, because the people I was around could do things I couldn’t do. You have to understand the people who were there. Well, Don Horning was there, for example. Another person who was there was Bob Cole, who — he’s a physicist, but he’s in the chemistry department at Brown. He won the Langmuir Prize of the Physical Society a couple of years ago, for his work on dielectric relaxation. Just a fabulous physicist.
And he was at —
Woods Hole. In fact, my immediate boss was a guy named Dave Stacy, and Dave’s boss was Bob Cole. Bob Cole is one of these — he’s the only person to do an experimental thesis with Van Vleck, but he’s a fabulous theorist, and he’s one of these people whose able to take very complex theoretical treatments such as Kirkwood gave of some of the explosive phenomena, and he could start at the beginning, and at the beginning make the simplifying assumptions that Kirkwood eventually would make, you know, by the end of the thing, and then just in the simplest of terms, derive the results. He’s Just like Purcell, in this regard, you know, this ability to pick the central thing which makes the whole problem transparent. You abstract the essence of it, and carry through a solution which just is so beautifully clear, because it has no excess garbage in it; nothing but the essence and very simple, just one of the clearest people. In fact, he told me when I went back that I ought to try to do a thesis with Purcell.
I would like to ask you what went into your decision to stay at Harvard.
Well, I had firmly determined to leave. I mean, I assumed I had to, because it was my thought that you always went to a different place for graduate school. But Van Vleck was my advisor, and as I say, I was taking his mechanics course from him, and he would — as I say, “Well, why don’t you come around after class?” He’d say, “Have you decided where you’re going to go yet?” and I would say, “Well, no, I haven’t,” and he’d say, “Why don’t you come around after class and we’ll talk about it.” Then I’d come around to his office and he’d say, “Oh, what did you want to see me about?” The perfect absent minded professor. Then we would talk about it, and then he always ended up the conversation saying, “Well now, you don’t want to overlook Harvard. After all, you’re not like —” Then I would give him these arguments, why I ought to go some other place, and he’d say, “Yeah, but look — you’ve been off for several years during the war. You’ve been at other places. Your family’s from Wisconsin. You know, you really have been around a lot more than most people who confront this.” And so I just finally ended up deciding I would stay there. I wrote to Princeton. That was sort of my first choice. And I got back letters from Princeton saying Princeton’s terribly hard to get into. I wrote them asking for application blanks, and they wrote me back saying Princeton’s terribly hard to get into. They didn’t even send me the forms. So I just — I wasn’t about to pursue that. I wrote to Illinois, and I got admitted here and I got admitted at Wisconsin and Cornell. Those places were eager to take people who had had some war experience, but not Princeton. Anyway, in my first year of graduate school, why, Van Vleck said to me, I guess in the middle of the year — I was taking a course from him in mathematical physics. He said, “Have you decided whom to do your thesis with yet?” I said, “Well, no. I haven’t.” He said, “What you really ought to do for your thesis is to work with Purcell, and look at paramagnetic resonance, and measure some of these paramagnetic atoms.” Well, I didn’t know what on earth it was. I didn’t know anything about it. I didn’t know what the problem was, you know. But I dutifully followed his advice, because I realized that he really knew the place and had my interests at heart. It was very clear; he didn’t think I should be a theorist. He thought I should be an experimenter. At that time, I wasn’t really sure what I wanted to do, and debated trying to be a theorist. He said, “No, you don’t want to be a theorist. A person has to be awfully good to be a theorist, and you’ve got all this competition from all these Europeans, and stuff like that. The thing to do is be an experimenter.” So, I went in to Purcell. I remember I was so nervous my voice was just quaking. I don’t know why, that’s how it was. Anyway, he said, “Yes.” You know, it’s so different from the way it is here now, where you go around and interview a variety of faculty. He just said “yes” right off the bat. I don’t know, maybe Van Vleck had talked to him. I can imagine he had. I was the only person working with Purcell who wasn’t working on nuclear resonance. It was so early in the game, the Radiation Laboratory series hadn’t appeared yet, and there weren’t many books around on microwaves. So I got help from him and George Pake and Bob Pound to design a microwave spectrometer — you know, design a cavity for it, and I put this thing together, and I started getting results. I remember —
— what were the relations to the MIT Rad Lab, if any?
None at all?
None at all. Harvard had had this Radio Research laboratory which I believe was a countermeasures lab, and Van Vleck had been associated with it, and they had a lot of microwave gear around, and I scrounged microwave gear from them. Purcell and Pound had been at MIT at the Radiation lab. Pound was of course a big expert on noise and on frequency stabilization, and I built a little Pound frequency stabilizer, to stabilize the frequency, and it was so early in the game that when I started out, people were not operating their bridges so that you could pick out the absorption separately from the dispersion — people who had been working prior to me, in this field, had a microwave bridge, with a cavity on one arm and a matched load on the other, and they would balance the bridge perfectly. And then the output is proportional to the square root of (ki?) prime squared plus (ki?) double prime squared. So you mixed the absorption, ki double prime dispersion, you mixed them. In nuclear resonance, people operated bridges unbalanced, and when a bridge is unbalanced, why, then they could see, you could pick up the absorption or the dispersion, depending upon whether you went off balance with frequency or with amplitude. And I remember how excited I was — and it’s not a very big discovery, but I realized, by God, you know, I can do this differently. I don’t have to do what those other fellows are doing. All I have to do is run off balance in amplitude, you know, and be tuned to frequency, and I’ll be able to pick out the absorption. That was very early in the game. So that was —
— and in fact, you stayed in this general area. It became your main theme —
— yes, that’s right —
— of the research —
— that I’ve been involved with since. Except when I came out here — yes, it’s the main thing. Resonance had just broken, you know, 1946, and when — I entered graduate school in 46, and I got my PhD in ’49 — so you know, it was just breaking. People were anxious to find people in resonance, and Wheeler Loomis came to Harvard, recruiting, and I remember, Purcell stopped me in the lab one day and said, “Has Loomis been here yet?” And I said, “No, who’s Wheeler Loomis?” He said, “Well, he’s the head of the department at Illinois, and he wants to talk to you about coming to Illinois.” So I made contact with him later in the day, and they invited me out, and I came out here, and he offered me a job. And I had heard the rumor that Seitz was going to come here, and I thought that solid state physics would be a great area. I didn’t know anything about it, so —
— how did you get that idea? This was very early in the game.
Because George Pake, who was just ahead of me, for his thesis in nuclear resonance had seen come interesting effects — he had observed that protons in water molecules, water of hydration in solids, that the proton-proton interaction gave a structure to the nuclear magnetic resonance line, and you could determine the orientation of the water molecules in the solid. And then he and Herb (Gatofsky?) working with his rig, had examined things like ammonium chloride and some other solids, in which internal motion takes place at certain temperatures. The motion is frozen out at low temperatures, and as you come up in temperature, the motion begins. And they could see these motional transitions in solids. So I realized that magnetic resonance — and of course, I knew from the paramagnetic relaxation stuff that I was involved with, looking at iron, iron alum and manganese and manganese sulfate, thing like this, that the crystal fields affected the resonance. So I realized that things in solids revealed themselves in resonance, and I just decided that much of the most interesting area of resonance was not measuring nuclear moments or things like that, but was to apply it as a technique for studying problems in solids. And so I just decided, you know, that was the thing to do. And Fred Seitz was Mr. Solid State physics. There was his book, MODERN THEORY OF SOLIDS, which was the Bible that everyone used.
Had you used it at Harvard?
I audited a course in solid state physics that Van Vleck gave, but I didn’t audit it very faithfully, because I was trying to do a thesis. So mainly, I knew about the book. I didn’t know any solid state physics. But I figured, this would be a good place to learn if he was coming out here. I’m surprised at my gall, when I think back on it — when Loomis offered me this job, I said, “Well, I heard that Seitz is coming. Is this correct?” He said, “Well, we’ve made Seitz and Maurer an offer, and I don’t know whether or not they’ll accept.” I said, “Well, can I wait to give you my answer until they give theirs?” And he said, “Well, yes.” And I realize in thinking about it, I said just the right things. You know, the point was that it was clear I was focusing on the essential thing, which is what kind of scientific colleagues was I going to have? And if those guys were going to come here, you know, and then it was clear that this was where I ought to be. The thing that was interesting was that Seitz had been asked to organize a solid state group. He hired, he brought —
This was Loomis who asked you?
Wheeler Loomis, who was the department head and he brought Seitz and Maurer here, and they brought with them Dave Lazarus from Chicago, where he’d just finished graduate school, and Dylan [???] who’d been a student of Bob Maurer’s at Carnegie, and a —
The five of you came in the same year?
Yeah. But I was not brought here as part of solid state activity by Fred Seitz. I was brought here by Loomis. I came because of Seitz. But Seitz brought the others. They were his pick, whereas I wasn’t.
What about Maurer? Did he come with Seitz?
Oh yes. He was with Seitz at Carnegie Tech, and Seitz brought him to be a senior person setting up the experimental solid state physics. [???] was a brand new PhD, and he was going to go into low temperature physics and get the place started in that and Lazarus had been a student at the University of Chicago, with Andy Lawson, and was a very bright guy who was well regarded as being a very promising young solid state physicist. So they were brought here, and — the first thing I wanted to do when I came here was to look for the electron spine resonance of F centers. And I mean, my knowledge of solid state physics just about took me to — I wanted to work on what I considered to be a real solid state problem. I didn’t think paramagnetic ions and paramagnetic salts was a real solid state problem, and to me that was an interesting sort of problem, that you wanted to get down to low temperatures and things like that, but I didn’t feel that that was in the center of solid state physics, whereas I felt that defects were. You know, I still believe that. As real solid state physics. Now, I learned — and I decided what I’d make was a microwave superhet and I was going to try to buy certain parts of it and make other parts, and was sort of assembling lists and things like this, when I talked to Bob Pound, and I learned that Berringer at Yale, who had a very sensitive microwave rig, had looked unsuccessfully for F centers. So then I said, “I don’t know why he didn’t see them, but of course there are lots of reasons you wouldn’t see a resonance.” And later on it turned out that if I’d gone ahead with the microwave superhet, I would have seen them. The reason he didn’t was because he had a rig which used a bolometer for its detector, and the point was, he had very good signal to noise, and I knew this, because Bob told me what the noise figure was. But a bolometer is only sensitive at high power levels, and the F center turns out to be a very easy resonance to over power with high power, you can always — so called saturation. It’s a peculiar case. It’s very easy, to saturate, and obviously what happened was, because they were using bolometers, they saturated it. But with a superhet, that wouldn’t have saturated it and it’s a great big enormous fat resonance. It would have been easy to see. Sometimes I kick myself that I didn’t do that. Then I would have been sort of the guy who opened up the color center field, you know, in spin resonance. But instead, what I decided I would do, would be to take a plunge into nuclear resonance, because there was so much nuclear resonance apparatus around and Irwin Hahn, who was out here, graduate student here –
— I wanted to ask you who else was out here when you came?
Well, Irwin Hahn (crosstalk ) — OK, Irwin Hahn had gotten his Ph.D. here, and was staying on for another year. He was a guy who had started working at the betatron, and had decided he didn’t want to be involved in a big large scale group activity, and he talked to a faculty member named J.H. Bartlett, who was a theorist, who in essence would take anyone on to do a thesis with him, because Bartlett’s belief was that people should do independent study and he’d be happy to sponsor anyone. He suggested that Irwin should read these papers by Bloch and maybe there was something in there for him to do. When I made that statement about Bartlett that was not meant to be a belittling statement. He had really the good old fashioned view of graduate study, which was that a student would pick his thesis and work independently, and the faculty member would be there to help and talk with him if the student wanted to come around, but the student was demonstrating independence in research.
Which papers of Bloch?
Well, the original Bloch, Hanson and Packard paper, you know, announcing the discovery of nuclear magnetic resonance, and there were several papers that came out around 1946. Hahn had been a radar technician during the war in the Navy, and he felt that you ought to do things by pulse methods, because of this, and it’s very interesting, on his own he set up pulse apparatus and started looking f or transient effects in nuclear resonance, and shortly before I got out here, he discovered this famous thing called the spin echo (?) which may —
You also did some work on that.
Yes, I’ve done some work using spin echoes in different things. But he — so he was here and we overlapped for a year. Then, Herb Gatowsky had come out the year before in chemistry. He got his degree; he and George Pake got their degrees in 1948 when I got mine in 1949. I remember so well, because George Pake gave two invited papers the first year out from his Ph.D. — one on his thesis, the other on the experiments he and Herb had done after he finishes his thesis at Harvard. I mean, this was fantastic. Gatowsky had set up over in chemistry. This was before there was any commercially available NMR stuff, and Gatowsky didn’t know any electronics, but he hired an undergraduate in electrical engineering to build him the circuits for an NMR rig, and he got — purchased a permanent magnet, because that was what people were using in those days, and started doing nuclear resonance across the street in chemistry. And actually, one of the things which were great for me was that there really were — there were no chemists who did quantum mechanical calculations in the department. There were chemists who did statistical mechanics and things of this sort, but there — the result of it was that I was sort of the, Herb’s theorist, for a period of time there, and I got interested in these chemical problems, and wrote several papers with him, one with one of his students — for example, a paper on what are called chemical shifts. Harlow [???] and I wrote a thing in which we explained why it was that in fluorine systems, the chemical shifts of the fluorine nucleus correlated with the electro negativity difference between the two atoms. And this was the first — I think this really is a very basic paper in chemical shifts, because I think it was the first paper which really gave some physical insight into what was happening, beyond the simple diamagnetism of F states, which people had been talking about. Then I got involved in the spin-spin interactions with Herb and Dave McCall, and in fact, an appendix to that paper was — we had been talking about how you could have spectra which had a couple of lines in it, as a result of the fact that a nucleus would sit in a couple of different positions, and then if the nucleus can jump back and forth, those two lines can merge into one. And I thought, you know, this would be sort of interesting, to work the theory of that out, and I found a way of doing it, using the Bloch equations, and I just showed it to Herb. I mean, I just did this on the side. This really wasn’t related. So Herb said, “Well, let’s put that in this paper.” So we put it in the paper as an appendix, and it turns out to be one of the most referred to things I’ve ever done. And it’s really very amusing, because it’s used all the time by chemists, and — in studying (weight?) phenomena. You can use it to study chemical exchange rates, just all sorts of things like that. So I had a tremendous influence, out of the fact that Herb was there, and we wrote a bunch of things together. It got me very much interested in chemistry and chemical problems, so I’ve done a lot of things off and on, over the years, related to chemistry. Everything I know about chemistry I learned really out of that. I never had a chemistry course in college. We didn’t talk about chemistry in quantum mechanics, you know. We talked about scattering.
Yes. So let’s see, the solid state was then firmly rooted within a year or two?
Yes. That’s right.
Ok, we were at the point — I just want to backtrack a minute and talk about —
— solids – right —
— the solid state was begun —
— it was begun in ‘49 with the five of us, right —
‘49, with the five of you, and then —
— and then the next year, Jimmy Kaylor came —
— and Bardeen?
— no, I don’t think Bardeen came that —
— he came in ’51.
What’s that? Yes, I think he came a couple of years later.
Yes, That’s right. And then in ‘52, Wheatley.
And (John?) Wheatley came, right. Wheatley came to be a postdoc with Hill, who was a nuclear physicist, and out of this, he switched over — involved with nuclear alignment, and he switched out of this into low temperature physics. In fact, this is how he got into low temperatures, and eventually became of course sort of the preeminent low temperature experimenter, American experimenter.
Around Seitz was a whole bunch of theorists. There was, to me the most important was Overhauser, who —
You mean he was here too?
Yes. Overhauser came here from Berkeley, where he‘d been a student of Cottrell. Cottrell was at Bell lab, started taking students while he was still at Bell, before he’d made the full transition, and Overhauser did a thesis with Cottrell away a good deal of the time. And Overhauser in his thesis discovered — his thesis was calculation of the spin lattice relaxation time of electrons, and in his thesis, why, he discovered that if you saturated the conduction electron spin, so that you equalized the up and down populations, that you polarized the nucleii. And this was, you know, the discovery of dynamic polarization. And when he arrived on the scene here, what he wanted — he immediately came to me and said, “Now, look, here’s this way of polarizing nucleii, why don’t you do it?” The problem at that time was that no one had seen a conduction electron spin resonance, and so, we set to work to look for it. And — I don’t know, you probably don’t want to hear about all this stuff.
Yes. Yes. There are many things that we can talk about.
Ok, all right. Ok, this is actually maybe a little interesting, because I really think that the dynamic polarization of nucleii started a whole bunch of things. And this actually happens to couple into my experiment on superconductivity. Well, when I was a graduate student at Harvard, why, there was a man whose name I’ve forgotten, I think it was Cook, who was interested in. Faraday rotation, microwave study of Faraday rotation, and he decided to look for conduction electron spin resonance in copper. He couldn’t find it, and he asked Purcell why. Purcell got thinking about it, and came to realize, well, after all, microwaves, the skin depth is very thin, and if you were going to be — and the electrons, conduction electrons, come from the body of the metal, out into the skin region, where the RF is penetrating, then go back in, so that the spin is only in the field where the microwave alternating field is for a very short time. Do I make myself clear?
So he used an uncertainty principle argument to show, therefore the resonance would be very broad, because, calculating the velocity of electrons at the Fermi surface and the skin depth, so you can account for it, so it would be very broad. So when Overhauser described this thing to me, he said, “You know, these relaxation times are quite long, so the lines will be narrow, easy to see.” And I described this problem, you know, skin depth problem. Then I got thinking about it. Then I had been a graduate student, I had tried the following experiment. It didn’t work. But it was sort of a nice idea, you thought. You know, I was looking at the diagrams for energy level for paramagnetic ions, as you apply the magnetic field, and they start out with certain levels at low field, and then the levels at high field are different, and in certain cases, they cross. And it occurred to me, why, heck, if they cross, the transition’s gone down, way down to low frequency, and that means you ought to be able to see them in a much lower frequency apparatus. I took a sample that I had, and I put it in George Pake’s nuclear resonance rig that was running at 30 megahertz instead of 1010 cycles, and so, this was a much lower frequency — I mean, in essence zero frequency — and I thought, well, I’ll see if I can vary the field, and spot the (zeros?) level crossing. Well, it was kind of a neat idea, you know — but it didn’t work. We never found it. I don’t really know — I think probably somebody subsequently has successfully done that experiment. I think part of the problem in this case was it turned out that the crystal field parameters turned out to be all different from what we had at that time thought they were. But that gave me the idea, you know, of doing a nuclear resonance, electron spin resonance with a nuclear resonance apparatus, and in thinking about this thing that Overhauser had taken — Ok, here’s a line which is going to be very narrow, very intense, and I realized: well, if I can lick the skin depth problem, I’d just go down and use a nuclear resonance rig. And then you can work with small metal particles, and the radio frequency field will penetrate entirely through the sample. Therefore, it doesn’t matter if the skin is spin is moving. It won’t move in and out of the radio frequency field, because the radio frequency field will penetrate the entire sample. So I realized I’d get rid of what was the line broadening mechanism that made it so the experiment didn’t work. So we made a little solenoid, and put together a nuclear resonance apparatus. Then I had a very bright student named Don Holcomb, who is now at Cornell, head of the department there for several years just a few years ago, and his thesis was going to be to find the conductional electron spin resonance, and using a solenoid. You see, we didn’t really know how narrow it would be. Was it going to be a mille gauss broad? Or was it going to be two gauss broad? Something like this. He looked and looked and looked, in all the good metals, you know. We looked in copper. We looked in aluminum. We looked in silver. We looked in gold — all those things which you think of as being good metals. And we didn’t find it. We even had a sodium sample around, which we tried, but it was sort of an old one, and we didn’t see anything. So then we gave up on that and switched to another topic, so to speak, which worked out nicely and so on. But a short time later, why, there was a publication from Griswold, Kipp and Cottrell out at Berkeley, in which they reported a conduction electron spin resonance in lithium and sodium, with microwaves. And the resonance was only a few gauss broad. Of course, this excited me enormously, because I realized now we could do the Overhauser experiment, because there were those resonances. My students were down in a seminar. Don Holcomb was looking at the nuclear relaxation in metallic lithium, and he had a sample in his rig. I went in and I grabbed the sample and I took it in. We had a little transitron oscillator, so called, for measuring magnetic field, and I stuck this sample in that and stuck it in the solenoid and hooked a Variac up to the solenoid, so I could sweep the field way back and forth — displayed the thing on the oscilloscope, and there was the lithium resonance, a great big fat enormous thing. If only we’d looked at lithium, we would have been the people who discovered it! It’s a terrific lesson, you know, about staying with an experiment, and trying and trying and trying. I mean, you never know when an experiment —
— how did they happen to look at lithium at Berkeley?
I don’t know how, why it was that they happened to look at lithium and sodium. We then saw it in sodium as well. We had a new good sample. These are reactive metals, you know, and very small articles, they can kind of deteriorate. So then — but now you see, I’d had all of this thinking, which was sort of a reversal of what everyone else was doing, which was, work down at low frequencies. You know, do electron spin resonance in the mega cycle region, instead of up in the microwave region. And we immediately set out to try to demonstrate Overhauser’s thing. One of my students, Tom worked on that. He’s now a professor at Princeton. We decided, first thing that we would do, we knew we had to have alternating fields around five gauss in amplitude, which is very strong, so we really wanted to be down in the region where you had big fat oscillators. Up in the microwave region, we just didn’t have that kind of power, and we had the skin depth problems and everything, so we set about doing this experiment down there, and it worked. I keep a picture of it up above my desk, because that experiment got me promoted from assistant professor to associate professor, and the next year to full professor! You know a whole bunch of things. And got Tom a job at Princeton.
Let me ask you, before we go on, about this new field of magnetic resonance.
It seems to be a productive one, particularly in that period. I haven’t looked into this yet, and probably will in years to come. I’m just wondering, where in that period the major groups were? One was at Harvard —
One was Harvard. There was a good group at Rutgers, Henry Tory and Peter Weiss. There was a group at Stanford, I mean, Bloch, Felix Bloch, and a group began, there was a group at Berkeley, and then, I guess the group here at Illinois. Actually we had as much action as anyone; I really believe that an awful lot came out of here. With Irwin Hahn and Herb Gatowaky both, you know, here — of course, Irwin was here for just a year after he finished his thesis. Then he went out and post-doc’d with Felix Bloch at Stanford.
All of these groups were coupled in this strong effort in solid state physics.
Well, I would say that’s stretching it. The people at Rutgers were interested in, Weiss and Wagner were interested in, of course, paramagnetic ions, and Henry Tory was interested in some fundamentals of nuclear resonance, although he did an analysis of diffusion effects. The people at Harvard were looking at sort of applications, sort of motion in solids. That was the principal thing that was going on there, although Purcell got interested in solid hydrogen. The people at Stanford were mostly interested in liquids. They were sort of on the borderline between physics and chemistry, really, in terms of the things they got interested in. They got interested in high resolutions spectroscopy, and of course there was a strong coupling between Bloch’s group and the people at Barium. He influenced the Barium people a lot, in what they did.
Barium is right in that area?
Yes, that’s right. It’s right on Stanford campus, it’s on Stanford’s own land, and — But I think it’s really the case that our group went after what I’d call the classic solid state problems. Dick Norvard’s first thesis with me — he was my first student — Looked at hydrogen and (poladium ?), and studied the motion of it. Then, he and I looked with Herb Gatowsky at self-diffusion in alkaline metals, stuff like this. I mean, we got off on these types of directions. But we got, I got strongly influenced by the many body people, because Dave Pines came along about that time.
When did he come along?
Well, now, I don’t remember the exact year but it might have been ‘53, probably.
I can check that.
Yeah,’53 or ‘54. Because he had been doing many body calculations, you know, of the electron gas, this work that he started with Bohm for his thesis, and one of the things which he’d done was to predict that the spin susceptibility would be influenced by the many body interactions. He and I were standing in the hall one day talking and he said, “It’s too bad one can’t measure the spin susceptibility by itself,” — you know, all susceptibility typically has orbit as well as spin , and at that time, we’d already seen the conduction electron resonance. I said, “Well, I know how to do that experiment,” because the point is that the area under the absorption curve is proportional, well known constant, to the static susceptibility. And if it’s spin resonance you’re looking at, it’s the spin susceptibility, spin’s contribution to it. Once again — I mean there’s a whole set of experiments here, with the same theme running through them, which is kind of my point here. I realized, the way to do that experiment was to use a nuclear resonance apparatus. Your problem is if you want to measure absolute intensity, you see, it’s a very hard problem. You’ve got amplifier gains. You’ve got losses in the circuits. And to measure the area under an absorption curve is in essence making an absolute absorption measure. And that’s very hard to do. The sorts of corrections were like 30 percent corrections, to the theory. So the question of, how are you going to do that in such a way that you have a measurement which you have that sort of confidence in, of an absolute sort? Well, I realized that in lithium, you have both nucleii and — like sodium — you have both nucleii and electrons. And if you use a nuclear resonance apparatus, and use it as an iron magnet, up at 10,000 gauss you can look at the nuclear resonance. And if you turn the magnet down to 5 gauss, you can look at the electron spin resonance. You have the very same apparatus, the same amplifiers, same coils, same sample — everything’s identical. And what you do is you compare the area under the nuclear resonance with the area under the electron spin resonance. And by area, I mean literally the area on an oscilloscope plot, the absorption line. And the ratio of those two areas is the ratio of the spin susceptibility of the conduction electrons, to the spin susceptibility of the nucleii. Well, the nuclear spin susceptibility is just given by the (Lanjaman?) Formula, so you know it exactly. So the point was, to change the thing into a relative experiment, and measure these terms — and that was an easy experiment to do. Bob Shoemaker did that then for his thesis. That’s one of the first papers which actually did a many body, which really measured a many body effect in a solid. Be use this was a many body correction, to the spin susceptibility. And then — and this got me really interested in many body physics. About that time, Bardeen came, and gave a talk about his early theory of superconductivity, and you know, he and Froehlich had these theories which were in correct. But he gave a colloquium on it, and I didn’t understand the colloquium very well, except I did —it’s not a slam against the colloquium. It’s just that, you know, it was pretty abstract topic. But what was clear was that something was happening at the Fermi surface, and in essence, he was describing something like an energy gap at the Fermi surface. But I knew that those were the electrons that provided the nuclear relaxation. And — in a metal — we’d been working in metals. So that if something happened to those electrons that made them different in a superconductor from in a normal metal, then the nuclear relaxation in a superconducting metal must be different from that in a normal metal. Now, the trouble is that a superconductor is a perfect diamagnetic. So it excludes the magnetic field. And Fred Rife — I’m not sure just at what stage he’d looked at the mercury particles, whether it was before I got this idea or afterwards.
Where was he?
He was at University of Chicago. He’d been a student of Purcell’s. And he made some very small particles of mercury, smaller than the superconducting penetration depth, so that he could actually look at the nuclear resonance in the superconductor. But you could do that, but that’s for mercury, where the nuclear relaxation is very fast. While I was sitting there, I suddenly realized how to do this experiment, which was to — you cycle the magnet, and use the magnet — turn the magnet on, so you suppress the superconductivity when the metal’s (normal?), — in the metals normally, you look at the nuclear resonance. You turn the magnet off. The sample goes superconductor. And the nuclear spins get very cold, because it’s an 80 [???] magnetization. And now they’re much colder than the lattice, and they warm towards the lattice, and then you turn the magnet back up again, and inspect when the thing gets to be a normal metal and you can look at the resonance again. If there had been no relaxation, there would have been no difference between the resonance before and afterwards. If there’s relaxation, why, the nuclear will be warmer and the resonance will be smaller.
Did Bardeen work closely with you?
No. No. This was —
Did you get started when —
Yes, what he had to say specifically describing that, I just sat there in the audience and —
— was it about ‘52 or so?
Yes, it must have been around then. ‘52 or ‘53, something like that. You know, just sitting there in the colloquium I realized first of all, I should do the experiment, and saw how to do it, you know, just sitting there. I left the thing, of course wildly excited — saw this is a great experiment. And I had Tom [???] who was still there, hadn’t left yet, and my original idea was that we would do this experiment using the solenoid which we’d built to do these other low field experiments. Because I realized that to turn the magnet off and on — look, I knew all the time that, a one second relaxation time in normal aluminum, so you have to be able to do the field cycle in a time on the order of a second, with no chance in the world you could turn off the big iron magnet in that time. But I realized that you don’t have to work at those high fields in order to do these things. I mean, there’s a whole stream of things here, you see, where we just did experiments that are totally — which were absolutely different from what everyone else was doing, because everyone else was using big electromagnets, up at high fields. And we just t hit the scheme of things, you know. Once you realize there are things you can do which other people didn’t think about, your mind just gets going in a different direction, and so — Actually, we ended up making a special magnet for this, using laminated iron and so on, and Chuck Hebel did this for his thesis. But it was really interesting, because Al Redfield got the same idea independently, for this experiment. And he described the experiment to Felix Bloch, and Felix Bloch said, “It’ll never work.” Al said, “Well, why?” He said, “Well, once you turn the field off, why, you’ll lost the magnetization.” I realized that wasn’t so. I’d taken a course from C. J. Gorter, a Dutch physicist, at summer school.
Where did you take it?
At Harvard, in summer school, in a course on paramagnetic relaxation.
Oh, 1948, something like that. And so I was familiar with ideas of spin temperature, and I’d been involved with it in my thesis, you know and thought about these things. I realized that, you know, you could think about it in these temperature terms. You know, you can demagnetize and you can re-magnetize, and down zero field, there’s zero magnetization, to be sure, but the order is still there. If you turn the field back on, you recover the magnetization, because the order hasn’t changed. If it’s a reversible process, you’ll get it back. You know — see, this isn’t the way resonance people thought about experiments. So that’s why Bloch wasn’t thinking that way. Redfield went ahead with it. He later told me that actually, he decided the experiment was too hard to work. And we were ahead of them, anyway, and we made it work. And it was — we made it work just before BCS came out, and we had this really remarkable result, which was that when the thing went superconducting, the nucleii relaxed at a faster rate than they did in the normal metal. And the reason that was peculiar can be seen if you think about a too fluid model. A too fluid model of a superconductor, you have normal electrons which can be scattered, and you have super-electrons which aren’t. And presumably, if the nucleii are relaxed, they’re only relaxed by the normal electrons, because they have to exchange energy. So those are the only ones they would relax, they would interact with. So a normal metal has the full amount of normal electrons, and a superconductor has less. Therefore, it should be less able to relax the nucleii, so the nuclear relaxation time should be slower. We got this idea that, you know, the idea of the energy gap had been mentioned at that time, and we got the idea then — you know, worked out the formulas for the way this went — that the, if you took the states, the uniform density of states in a normal metal, and you put an energy gap in it, (you’ve got to conserve states, however, you don’t conserve or destroy states), and you put the gap in at a given place, so the states must pile up on the edge. So the density states must be higher in the neighborhood and then fall off. This is pre-BCS. So we tried approximating, taking a gap — people had a rough idea what the gap ought to be. The [???] concept [???] had been talked about then. I mean, the concept of an energy gap wasn’t that odd, we didn’t originate it, and people had talked about it. In fact, we saw that could —
This was now about ’56?
‘56, yes, that’s right. We could see that in fact, you could enhance the nuclear relaxation rate, if the gap was rather small and the states were piled up on the side. So this was really great — I mean this appeared to confirm that you had that sort of density state. Later, after BCS theory came, which was just immediately after that, the ultrasonic experiments had this other result, which was that the scattering of sound waves drops as you go below the TC. The scattering rate of nucleii goes up. And if you have a simple one electron theory, they ought to both do the same, because they’re both in essence zero energy exchange processes. I mean, the quanta are so small — you know sound waves or nuclear spin flip energy differences — these are such low energy differences that it’s in essence a zero energy transfer relaxation process. The only difference between the things was the matrix element. But I mean, you know, you have the square of the matrix element. You have the density of final states, from the Fermi Golden Rule. Then you have the probability that the final state is empty, which is one minus the Fermi function. Then you have to add up over all the initial states, which is proportional to the density of the states at the initial energy, which is the same as the final energy, multiply by the Fermi function, and integrate it over all electrons. So, we’d have the square of the density of states, F times 1 minus F times the square of the matrix element. And you know the answers got to be the same, no matter what matrix element you’ve got in there — unless that matrix element does wild things, at the Fermi energy as a function of energy. I mean the difference between one thing and another is just sort of the magnitude of the matrix element. So ultrasonics and nuclear relaxation should have the identical temperature dependence. And of course what BCS did was to show that in fact, there were two terms which interfered, and that the way this thing should be calculated was, you should add two matrix elements together and square them and in one case, the two matrix elements had the same sign; in the other case, they had the opposite sign. So it is this interference phenomenon associated with pairs which makes for the difference between nuclear relaxation and ultrasonic stuff. So this was really one of the first things. I didn’t really grasp the full significance of this immediately. I mean, I knew the nuclear relaxation told us something, but the full import of exactly how the contrast between them worked, required Bardeen to really explain it.
Did you go to Bardeen with the results?
Oh yes, we went and showed him our results, and told him, gave him this trial explanation, that is, lumping up the density of states. And he said, “Yes, that seems very reasonable.” But then after they had BCS theory, of course, why then, we wanted to try to calculate it, using the nuclear relaxation. And my student Chuck Hebel and I, sort of getting lessons from Bardeen, carried out a calculation. And that was rather interesting, because in the process — and this is referred to in a footnote in the BCS paper — just going back to first principles — I mean, Bardeen could do these things in his head. But I had never done a calculation using second quantization before, so I had to go dig the books out and stuff like this, and was doing it more mechanically, and we did come across these problems of the interference effects. There’s a mention of that in there. Did you ever read their paper?
Ok, well, you’ll find a reference to “Hebel and Slichter” in a footnote in there, having to do with these matrix elements, and the interferences.
Yes. Did you interact with the other people, with Cooper and [???] also while you were here?
Yes, but I interacted most with (Schrieffer?).
In what way?
In that we were talking with him about the results of our calculations. For a while we kept making mistakes. And then we’d go — you know, we had questions about what was going on in the thing, and we would go ask him and talk with him. Bardeen showed me how to do the original thing. And then I went through a sort of — in other words, he sort if explained how one did these calculations. But it wasn’t a very systematic thing, in the sense that — I mean, he told me the difference in energy between this state and that, and I finally had to sit down and sort of work out the Hamiltonian and, you know, fill in a bunch of steps f or myself, to understand what was happening. But he, in essence, outlined what one was to do, and you know I was just trying to understand what he was saying. But Hebel and I, in essence, did, on our own, carried through an independent calculation of the relaxation time. In fact, we each did it independently, so that we could check each other, which was sort of nice. That’s what I always do with my graduate students — if there’s a theoretical part to something, and I have done it myself, why, then I have them do it, without letting them see what I have done. That’s a really important thing, because that way they really feel that they did it and you know, if a guy’s done an experiment, and you go in and work out the theory on it and then, that’s it — you swipe from him a really important part of his work. You’re ladling off the gravy, so to speak and he’s left with, sort of feeling like he’s a pair of hands. I think that’s the worst thing you can do. So I’ve always felt that — I’ve always done this. If I’ve calculated something — no, I didn’t do it with Dick Norbert. That’s where I realized that I should have, that one should sit down and do the calculation, and not show them. Tell them, “Why don’t you do this calculation?” It’s a very good technique, because it checks out the numbers.
Sure. Was your work with Hebel that we were just talking about — was this one of the first tests of the BCS?
Yes, I think it really was. I think the — well, put it like this. Of course there was a whole mass of experimental evidence, and one of the most convincing things about BCS was that it explained all of these experiments. But I think that it’s really true that the contrast between the NMR and the ultrasonic was a — was viewed as a very specific, concrete verification of the fundamental idea of pairing, whereas the other ideas — there wasn’t anything that was especially unique. I mean, you could explain the heat capacity — of course, it was a great triumph that you could get the temperature dependence of the heat capacity, and, you know, the temperature dependence of an energy gap, and other things like this. But this was so, appeared to be so specific to the BCS ideas, that I think that Bardeen and Cooper and Schrieffer really felt that it was, maybe the first. Well, the other thing I guess was, what happens on the electromagnetic penetration; transmission through a thin film. You know the work that Tinker had done.
Oh, yes. Before we leave this paper with Hebel, I notice in the paper that — the Letter — that Hebel had a pre-doctoral fellowship from GE. Now, I keep on hearing that the solid state program here has had strong connections with industry, but I haven’t yet figured out what these connections were, and I was wondering whether this was one of the important ways in which industry connected with research here, or were there others?
Well, we did benefit, in terms of having fellowship support from industry. But I would say that, a more important thing is that there’s a lot of solid state physics going on in industry, and a lot of very goal solid state physics. People like Bardeen had been in industry.
How does that relate to working with [???]
Well, I think really just in that we had contacts with the people, and we knew a lot of them, and we knew a — you know, in a lot of fields in universities, people sort of look down their noses at industry, sort of feel, you know, that it’s beneath them. “Those people work on applied things rather than pure” — stuff like that. That never was the case with solid state physics.
The connections were quite informal.
Except for the fellowship.
Yes, I would say so.
Were the fellowships common or was this unusual?
I don’t know how many we had. I suppose we had probably a half a dozen fellowships. I want to make one other comment about that paper before we leave it. That’s the best paper I’ve ever done, without doubt.
Speaking of the one with Hebel?
Yes. Yes. And the thing which is funny about it is that there’s no one audience that can understand all of it. I mean the significance of it because that paper has another thing in it. That paper has the first calculation of a nuclear relaxation time with the magnet, turned, off. Usually when people calculate nuclear relaxation times, why, you have the spins quantized in a strong field, and so you can begin with a strong field approximation as to what the energy levels are. But when the magnet is off — which is the situation in this case — you don’t know what the quantum levels are. And we solved that problem, as part of working out the theory. I mean, we solved it for the normal metal and we also solved it for the superconductor. The thing is, you know, unless you‘re a resonance person, you don’t realize that that was a major advance in resonance theory, you know. And of course the superconducting people don’t — they don’t know about that, you know. Then, I think a lot of people don’t know that it was sort of a tour-de-force, you know to be able to measure relaxation in a superconductor.
The theorists, you know — these days, if you read a book, it says something about superconductivity, and then it makes a little mention that the nuclear relaxation time in a superconductor is, you know, reveals something about it. But the fact that figuring out how to measure something in a superconductor — the damn superconductor excludes the magnetic field, you know — and things like this — people just don’t even bother to mention that, that it’s a problem. So, it’s funny, I guess I really feel that I’ve never — I guess there are two people, maybe there are three people in the world who really know what’s in that paper. You know, Chuck Hebel and me, and I guess — well, say John and Bob Schrieffer and Leon Cooper, I think, understand those other aspects of it, that — that, you know, that paper could have been actually maybe three papers, in terms of the [sorts ???] thrust of things that are in it. So I really — that’s a neat paper, I think. Of all the papers that I’ve done, I have a special feeling for that one.
I was very impressed when I looked at it, although I certainly didn’t understand it in detail. Let’s turn to some general questions I have. I’m getting to the — the interview that you did with Joan Warnow and Bob Williams, will be filed together with this interview, and so we don’t have to go over the same ground, but I do have some questions to ask you, after my re-reading of the transcript. You pinpoint 1950 as a special year, because first of all, that was the year of the experiments carried out at Rutgers by Bernie along with Reynolds, right, and Nesbitt, and also of E. Maxwell at Bureau of Standards, which showed that the transition temperature of a superconductor varies inversely with the square root of the of mass.
Which I guess then made very clear that superconductivity was due to the interaction between the electrons and [fermions???]
The other important event that you mentioned in 1950 was the theory and Bardeen’s unsuccessful attempt to develop a theory based on the self-energy of the —
— right —
— electrons in the field. Now, you don’t mention the Russians at all. And 1950 was also the year that the papers of Ginsberg and Landau came out in JETT and I heard that these were — I gathered that those were among those that were dumped in the harbors in the midst of the McCarthy era, and I wondered if these papers played any role at all in what was going on?
No, they didn’t play any role at all in the BCS theory; in the development of the BCS theory.
No. No, I don’t think so. They really — they are —
— I’m not even sure they were known in that country.
Well, I’m not sure they were known, either, and I don’t — there are a couple of things I’m not sure of, factually. I don’t know what year we started having translations of Russian journals available, because no one was reading the Russians — you know, with very few, exceptions. I don’t know whether we were getting Russian journals. I’d have to go look the facts up, on that. But even if we were, though, the Ginsberg-Landau stuff, of course, is very important, in superconductivity. But it is in some ways, a somewhat independent intellectual exercise, from the microscopic theory of BCS. That is, you don’t — it doesn’t —
We were talking about the independence of the work being done in Russia in the fifties.
Yes, at least the Ginsberg-Landau stuff.
That’s really a phenomenological theory.
Yes, but I get the impression, I haven’t studied this — I hope I can get to in months to come — that this, then, led to more detailed work, like [???] and others.
Which brought the two theories in some sense together?
Yes, the BCS theory contains the Ginsberg-Landau theory in it, and people have in essence derived the Ginsberg-Landau theory from the BCS theory, within certain approximations. The Ginsberg-Landau theory is an exceedingly useful theory for discussing a whole bunch of phenomena and I mean, it’s a great intellectual triumph, and very important, especially, however, when you’re talking about type 2 superconductors. I think this is where it had a big role.
But it played essentially no role?
I think that’s fair to say. Of course, I can’t speak for Bardeen, but I don’t think it played a major role. There are certain ideas which are in it, which [Papard???] also was talking about, and I don’t know to what extent [Papard] might have been influenced. This had to do with the idea of a coherence length, and the questions of the surface energy, at the interface between a normal and a superconducting metal. And Papard of course had developed a set of ideas about superconductivity, based upon his microwave impedance measurements, which led him to believe that superconductivity was very simple, a very simple phenomenon. In fact, he was traveling around the country giving — I heard him give a colloquium entitled, ‘The Simple Facts of Superconductivity.”
Well, this was just a year, I guess, before BCS. Maybe two years. Something like that. It was right about that time. You know, it’s an interesting thing — certain people had a very deep physical insight into what superconductivity was. Papard was one of them. Bardeen’s another. A very deep physical insight. But there’s a long way between the deep physical insight and a mathematical formulation.
Let’s see — some of the other people who were working on it were: [???] Black —
Black and Shack [cloth?], yes. They were on sort of an [orthogma???] course, Black and [Shaftwell???].They were trying to view the thing as a condensation, and Bardeen always felt that that was not a — that that was not a legitimate idea. They wanted to have pairs with bosons, pairs of fermions with bosons, but the point of, size of the pair is so great that the pairs are strongly overlapping, so, my recollection of what Bardeen said was, “Well, that’s rather an artificial thing to say, when the pairs overlap so strongly, that you can’t really just say, well, therefore, they’re fermions, but they’re in pairs so they act like bosons.
What about some of the other people? Was Heisenberg still interested?
I think always people were interested, but I don’t think they were working on it.
They weren’t, Heisenberg and [???]
Well, he was still very much interested. Well, let’s see — there are two London brothers, but one of them was Fritz, I forget what year he died but he died prior to BCS.
I was wondering whether these people were in connection with each other, or —
Well, yes, there were a lot of people in communication. I think if you look at —you can see reference to one another’s papers, about that time. But I would say something like this. Someone like Bardeen was working very much on his own, and he and Cooper and Shreefer, when they were working together; it was a very private sort of operation. They weren’t going around talking to lots of people. The way I describe it was, you know, people knew that this was one at the really big unsolved problems. And what people wanted to do was to solve it themselves. They didn’t want to go shoot the breeze with a bunch of other people. They would maybe — you know, they would publish things. It was, of course, a highly competitive game. And when Bardeen published his paper about the energy gap, in which he pointed out — very famous paper, in which he pointed out that if an energy gap existed, that it would almost surely give you a Meissner effect, and —
Something like that, yes. And he published this thing, and then it was immediately attacked. I think it was Buckingham who attacked him, on the grounds that his argument wasn’t gauge invariant. And then Bardeen came back with a reply — and this was a very important thing, because the only serious attack that was made on the BCS theory, I would say, after it came out was the question of gauge invariance. And Bardeen had such a deep intuitive understanding of why the London gauge was the right gauge to work in — you know, to prove that — where — whether something’s gauge invariant or not, there’s nothing wrong with taking a gauge and doing a calculation. The problem is, when you change the gauge, do you know how to do it? And there are some collective effects which take place when you change the gauge, which other people haven’t included, and Bardeen understood that they ought to be there. I think this is a way of phrasing it. Actually, Pines and Shreefer, over in Paris, the year after BCS, I think did some very nice stuff on the gauge transformation proving the gauge invariance, carrying the details through. But it’s interesting, the essential physical arguments, I believe, are given in Bardeen’s reply to Buckingham’s attack on his paper on the energy gap.
I have two questions. One, this energy gap work of Bardeen’s — did this follow the experimental work of the group?
Yes. Yes, that’s right. Bardeen was always up on the experiments, you know, who was doing what. He’s fantastic in this regard; one of his really strong points. And he knew the concept of this energy gap was there, that people were seeing it, and you know, his early theory had something like an energy gap in it. It would be interesting, probably today I’d understand those papers — you know, if I went back and read them — although I didn’t understand them at the time, and — but I think, you know, it had something like that in it. It was a — but it arose for a different reason. It was a self-energy instead of what you might call an off diagonal energy channel.
Do you know if these people who were doing the experimental work understood the significance?
Well, they understood yes — you mean, did they understand that their work implied an energy gap? Oh yes. Yes, they did, they saw — the point was, for example, in the specific heat experiment, was that they had an exponential of 1 over T, they said, “This looks just an energy.” So you had a discrete energy level, up above the ground state. And they pointed this out. But the interesting thing that Bardeen did was to say — now, let’s just take a very simple thing. Let’s just assume that we have an ordinary metal, in the way that we’re used to talking about it, with those wave functions and everything like that, and we have the density of states which is just the same as usual, except we have a little region above the Fermi energy where we don’t have any states. And what he treated was, the ease of a density of states which is uniform, drops to zero, then comes up again. What Hebel and I found we had to do was , pile the states up, for our experiment, and BCS theory then in fact tells you exactly how the states pile up. But just taking a very simple picture, like this — I mean, you know it takes a lot of guts to do this. He just so understood what that would mean and he just took that picture and took a simple product wave function, of the sort that you’d do f or free electrons and just used that different energy spectrum, and did a perturbation theory to calculate the electric current which would flow when there was vector potential present. And he showed that the relationship between current and vector potential was in essence the London equation in the non-local form which Papard had proposed. And so he said, “Almost surely, if you come out with this, you’re going to get the answer.” So he realized, all he had to do was find the energy gap. You know, it’s interesting — all of this previous work that he had done, he understood the difficulties that he had in all of his previous theories. His first theory, you know, he had the following trouble; he found that the KT of C [??? constant times] was about equal to H Omega, where Omega was the Debye lattice vibration frequency, so it was like K theta. And he said, the trouble with that theory was that it made the superconducting transition temperature come out about the same as the Debye temperature. Well, that would make all superconductors be up around several hundred degrees, and he realized that his big problem was how — he had to get a factor times that lattice vibration frequency. He knew he had to have the lattice vibration frequency, for an isotope effect, but it had to be something times H Omega. But you had to have a factor which really knocked the thing down, because it had to take it from several hundred degrees down to a degree or so And the big thing which really hit them, you know, where he really realized that they had the theory, was when, in the ground state, energy of the thing, he had exactly such a factor. I remember his remark, that he knew then that it was really right. And then of course, the other thing was, he could see immediately why he got an energy gap. He had all these pairs, and if you broke up a pair, why, breaking up the pair cost you a specific amount of energy, so there was a gap. I mean, before he had gotten the wave functions for all of these things, he still had a whole batch of stuff to solve, for temperatures other than absolute zero. They’d just gotten an absolute zero thing. But he knew he had the right answer. You know, just because of all of these previous efforts.
They all contributed. They were part of it.
Yeah, very much so, and one of the reasons they moved so fast, I think, after they had the wave functions, was that —
— they had all this —
— they knew all the things to do.
I wonder if you could clarify a little bit for me what David Pines’ role was in this development. I know he’d been working on the interactions of electrons with protons.
Yes. What he did was to work with Bardeen on what the effect of the lattice distortion is on the electron-electron interaction. In other words, as the electron moves through the lattice, the ions react to the electron and displace. In an insulator, this gives rise to what’s called a polaron, and Pines, we, and Low had done a famous paper on the Polaron, which is the way the effective mass of an electron moving through an insulator is changed by the lattice distortion. And then Pines — he did that while he was here as a post-doc working with Bardeen. Then he turned his attention to how you do the same thing in metal. And the point is that in the metal — in the insulator, when the electron moves, its electric field acts directly on the ions, so it’s an electron moving through silver chloride, for example — the silver ions and the chlorine ions are either attracted or repelled by the electrons, so you can see those lattice distortions. In a metal, the thing is complicated by the fact that the electric Interaction is screened by the conduction electrons. And this whole business of the way you handle screening, with plasma effects and things like this, was exactly what Pines had been working on in the electron systems. That was the type of thing that he included in his famous calculation of spin susceptibility and stuff of this kind. So in essence, what he did then was to re-do the polaron problem, setting up the interaction, with all of the many body effects of the metal. And so in essence, they had developed a Hamiltonian which described this interaction. I think some of that had been done also independently by Froehlich, but I’m not sure how much of that Froehlich actually had, of the proper form for the metal. That was a basic ingredient that you had to use, to calculate the — what was going on in the superconductor. So to speak, it gave them a starting point, gave them the starting Hamiltonian. What BCS then went ahead and did was to figure out how to solve that Hamiltonian. But it was typical of Bardeen that he would try to break the problem apart. I mean, so he said, “Let’s just examine how to treat the electron lattice coupling, and the electron-electron coupling, through the lattice,” and they, you know, did this careful job, to include screening and all of those effects. I don’t think Pines ever worked on superconductivity here, per se. After he — he left here to go to Princeton, and was on the faculty. He was there at the time BCS theory was done, and he —
Cooper came here, and Pines went there.
Yes, well, Cooper was at the Institute for Advanced Study, whereas Pines was in the physics department. I think he was a physics — an assistant professor, something like that.
When Cooper — I guess I can get this from the people themselves, but just as someone who was here when this work was going on, the work of the three, Cooper and Schrieffer and Bardeen, did they talk to other people about what they were doing? Or did they work very closely together?
Well, they certainly didn’t talk to me. I don’t know they might have talked to Frances Low or someone like that. But —
— they didn’t — give seminars?
No. Not that I recall. I don’t know, there might have been theoretical seminars, but I know darned well —I was working on this experiment at the time, and I certainly, I didn’t even know Cooper was working with Bardeen, on superconductivity, I don’t believe, at that stage.
Ok, now, when you first heard about superconductivity from Bardeen — that’s covered on the other tape so we don’t need to go over that now.
You do mention in that interview, you went to a conference in ‘51, was that [???]? On the way home, people were talking about —
— yes, that was the Gordon Conference —
The Gordon Conference —
On — that was many body effects of solids, something like that.
And so the general reactions to the theory I guess were discussed there.
Is that a place, if I looked in the Proceedings of that conference —
Well, there are now Proceedings of the Gordon Conference. The ground rule of a Gordon Conference is that there are no publications. Everything you say is off the record. And these comments were simply — this was just a discussion in an automobile, driving from New Hampshire down to Boston. Dave Pines and I think Paul Martin, and Brian Papard and I. You might ask Dave Pines if he remembers this conversation, because I remember he told — did I tell about Feynman in that? I think I did. About Feynman’s reaction to that?
Oh, you mentioned that some major person who had been working, major physicist who had been working on superconductivity couldn’t bring himself to read the paper for a while.
Yes, that was Feynman — until he’d had a big triumph that was Feynman, and the other person was Papard. Interesting, I think that — I really believe that an awful lot of things that Papard has had to say, in subsequent years about how solid state physics is all over and there’s nothing exciting to do and stuff like that, is really just a, in large measure the result of the tremendous bitter disappointment that he felt, that he didn’t solve that problem. He felt, as I said, he felt it was an essentially simply problem. But you know there’s not a chance in the world to what he ever would have solved it, for the very simple reason that he’s never done quantum mechanical calculations, and this was a very sophisticated kind of quantum mechanics. Papard has the intellectual power to solve the problem, but he just didn’t have the tools in his tool set. It really took a specialist. This is a perfect example of a theoretical problem which would never be solved by experiment. There’s some theory that experimenters can do, but no experimenter could have done this one.
I’d like to close this session with some general questions on, why is superconductivity so important; the implications of superconductivity? First of all, what it did for the department here. Secondly, what it did for solid state physics in general. Third, what it did for physics in general, in terms of other fields that it started and finally, society starting with the department.
Well, of course, I mean, it was a source of fantastic excitement.
Would you call it a major development, or is it just one of many?
Well, I would — yes, it’s probably — no that work has to be considered the most important. If one had to list, if one had to pick out a single most important contribution to physics, I think that would be it. Now, people can argue that of course. You know, this is a Judgment. But if you measure it against several things — first of all, the difficulty of the problem, and the people who had worked on it unsuccessfully, that’s one way of measuring things. Another way of measuring things is, in terms of the impact that something has had on a field, and this theory has influenced not only superconductivity — it’s just spawned enormous activity. It has had a tremendous influence on understanding condensed matter. I mean, it’s important in underlying nuclear physics, in astrophysics — it’s — you see, this is the only real many body theory which has ever been done. I mean, by many-body theory — theory in which you really take the interaction into account, and don’t sort of get there by perturbation theory. You know, this solved the whole damn thing in one big bite. There have been a lot of other things that people have done that have been modeled on it. It’s had an enormous effect, I think, on the direction of theoretical physics, the thinking, the concepts that people apply. Just because, as I say, it’s the real many body problem, makes people realize. They understand so much more about what many body effects are. They just think at a much deeper level, as a result. It’s a truly major insight. There’s nothing like it that’s come along before. I mean, here’s a complete theory of a phase transition, worked out on a microscopic basis. You go from absolute zero right up through the phase transition, and the spin goes normal. And the only other theory of phase transition that there is, really, you know, is the Curie-Weiss theory of ferromagnetism, where you assume a molecular field, and most other theories are that, in one form or another. But this one is really basically different; I think it’s fair to say. It doesn’t sort of start out with a simple ad hoc assumption, you know, and make the thing essentially classic. So what’s your next thing? You said science as a whole?
Yes. I think you’ve answered it.
Yes. It’s just really a very major and sweeping thing, and its importance has grown, rather than diminished.
Thank you very, very much.