Edward Purcell

Sopka:

This is Katherine Sopka. I'm visiting today, November 23rd, 1976 with Professor Edward Purcell in his office in the Lyman Laboratory. In the interest of compiling a history of the Harvard Physics Department in recent decades, Professor Purcell has kindly consented to share with me his recollections of trends and events that have shaped the course of that history since his arrival as a graduate student in 1934. Professor Purcell, perhaps we can begin by asking you to recall your impressions of life within the department when you first came.

Purcell:

Well, the contrast with today that stands our most sharply when I try to think back is how serenely uncrowded the place was in those years starting in '34 with me. I was of course very green, and never having been at a place like Harvard before and having studied engineering rather than physics mostly before, although I'd just had a year in Germany in physics, so I saw the department very much from the worm's eye view, but still there weren't very many students. As I remember there were perhaps 20 or 25 graduate students. The classes were small. The professors were on the whole I thought very, very kindly and really there were some outstandingly kind people — Saunders and Oldenberg come to mind from my first year. I had courses with Kemble, a great teacher, although that wasn't the first year and Chaffee. I remember that when it became time to find a thesis topic I had a room, a whole research room, all to myself even before I had a topic. The same room now is just below this room here. It's crowded with apparatus and graduate students. I remember that the picnics in those days were rather congenial, friendly affairs. Somehow the intense pressure that's on the present graduate student didn't seem to be so acute. It wasn't that there were a lot of jobs, it's just that there were I guess fewer people. Some of my most intense intellectual experiences at Harvard were actually — one of them was in mathematics where I for the first time got a glimpse of what mathematics really is by taking Math 13 as it was then called. And of course the university was a very exciting place for a boy from Purdue, and I remember going to hear, as an auditor, Whitehead's lectures on cosmology and things like that. And we lived in the dormitories and I lived in Conant Hall and Perkins Hall at various years. There were no eating facilities whatever for graduate students, so we took our meals up and down the various little restaurants on the northern stretches of Massachusetts Avenue. It was a very, really a rather — jolly is perhaps not quite the right word, but it was a, there was a good fellowship among the mathematics and physics students in the halls. I set out to do my thesis. Well, first Professor Chaffee gave me a thesis topic. I had taken his course in electron physics the first year I was here I think. He gave me this thesis topic which actually would have been a superb topic. Indeed if I had had the sense to stay with it, it would have resulted in the invention of what was later called the omegatron about ten years ahead of its time. Chaffee had the idea that it might be interesting to run a gas discharge in a magnetic field and then measure the impedance of the discharge with an RF our bridge and maybe you would see something funny happening at the cyclotron frequency of the ions in the gas, and if it did, well then you could obviously have a mass spectrograph made that way. And so I started to get the apparatus together to try this, and I had the apparatus really pretty well assembled — at the least the coil for the magnetic field and a bottle for the discharge. And I started doing a lot of calculations and convinced myself that it couldn't work, and I took the calculations around to Chaffee and showed him and the argument was that the ion wouldn't stay in the field long enough to have it display any resonance before it had collisions and things like that. And Chaffee was persuaded by my calculations, and so we dropped the thing, which was really a great pity, because it was in fact a good idea and if I'd only pursued it a little farther I would have been quite likely to stumble on a method to eliminate the difficulty with the lifetime. And in fact it wouldn't have taken very much inspiration starting from Chaffee's idea to invent the omegatron. So then I, looking around again, did some experiments at the suggestion of Bainbridge and finally ended up with a thesis topic that he had suggested which turned out to be doable and interesting, and I did it. But that a graduate student could shop around like that in those days and have a room to himself while he tried out various ideas is in rather sharp contrast to the situation today. The courses were small. I, of course, took Van Vleck's electric and magnetic susceptibilities course which was the course he gave based on his marvelous book. And Malcolm Hebb and I were the only registered students in the course. I think there may have been one or two auditors. But Van lectured to us for a semester, and then he assigned a term problem to work on at the end of the semester which Malcolm and I literally spent the next year working on, and which turned into be really one of the best papers I've ever had my name connected with. And I remember we took the final exam, and in those days the final exams of course were printed, set up in type and printed on a printing press, and of course serially numbered. Malcolm and I had serial number 1 and serial number 2. I often wondered how they stopped the presses before they turned out a very large excess of copies of that examination. But that's what made it really such a wonderful time for a graduate student, with Van and Kemble, and then Bridgman of course, a very great man from whom I had thermodynamics, my third course in thermodynamics actually. The first one was taken at an engineering school and was an absolutely travesty on the subject. The next one in Germany which was a really eye opener, and finally Bridgman's, which was an eye opener in many other ways. Well the cyclotron then took over our interests toward the end of the '30s, and many of us worked on that, building it.

Sopka:

Professor Street told me yesterday that you were responsible for one particular development which helped get the cyclotron in operation.

Purcell:

Well, let's see how it was. My job was to worry about the magnet, the power supply and the control circuit for the big cyclotron magnet — well, big in those days. It doesn't seem so big now. And I built the control circuit for that with other people's help, and we all worked over there putting in wiring and things like that. I then, after the cyclotron got running, proposed and developed an innovation which was an automatic tuning of the cyclotron which was a rather Rube Goldberg kind of system which modulated the radio frequency by a rotating capacitor plate which its vicious whine really scared people in the vicinity, and this thing changed the frequency of the cyclotron up and down a little bit, and then we looked at the modulation of the beam current and used the phase of that modulation to control the main generator indirectly by controlling the field on the exciter generator. It was a negative feedback system in the days when such were not well understood by most people, especially including me. There may have been some people in those days who understood the stability of the feedback system but I was not one of them. But I finally got it working by stuffing in enough RC circuits here and there to keep the thing from, uh, to make it stable. And it worked. It automatically tuned the cyclotron, which it was actually enormous help. But it was a lot of fun. I mean, I look back on that cyclotron building period. It's certainly one of the most interesting episodes in my professional life, learning so much and I remember when we finally got the beam out and how hard it was. Of course we were copying Berkeley, as everyone did. Our machine was a slightly scaled down version of the then Berkeley machine, and we were following them, using their techniques, including the Berkeley shims that you had to shove around to try to get the beam to come out. But it was quite a good machine, thanks in no small part to Ken Bainbridge's very careful engineering. Of course like all cyclotron workers in those days, we were in retrospect ridiculously careless with radiation. But fortunately I think no one in the Harvard cyclotron suffered any ill effects such as cataracts that other people occasionally got. We let the beam come out in air and take pictures of the curing deuteron beam and air and stuff. Well, it was very shortly after that that I vanished from Harvard into the MIT Radiation Laboratory, so that there is a period at the end of 1940 to the end of 1945 when I practically didn't set foot in the laboratory here, and of course in that period, the wartime period, there were — the activities here were mainly taken over with training naval electronics people and things like that. Weren't they?

Sopka:

Yes. There were almost no advanced courses given. It was all army and navy STP programs. There were a number of Harvard physicists down at MIT though in the Radiation Lab?

Purcell:

Oh yes. There were, well, let's see if we can —

Sopka:

Or people who have come to Harvard since.

Purcell:

Well, the people who went down there from Harvard included of course Bainbridge and Street. Somewhat later Furry, who was in the theoretical group at the radiation lab starting about '41 — maybe '42, I'm not sure. Let's see, who else? Jack Pierce was there working with, who had been working with Mimno and Pierce and G. W. Pierce here. Working mainly with Mimno, Jack Pierce was down there in the Loran business with Street. Bill Preston was at the Radiation Lab, and then lots of people who were here after that — Schwinger and Ramsey and Pound. Others I guess. I don't remember what happened to all the cyclotron people. Jack Livingood was one of the people on the Harvard cyclotron; Curtis, Biggs Curtis [?]; and Ruby Scherr. Scherr came to Radiation Labs. That's right. I guess and later went to Los Alamos, as did Norman Ramsey and Bainbridge. Well at any rate, as far as I was concerned I didn't know what was going on here at all, although there was the so-called Radio Research Laboratory at Harvard which absorbed other Harvard people, including Van Vleck.

Sopka:

Yes. I believe the Radio Research Laboratory was concerned with the anti-radar.

Purcell:

Yes. That was the radar countermeasures as it was called, Felix Bloch was there, and it was organized under Fred Terman. We had rather little to do with that, because since they were doing countermeasures they were sort of one level higher security and they were supposed to know what we did, but we weren't supposed to know what they did, more or less. But as it happened, my work at Radiation Lab had very little connection with countermeasures, so I paid almost no attention to the Radio Research Lab up here.

Sopka:

Did you feel that the work you did at MIT during those years significantly contributed to your own scientific development, or was it a hiatus in your —?

Purcell:

No, no, no, it was tremendously — in fact everything I've done since, everything since radiation lab in my professional activity almost directly stems from the influence of the work there.

Sopka:

That's very interesting.

Purcell:

Not only the work, but actually the people. That is, I spent most of the war working in groups that were with whom Rabi was closely associated. In fact he was the nearest thing to my boss through the Radiation Lab was I. I. Rabi, and I worked very closely with people who had come up through Rabi's laboratory like Ramsey and Zacharias. And I also worked with the people who were at Columbia during the war in the Columbia radiation laboratory — Kellogg and Kusch and Lamb and Nordsieck, so that I — well, see the two most powerful influences on my own subsequent work were getting to know, being immersed in the physics attitudes of people from the Columbia, from Rabi's Columbia laboratory, which necessarily made you, stimulated you to think about resonances of one sort or another and how you could detect them. And then of course getting to know all the new microwave electronics techniques, particularly microwaves and problems and questions of signal and noise and things like that which were crucial in all experiments that I was going to be doing in the next several years. You see, the point was that in the radar business we had to understand for the first time for most of us the general signal-to-noise problem. How do you go about detecting a weak signal and what in principal is the weakest signal you can detect and so on, so that we were able to calculate on the back of any envelope just how we would go about detecting an effect of a given strength. And all the postwar magnetic resonance experiments for example, whether they were by ourselves or other people, hinged on that as did the experiments in radio astronomy, including 21-centimeter observation. That all just really traces back very directly to what we learned in working on radar. The whole idea of wave guides we picked up there, absolutely new to most of us. When we went to radiation lab, we had no idea about things like that. Even though I was trained as an electrical engineer in the early '30s, that stuff was completely new to me because, in my training we had never talked about lossless transmission lines, let alone wave guides. And then there was wave propagation, and that came into radar in a very important way, the kind of thing that Furry worked on through the war. It was a tremendous education that we were fantastically lucky to come out of it with that great advantage.

Sopka:

In the case of radar research during the war, I assume that people like yourself could see the total picture in a different way than the people who were on the Manhattan Project where unless you were quite high up in the project you only saw a very small portion.

Purcell:

I think that's true. Certainly in the radar business there were no internal compartments at all, only those that just made by a lack of time and concentration on one problem at the exclusion of another. But we were very close to the applications from the beginning in contrast to the Manhattan District people. That is to say, radiation lab people were working very closely with the military and we were, apart from that we were very close to the actual war because of our close association with our British colleagues who were in the thick of it, so that the very first months of Radiation Lab we had people fresh from the battle of Britain which was then going on telling us what the problems really were in trying to shoot down bombers with night fighters for instance. Later in the war there were branches of Radiation Lab in Europe — in Britain first, and then in France after D-Day.

Sopka:

Did you go to the Radiation Lab before Pearl Harbor Day?

Purcell:

Oh yes, oh heaven's yes. No, the Radiation Lab, I went to radiation I think in December 1940.

Sopka:

I see. A whole year ahead.

Purcell:

A whole year ahead. And no, our first job at Radiation Lab was to try to make a microwave radar for a British night fighter using the British 10-centimeter magnetron, which had been invented at Birmingham by Oliphant and Boot and Randolph and brought to this country by the Tizard mission and successfully copied at Bell Telephone Labs by Jim Fisk and his group there. So that was our start in radar. We ended up with many other things. Of course the lab ended up with four thousand people, but when I went down it was just, oh, I guess 30 people or so in a room, big room [laughs] with this one job to focus efforts on, although inventing a lot of things, because half the things that were required we had no idea how we were going to do. Then I stayed rather long at the Radiation Lab because I was one of the large group that stayed to write the books, Radiation Lab series, and I think there even was a period when I was sort of half time back at Harvard and half time down there trying to finish the book writing. But in any case we came back to Harvard to do the NMR experiment.

Sopka:

Was it clear that you were coming back to Harvard all the time that you were at MIT, or was that something to be renegotiated?

Purcell:

Well, my Harvard rank was faculty instructor I believe, on leave. It was a time for a few years when the faculty had abolished — the rank of — what was it?

Sopka:

Assistant professor?

Purcell:

Assistant professor. That's right. There was no assistant professor. You were a faculty instructor, which was a 5-year appointment, I believe. But before I actually came back I remember being offered a permanent appointment, because I remember that Ted Hunt and somebody else — maybe it was Street, no I think it was Ted Hunt came down to see me at lunch at Radiation Lab and I had seen him — He was of course involved in the underwater place called you know the sound lab or whatever it was here, the sound laboratory. He was director of it. So he came down to tell me that, to ask me if I would be interested in coming back to Harvard as an associate professor, to which I said I was, and that was all there was to it. I can't remember, can't date that but it must have been, I think that must have been the summer of '45. I think that had been settled at the time when I came back and we did the nuclear resonance experiments here, I think otherwise I would not have been quite as — not have been so free with coming back and using this stuff and getting things done in the shop. So I think that's probably right. We came back and saw Curry and borrowed the magnet — the old cosmic gray magnet for that purpose. And started that work in the fall of '45. Although all of us that by that I mean Pound and Torrey and myself were still Radiation Lab employees. In fact our resonance cavity that we did the first experiment in was made in the Radiation Lab in the machine shop. Our first publication lists us as from Radiation Lab not from Harvard.

Sopka:

I see.

Purcell:

In fact I tried to borrow a magnet down at MIT and did not succeed. That's why we had to come back and do it here. We would have done it down there if somebody had given us —

Sopka:

They agreed to make your cavity but not to let you use —?

Purcell:

No, they had agreed to make our cavity. It was just made

Sopka:

Oh. Unbeknownst.

Purcell:

Well, the IF group, the machine shop, yeah, they were accustomed to making cavities and Pound just put in a drawing and had it made. So that they had the guy make that cavity. I met the fellow just a week ago that made that. We had a big Radiation Lab reunion at MIT.

Sopka:

Oh, I read about that, yes.

Purcell:

Yeah. And this fella came up to me and he said, "You know, I made that cavity." He was a machinist in the Group 53 shop that made it.

Sopka:

How many people came to that reunion?

Purcell:

And I told him it was at the Smithsonian now. He could go see it if he wanted to. Over three or four hundred people. Yeah. It was an interesting affair. Rabi was there, and DuBridge, Killian and Jerry Weisner [?] of course, lots of the old fellows came there. H. Guiford Stever [?].

Sopka:

Oh. NSF.

Purcell:

Bob Dickey. A lot of people I hadn't seen for a long time. Getting kind of old. Well when we came back to Harvard after the war, of course for me it was a very exciting time, because we were into nuclear resonance and things were really popping. Bob Pound came back with me as a Junior Fellow, and we were all full of the microwave stuff and everything, and then of course at the time in nuclear physics the whole world was opening up in a most exciting way also, so that there were so many things to do and teach. The Mark I computer was here and then it left — It chugged away through most of the war here, didn't it, the Mark I?

Sopka:

I believe it began operation during the war.

Purcell:

During the war. That's right. It spent most of its time doing Bessel functions order one-third, and there was a good reason for that. One of the unexpected phenomena in microwave radar was called trapping, which was the radar waves normally which would go in straight lines and therefore could not see a ship around the horizon would sometimes see around the horizon for a hundred and two hundred miles, owing to, as it turned out, the fact that the lower few hundred feet of the earth's atmosphere over the ocean has a gradient of water vapor in it, and this makes a gradient of refractive index which is capable of trapping and guiding a wave along the surface. And the theory of this, which Wendell Furry worked on, is one of the leading workers on developing the theory of this thing; propagation in a duct like this involves Bessel functions order one-third. And since the navy owned the computer, at least Howard Aiken was by then the commander, and the navy had purview of the ocean and used radar on it. Bessel functions order one-third were a high priority item. Of course it was before that the real age of computers in physics when we came back. Well we came back to find that the ionosphere research, which had been quite important at Harvard in the '30s under Mimno and Pierce working under Mimno, was still going on, and this was rather distressing because Pound and I were trying to do nuclear resonance experiments here in Lyman Laboratory which required fairly good receivers and keeping the noise down and all that, rather delicate measurements electrically, and the antennas up on top of the laboratory were sending up these microsecond pulses to the ionosphere, you know, which were just making a noise all over the spectrum. And after suffering this interference for quite a while we finally got enough clout (as they would say now) to cause Mimno to move his show out into the country.

Sopka:

I see.

Purcell:

And turn off those terrible pulses that were smearing ink all over the estolen [?] and angus [?] record. It was certainly an exciting time, those years, which were of course also the years when the graduate school here was expanding, and of course also the years when those of us who came back as young teachers full of enthusiasm and everything, had the marvelous experience of having those returned GIs to teach who were so tremendously eager to learn, you know, to work so hard. I've never seen anything like it before or since, certainly not since.

Sopka:

The classes were certainly large.

Purcell:

Yeah.

Sopka:

You know, I was in your Physics 28 class — in 1947-48.

Purcell:

Right, right. Well you remember those people. They were just fantastic. I mean, they would work like, you know, there was just nothing you could give them they wouldn't do, and — Yeah. Well the vacuum tube was still king. Chaffee's book was still full of important material. Saunders retired during the war — isn't that right? — sometime. He left.

Sopka:

Saunders I believe retired in '41 —

Purcell:

Maybe it was '41, something like that.

Sopka:

Kemble became chairman.

Purcell:

Kemble became chairman. That's right.

Sopka:

Apparently there was a change at that time. Saunders had been chairman for 15 years, so since that the chairmanship has been a short term duty.

Purcell:

Yeah. Well Saunders of course was a wonderful person to have as chairman for a young graduate student. I remember him with great affection, and he was awfully nice to us. My wife and I were married in '37. I didn't even have my Ph.D. yet. I was just a graduate student. You know, doing half time as a teaching fellow, and Saunders helped us find a place to live and then in the summer of '37 they let us live in their house over in Berkeley Place all summer while they went to Ireland. And oh, it was fantastic. We have never reached that level of living in our life since. We had a gardener that came around, took care of the grounds, and a lady that came in and cleaned, and all we had to do was to keep the bird baths full. You remember he was a very serious bird person.

Sopka:

I didn't know that.

Purcell:

Yes. They banded birds, he and his wife. It was one of his many hobbies. Of course music, as we know, is his very important hobby, and he had started working on violins already.

Sopka:

Yes, in that period.

Purcell:

In the late '30s. The late '30s he was doing violin stuff. Right. He had his spectrum analyzer, his acoustic spectrum analyzer down here in the foot of the stairs here, in Cruft in that room. And if you go clear down to the basement and just inside that door was his thing. On the same level as the Mark I.

Sopka:

I see.

Purcell:

Of course prior to the Mark I there was that famous battery in there. Did anybody put that into your notes?

Sopka:

No, no.

Purcell:

The 100,000-volt storage battery?

Sopka:

I had read that there was one, but I haven't talked with anyone who ever saw it.

Purcell:

Oh, that was a fantastic thing. That filled the room that was later occupied by the Mark I computer down in the sub-basement of Lyman there on the east side. The money for it had been given to Professor Duane who did that famous work on determining the short wavelength limit of the X-ray spectrum, and he was to use that for his X-ray tubes so he'd have a very constant high voltage source. And this was built — I don't know, it was here already when I came as a graduate student, but it couldn't have been more than a few years old then. Duane died — when was that?

Sopka:

He died about 1938, but his work —

Purcell:

He was retired, because I don't ever remember seeing Duane as a graduate student.

Sopka:

He was in poor health —

Purcell:

Yes.

Sopka:

— from about 1930 on.

Purcell:

Yes.

Sopka:

His Duane-Hunt Law was done back in 1916 —

Purcell:

Yes.

Sopka:

— so that this battery that you're speaking about, was this something that had been in existence before the building of the Lyman Laboratory?

Purcell:

I don't think so. I don't see how it could have been, it was so big.

Sopka:

I see.

Purcell:

And it looked very new, and it was very elegantly arranged in stacks like book stacks in a library. You know, the whole room was full of these stacks. I mean, that's 50,000 lead-acid cells it was. Each cell was perhaps a little glass bottle that was 2 or 3 inches in diameter and perhaps 7 inches high. And there were 50,000 of those things all strung along very neatly and connected up with bus bars in different ways so it could all be connected in series or two halves in parallel. And we had a fellow named Charlie Lanza who was a lab assistant technician, also had a brother Sam Lanza. I think it was Charlie Lanza who one of his main jobs was nursing that battery, because you know he'd go around and put water in, and it takes quite a lot of time to keep 50,000 cells. But by the time I was a graduate student, at least by the time I was doing my thesis, the battery had suffered a certain amount of depredation because people had gone in from time to time and walked off with a few thousand volts' worth.

Sopka:

Oh.

Purcell:

Because that was, you see, just before the days of voltage vacuum tube regulated power supplies, and if you wanted a really steady DC source in, shall we say 1936, '35, there was nothing better than several hundred 2 volt batteries strung together.

Sopka:

That's related then to the Byerly battery.

Purcell:

Yes. Well, we had battery sources for the laboratories for DC, but I think probably the Byerly thing was just a 50 volt battery or something like that probably, or it wasn't a very high voltage battery.

Sopka:

It was higher than 50 voltage, 50 volts. I'll have to see if I can determine how high.

Purcell:

Well, of course there is the thing that I mentioned to you some time ago that I learned in my astonishment that in — when was it? — in 1900 in Jefferson there was a 20,000 volt battery that was done, that was used by that fella who was repeating the Roland experiment.

Sopka:

Edwin Adams.

Purcell:

Edwin Adams, yeah.

Sopka:

Who ended up at Princeton.

Purcell:

Yes, that's right, and according to his account he used in his experiment a 20,000 volt battery that Professor Trowbridge had apparently built over in Jefferson, which is no small piece of equipment.

Sopka:

No.

Purcell:

Well what happened to this one was that in about, it must have been about 1936, it was suddenly apparent that the regulated power supply would be the wave of the future and in fact the definitive paper on it was written by Hunt and Hickman, a very important reference. Roger Hickman and Ted Hunt wrote a paper on vacuum tube regulation of voltage sources or something like that, which was the basic bible for building and, theory of regulated power supplies. And it became clear that if you ever needed 100,000 volts DC very accurate, that the thing to do was to make a regulated power supply, that it would cost less than what we were paying Charlie Lanza to take care of the battery. So although the thing was in perfect working condition, it was clear that it was uneconomical to keep it. And I remember seeing at that time trucks pulling up right out here, outside the corner of Lyman, and just shoveling in these hundreds of bottles and the truck going off piled —

Sopka:

It was just junked.

Purcell:

Just junked, piled with a ton of little glass bottles. They junked the whole thing — which is the only rational decision at that point.

Sopka:

Yes.

Purcell:

No one was actually using it at the time, but if they had wanted to use it, it would be crazy to do it, make a power supply. So they junked it, and then of course then that emptied out. I'm sure that one of the arguments was that if we junk it we'll have that enormous room then free for other things. And shortly thereafter the Mark I moved in — Whether Mark I was already in the wind when the battery was junked I'm not sure. My impression is that the battery junking occurred before Mark I, which after all; Aiken didn't have the idea for that until '39 but —

Sopka:

I believe so. I was told that the Mark I was actually constructed at the IBM laboratories in Endicott and was shipped when completed to the Harvard basement.

Purcell:

Yeah, okay. Well then the junking of the battery occurred before the — I don't know what was in the room then between, after the battery was — That was a period when I was then an instructor. I'd gotten Ph.D. in '38, and '38-'39 I was an instructor and Jack Livingood and I shared an office up in Jefferson while we were both working on the cyclotron. The teaching I remember from that — Well, my first teaching at Harvard I think was under N. H. Black.

Sopka:

Yes, that's quite likely.

Purcell:

I was a teaching fellow in Physics B.

Sopka:

Yes, that was his course.

Purcell:

Which he ran in his very well planned and organized way. So that was when I was — I guess in '37 probably was my first teaching in that. I was a lab instructor. The laboratory was up where our tea room is now, you know, at the end of the building. That was the old Physics B lab.

Sopka:

I didn't know that. I took Physics B at Radcliffe where we also had Professor Black.

Purcell:

The room then became our undergraduate library — later.

Sopka:

Yes.

Purcell:

That was the Physics B lab. Yeah. Over at Radcliffe of course everything was given over there in double. And he ran this well organized course. He was actually — I have great respect for Black. He didn't know a great deal of physics certainly, and of course was not one that we would now hire — oh, it was pretty much practical physics, high school physics kind of. But it wasn't all that bad, and —

Sopka:

You may be amused to let me put in a little comment of my own about Black.

Purcell:

Yeah.

Sopka:

At the end of the year a classmate of mine and I decided that physics was for us, and we went up to tell him. We thought that he'd be pleased.

Purcell:

Yes.

Sopka:

That on the basis of his course two girls are going to physics. And he got the most peculiar expression on his face and said, "But physics is no place for a woman. Why don't you go and take chemistry or biology?" Well that particular reaction was just sufficient to confirm any doubts that I may have had in my mind physics was for me. [laughs]

Purcell:

Very good. No, I can imagine him saying that. He was very old fashioned of course in his views about things of that sort. He'd come, you know, from Roxbury Latin where he had been Conant's teacher.

Sopka:

Oh, no, I didn't know that. I knew he had taught, Black had taught in the high schools.

Purcell:

No, he was quite a well-known secondary school teacher, and as I say, young Conant had been in his class, Roxbury Latin I'm pretty sure it was. And Conant brought him to Harvard.

Sopka:

Oh.

Purcell:

One of the early, first things Conant did as president, brought Black over into the School of Education.

Sopka:

I see.

Purcell:

And Black was supposed to be the School of Education teacher. And Black didn't like it in the School of Education. To his credit he found that what was going on there wasn't — So he came over and taught the elementary courses in physics. And he remained an assistant professor, you know, his whole life.

Sopka:

Yes, I knew that he never did any research.

Purcell:

He never did any research, never pretended to. As I say, I had great respect for him, because for all his limitations he was a very straightforward, modest, you know, and a great, great character.

Sopka:

And certainly for physics at that level, he was a very effective teacher.

Purcell:

Very effective, yeah, yeah. It was a well-organized course. The students knew always, as you remember yourself, what he —

Sopka:

And he had at least two books that he had written.

Purcell:

Oh, he had innumerable books. You know his books — I studied high school physics from his book, Black and Davis, which was the high school physics text of the 1920s. There's no question about it. And then he had a chemistry book that he wrote with Conant. I'm sure Conant wrote most of the book, but — and Harvey Davis, who was the Davis of Black and Davis, was at Harvard then, Dean of Engineering or something. But Black's books sold more than 2 million copies total I learned one time.

Sopka:

Oh. Because then he had also an elementary physics book that was just under his own name.

Purcell:

Yes. Practical Physics it is called I think, something like that. That's the book we used in Physics B.

Sopka:

Yes.

Purcell:

That's right. This high school book was Black and Davis, and Black and Conant was the chemistry text. They were sort of secondary school level books, even for those days, but they were practical. You learned how an automobile works, and things like that.

Sopka:

Sure.

Purcell:

And then I worked under Saunders in all of Physics C, which was I guess you know [???].

Sopka:

Physics C was, according to the catalog, only open to students who had already taken physics in high school.

Purcell:

Yes, yes. Physics C was the present equivalent of Physics 12.

Sopka:

Yes.

Purcell:

And Physics B was the present equivalent — was equivalent to the present Physics I, a very parallel thing. Physics B was the thing that premeds took and people that hadn't had physics before and weren't going to take it again and had to take it for one reason or another. They were in Physics B. But Physics C was Saunders, and, that was quite a education for a future physics teacher because Saunders was a person who talked with great — or very debonair and charming figure. And I remember his famous lectures in acoustics in Physics C, you know, when he would play instruments and stuff like that were so — One time he — if you remember the demonstration called the "Chladni Plates"?

Sopka:

Yes.

Purcell:

You know, where you have those —

Sopka:

The patterns that sound —

Purcell:

Patterns. Yeah, yeah, you have a square plate and a round plate and a triangular plate and then you sprinkle sand on 'em and stroke them with bow —

Sopka:

Yes.

Purcell:

And then you get these patterns.

Sopka:

Patterns.

Purcell:

Well I'd seen Saunders do that, and it looked absolutely easy to do. There's just nothing to it, you know. And that semester, one semester he was going to be away, had to go to a meeting or something, and he asked me to give the lecture, that particular lecture. So I did practice the day before. I went over there and got the Chladni plates out and got the bow, put the sand on the plates, and I couldn't make the patterns come out at all. You know, and it was a mess! And the more I tried, the worse it got. And I tried it in lecture and it didn't work either. And of course but Saunders was, you know, he was a musician, played the viola, and when he took a bow in his hand —

Sopka:

He knew what to do with it.

Purcell:

— he knew what to do with it, whereas — [laughs] Oh dear. And all the other things. Blowing — I think you had to blow a clarinet or something and it was a real fiasco. Well, and of course Lyman was around. When I was a graduate student I listened to Lyman's lectures in optics, physical optics.

Sopka:

Oh.

Purcell:

And then Lyman was a figure there, always in the colloquium, he'd be sitting there, front row. Of course Lyman would have given you the same advice that Black did probably, about majoring in physics.

Sopka:

That is very likely, because I found a letter that Bergen Davis wrote to Lyman about a girl who had taken her degree with Bergen Davis, her undergraduate degree, and was looking for graduate work, and she was from the Boston area. And Bergen Davis wrote to Lyman and suggested that she do graduate work at Harvard. Lyman's answer, the gist of it was that he didn't know how to cope with a woman at Harvard [laughs].

Purcell:

That's right, that's right.

Sopka:

They'd never had one, and he didn't know what he'd do with her if she came, and couldn't she go someplace else!

Purcell:

I was thinking of Lyman one time years ago and I was colloquium or seminar or one of our very bright physicists was sitting there nursing her baby, you know, calmly, but I was thinking of Lyman in there turning around and seeing that one [laughs]. He was a very dignified gentleman.

Sopka:

Apparently Lyman never married.

Purcell:

Oh no, no, he never married. No, oh, no. He was — As was Professor Hall.

Sopka:

Oh, I was never aware of that.

Purcell:

Professor Hall was an interesting figure too in those years, in my years as a graduate student.

Sopka:

He was still around, but he had retired.

Purcell:

Oh no. You see, he came in to the laboratory every day and he did experiments on the Hall Effect.

Sopka:

I didn't realize that he was still active in the '30s.

Purcell:

Oh yes, yes. He didn't produce anything much, but he had the room, which is now I guess Arthur Jaffe's office right up here.

Sopka:

Yes?

Purcell:

Although it's a smaller room now. And in this room he had a magnet and he did research on the various effects, you know, of the Hall Effect, etc. in that magnet.

Sopka:

I see.

Purcell:

And he — by then he was quite old. When I was a graduate student he was really quite old. His hands shook quite a bit. But he had the magnet fixed up with sort of wooden guides so that when he got his little sample together and everything, if he could get it started in the guide he could push it into the magnet alright.

Sopka:

Oh. Good for him.

Purcell:

And he had thermocouples and galvanometers and he worked in there all by himself doing this. And he came to the laboratory every day, and in the afternoon he'd go into the research library and take a nap, sitting back in his chair in the corner, take a nap. And you know, since I'm approaching that age myself, I say it's a very, very good thing to do. In fact I remember a story about him that he was a man with a — very unobtrusive, didn't say very much, but he had a kind of a dry sense of humor whenever he did talk. I had a colleague, I had a friend who was in graduate school when I was a graduate student, a very brash fellow, really — those days, then nowadays — But he was fascinated by Professor Hall and one day he said to Professor Hall something like this: "Professor Hall, it seems odd that after all the distinguished work you've done you wouldn't sort of take it easy, that you still come to the laboratory every day and work," you know, something like that. To which Hall just looked at him with a sort of twinkle in his eye and replied, he said, "Well," he said, "you see I have to be near a toilet." [laughter]

Sopka:

Like a very earthy gentleman.

Purcell:

He put him down [laughs].

Sopka:

I've been impressed with the longevity of the Harvard physics faculty.

Purcell:

Oh yes, my goodness.

Sopka:

And not only their physical, but their research longevity.

Purcell:

Yes. I remember a terrible crisis at that time when somebody — Hall had made his thermocouples out of a particular spool of manganin and advance wire, whatever combination he used, in the stockroom where they had, you know, spools of different — And he did that for the very good reason that you didn't want to recalibrate every time. You want to use the same alloy.

Sopka:

Yes.

Purcell:

And some graduate student had checked out that spool of wire from the stockroom and it was missing. Sparks were really flying for a while. I don't know whether he got it back or not. No, Hall was an odd fellow. See, in a certain sense he never understood the theory of the Hall Effect, because the Hall Effect was only really explained by quantum mechanics.

Sopka:

Yes. That's right. So it would come too late for him.

Purcell:

And he had a theory, he had developed a theory of his own which he called a dual theory of conduction. He had a little book on it. It probably should be still in the library — a dual theory of conduction.

Sopka:

Oh. I'll have to look that up.

Purcell:

Yeah. And I remember one time when, it must have been — I can date it almost, it must have been '38, '37 perhaps. When R. H. Fowler was here he gave a colloquium — you know, the great British theoretical physicist, R. H. Fowler. And Fowler was talking about something having to do with the electron theory of metals. I forget the exact topic. And Professor Hall was sitting in the front row where he always sat, and at the end of the talk, the question period, Professor Hall mumbled some question that he didn't, you know, showing his general lack of enthusiasm for what was going on, to which Fowler may have said the most graceful reply. He said, "Well Professor Hall," he said, "I should think you would welcome this theory because this is the first theory that explains the positive Hall coefficients." [laughs] Of course Saunders kept on teaching for many years after he left Harvard, out at Mount Holyoke.

Sopka:

Oh, I wasn't aware of that.

Purcell:

Oh yes. Saunders retired to Mount Holyoke.

Sopka:

I see.

Purcell:

And he and his wife built a house at South Hadley. I think maybe she had had some association there. I don't know, they'd had some association somehow with Mount Holyoke before, and Saunders taught there for many, many years.

Sopka:

Oh.

Purcell:

We had been out to visit them back in those days.

Sopka:

I believe Saunders lived until 1960 or '61.

Purcell:

Yes, and of course she just died a couple years ago.

Sopka:

I see.

Purcell:

And there he continued his violin work there, very seriously. So he counts in the longevity list too.

Sopka:

The nature of physics certainly has changed in terms of what things one does if one is a physicist within the Harvard Physics Department.

Purcell:

Well, yeah, the tools are of course different so that the experimental work one now uses much more powerful aids, so that every experiment is festooned with things we didn't have — online computers and vacuum pumps unlike the old ones, and stuff like that. But I don't see that, I don't feel that is any real change in kind.

Sopka:

No, but doesn't it appear that many of the things that would not be done within the physics department, for instance if one wanted to study violins or trumpets would that be done within a physics department? Where would you do that?

Purcell:

Well, it is being done in some physics departments. I mean, there aren't many people doing that, because I don't think there's all that much to do in that, but there are two or three people like Arthur Benade at Case, whose career has been the physics of instruments.

Sopka:

Oh yes. I remember his book.

Purcell:

And that's — it's not a thing we'd ever — In fact, Saunders didn't consider that he was doing, that it was — It was sort of his end of career hobby so to speak I think even when it began right here in the department. Although it was interesting at the time, because he was well in advance of others in looking at the acoustic spectrum environments to find that new technique, then new technique, to the problem. Well, I don't know. If you — the high-energy work of course has developed a very different style because the one that perhaps could have been predicted already in the '30s, because now the machines are so big that they can't be at the universities, and so people have to do their work by these enormous cooperations at the big laboratories. Whereas in the '30s and immediate postwar years, if you were a physics department you had your own cyclotron and I think physics came out of the Berkeley cyclotron or the Chicago cyclotron, you know, or the —

Sopka:

Did you have anything to do with the construction of the postwar cyclotron at Harvard?

Purcell:

No, not a thing.

Sopka:

I see.

Purcell:

By then I was completely absorbed in the nuclear magnetic resonance stuff. I was aware of the planning and heard people discuss it, but I hadn't —

Sopka:

You didn't have any responsibility.

Purcell:

I had no responsibility and no input to it at all.

Sopka:

I see.

Purcell:

Theoretical physics of course is — I suppose one might say there's a change in style there too, but only in the scope of the enormous number of people working and it —

Sopka:

The acceptance of the importance of the theoretical activity on the part of physicists seems to have changed over the years. David Webster got his degree in 1913, and for an experimental thesis it apparently was very difficult for him.

Purcell:

Yeah.

Sopka:

He told me that he was told that Harvard would never give a degree which was not based on some experimental work.

Purcell:

Right. Yeah, that's certainly changed here in this department. I guess Kemble was the first one to get such a degree, wasn't he?

Sopka:

Yes, but even Kemble's thesis was part experimental.

Purcell:

Uh-huh [affirmative].

Sopka:

And Van Vleck was the first quantum physics —

Purcell:

First purely theoretical.

Sopka:

— purely theoretical quantum physics.

Purcell:

Slater's thesis was experimental.

Sopka:

Experimental. Yes.

Purcell:

Yeah, yeah, right.

Sopka:

On the other hand I, in reading the annual reports, I found that by the end of the '50s the department was concerned that there were too many theoretical students in proportion —

Purcell:

Yes. Oh yes. No, that's right. Everybody had — Well, I think it seems to me that experimental physics has really gotten quite difficult in that if I were, you know, I started a young graduate student or young postdoc in experimental physics, it's a rather forbidding situation in that it's hard to think of anything you can do that doesn't involve so much of an investment in equipment that the only way to do it is to join an ongoing research operation. You can't just go off yourself and have an idea and go off and do it, because you can't afford to buy the stuff it would take to do it, and furthermore there are so many people hacking away at the business it's kind of hard to scratch up something that somebody isn't already doing.

Sopka:

Yes.

Purcell:

Whereas in my time as a graduate student, looking back on it, the difficulty in finding something to do that nobody was doing wasn't serious at all. I mean, almost any reasonably good idea would have a chance of being new. You know, it's so that the individual experimenter who is able to have an idea himself and go and try out that wonderful experience is pretty hard to have, you know.

Sopka:

Does the graduate student today who wants to be an experimental physicist include in his training shop techniques such as were done 50 years ago when graduate students had to learn how to blow glass and operate lathes, a standard part of their training?

Purcell:

Well, I don't know. I'm not a very good person to ask, because I haven't had experimental graduate students for a long time, but my impression is that they have to know some electronics.

Sopka:

I see.

Purcell:

I suppose that for many of them, for some of them here, certainly they have to get some shop experience or pick it up one way or another, but many of them don't. Glass blowing of course is totally obsolete.

Sopka:

Yes.

Purcell:

But I believe that for the typical present day experimental graduate student the thing that he's absolutely got to have is familiarity with modern electronics.

Sopka:

Including computer science?

Purcell:

Oh yes, including, yes, yes, include that in. In fact he's got to — well, what he needs to know is most accurately defined by the content of Paul Horowitz's course 123.

Sopka:

Oh, that's interesting to check out there.

Purcell:

That is exactly what, you know, if you had any one thing that a present experimentalist needs, it's that.

Sopka:

It's in that course.

Purcell:

Now there are all these other things that he has to have some lower on depending on his field, like high vacuum stuff. See, one of the changes between my period as an experimentalist and now is just in the degree of vacuum that you can get. People don't realize how much it's changed.

Purcell:

By five orders of magnitude. If you really want a good vacuum now.

Sopka:

Well, I wasn't aware of that.

Purcell:

Yeah. It used to be, you know, 10 to the minus 5. We'd get down to 10 to the minus 5 and the —

Sopka:

That's pretty good.

Purcell:

Yeah, yeah, yeah, it's pretty good. You know, the old pump would be going there clack-clack instead of glug-glug, and the ionization gauge would be reading 10 to the minus 5, well you were home free.

Sopka:

Sure.

Purcell:

In a vac'd out small system now they think nothing of 10 to the minus 10.

Sopka:

Oh. That's —

Purcell:

And they'll go 10 to the minus 7 in a great big accelerator, and the pumping is an entirely different business, those titanium pumps.

Sopka:

Oh. Yes.

Purcell:

No mercury around the place. A modern laboratory is distinguished by the total absence of mercury in any form. I guess a little bit into fluorescent lights, but otherwise —

Sopka:

And none rolling around the floor like they used to —

Purcell:

It's not rolling around the floor.

Sopka:

— warn us about.

Purcell:

And boiling in the pumps. God. And those traps we used to have to keep closed. Well that and of course then the whole rebirth in optics is one of the great things, on account of lasers and everything, that whole physics of, the experimental physics of light is just a totally new subject now, absolutely remarkable.

Sopka:

Oh yes. Would you be inclined to comment on the events surrounding your being awarded the Nobel Prize? You've mentioned the kind of work, your nuclear magnetic resonance research.

Purcell:

Well, I don't know. I think the more interesting comments are perhaps on the experiment itself and not the Nobel Prize.

Sopka:

They're in your lecture, your Nobel lecture presumably.

Purcell:

Yeah. But our relationship for instance to the Stanford work is sort of interesting, and I'd like to say a word about that just to have it on record.

Sopka:

That would be fine.

Purcell:

We began thinking about the experiments. I really had the idea for the experiment in what I think was August 1945, and enlisted the help of Henry Torrey and Bob Pound. Bob, because he was, I knew him, he was a good friend, very young then, but he was of all the people at the Radiation Lab, Bob was the sharpest on questions of signal-to-noise receivers and stuff, and Henry was mostly involved in that too. Henry Torrey moreover had been a graduate student at Columbia in the molecular beam, at that time the beam stuff, and knew that side of it, and furthermore was a very competent theorist. So we made a good team in that respect, and we couldn't have done it I think — See, every person was indispensable. We couldn't have done it without Bob, couldn't have done it without him. And we had no knowledge whatever of Bloch's idea and what he was working on at Stanford. Totally. Although Felix I think had been thinking about it maybe already before he left Cambridge to go back to Stanford, but at any rate I didn't even know him then. Knew who he was and seen him, but I — So the only thing that gave us pause in the course of working on the experiment, and it occurred after we'd I guess even after we'd tried it once and failed, that we learned that two or three years before Gorter at Leyden had had essentially the similar idea and tried it and it failed. We learned this because we got a photostat or something of the article which hadn't been coming over during the war, but somehow I came over and found it here and read it. It wasn't a photostat, you know, it was a what do you call it? Microfilm I guess. And some — I don't know who put us onto this or who heard this rumor, but we looked it up and sure enough Gorter had tried what ought to have worked to measure nuclear resonance, with crystal of — I forget what the crystal was at the moment. Well at any rate, we decided we knew why it failed, because it was the thing that we had already been taking into account and preparing for connected with the relaxation time. We thought he had saturated this stuff before the resonance, so we decided to go ahead and try again, and worked on it, and finally we made it work — all of this before we even knew what was going on at Stanford. And then we heard about that, and they, after they were working on it, and they finally got the resonance about a month or so later, Stanford, using Bloch's technique of the crossed coils. The first actual communication we had with that group was with Bill Hansen, a very interesting man who collaborated with Bloch on this experiment but who was very well known to the rest of us as one of the inventors at Klystron and had managed to spend some time at Radiation Lab lecturing on microwaves, and we knew it would be a tremendous expert and very fun fella, but Bill Hansen came east I guess it must have been perhaps March of '46 and we started explaining our mutual experiments to one another. You know, it was clear that they were basically getting the same result, but we had approached it from such a different direction that literally we talked with Bill for about a half an hour before we established where we could each see it from the other's point of view.

Sopka:

Well, that's very interesting. I wasn't aware of that.

Purcell:

Yeah. They had always thought about it, in a perfectly correct way, in terms of the precession of the expectation value of the moment so to speak, so they had just thought about it classically and the precessing moment, which was the basis for the two-coil method, one coil drives and the other coil picks it up. And we had thought of it in terms of inducing transitions between two energy levels purely quantum mechanically and thought of the absorption of energy from that, and we had detected it therefore not by looking at the precessing out of phase component but by detecting the absorption of energy to resonant cavity.

Sopka:

I see.

Purcell:

Now of course these are actually identical, of course identical, things, and in class if I were lecturing on it I would do it both ways, but we looked at it in such a different way that — Then among the very important things that happened here in '46, I ought to mention, was that Bloembergen arrived, and Bloembergen turned up one day, and it must have been — oh, it must have been, I don't know, again April, perhaps April '46, something like that, here turned up this young man who had just gotten his AB at Utrect and who came without any advance notice or introduction or anything. Nobody knew who he was; he just showed up here one morning and said he had gotten his Bachelor’s degree in Holland and he was interested in physics and wondered if there was any job for a laboratory assistant. And I talked with him for a while, and we had not thought we needed a laboratory assistant, and the thought hadn't occurred to us. We didn't have much money to pay anybody. But after I talked to him I decided it would be a good idea to. It was the smartest thing I ever did. So we hired Nico as laboratory assistant. He was practically fresh off the boat from Holland. You know, and within a couple of weeks he was writing, he was ordering things, building stuff, doing experiments, you know, really going.

Sopka:

Good.

Purcell:

And it was therefore the work that he did then the next year that was really the basis for all the work on nuclear relaxation. He went back to Leyden then and got his Ph.D. at Leyden but with a thesis that he did here. So in a very real sense my first graduate student got his Ph.D. at Leyden with Gorter, and amusingly, while he was there he got the crystal that Gorter had actually tried the experiment with and got the resonance in it.

Purcell:

So it really was saturated.

Sopka:

Oh. That makes a nice fulfilling of the original idea.

Purcell:

Yeah. It's really too bad for Gorter, because Gorter in a way almost you could say deserved to do the experiment first, because he was one of the first to point out the importance of nuclear magnetism and things like that, so that it was a case where Gorter — well, the only way to say it is Gorter was unlucky and we were lucky in that respect. There is no other way to describe it. Bloembergen then gave — we published then. Certainly the paper I had been associated with had more references than any other; it must be by far, an enormous margin. This was a paper that came to be known BPP, published in '46, which was our first big paper on the relaxation stuff, Bloembergen, Purcell, and Pound, and was really Nico's thesis sort of turned in — I wrote the paper, but while he was writing his thesis, and it was sort of published.

Sopka:

That's very nice to get that story down. We're coming to the end of this side of the tape.

Purcell:

Oh, alright.

Sopka:

So that I don't think that we want to start a new topic.

Purcell:

Have you ever seen the pictures that we have of those old people? Did I show you those? You know the stereo pictures we have? Do you know about those?

Sopka:

No. Maybe I can turn this off anyway now.