Oral History Transcript — Dr. Felix Bloch
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
See the catalog record for this interview and search for other interviews in our collection
Interview with Dr. Felix Bloch
Felix Bloch; December 15, 1981
ABSTRACT: Personal highlight in solid state physics and magnetism, 1928-1933. Reflections on Werner Heisenberg in Leipzig and Wolfgang Pauli in Zurich; early work on order parameter. Interview conducted as part of the International Project in the History of Solid State Physics.
Hoddeson:This is Lillian Hoddeson and Iím interviewing Professor Felix Bloch in his office at Stanford University on December 15. Since you lure already been interviewed in some detail by Kuhn in 1964 and Weiner in 1968, it will not be necessary to discuss your entire life starting from 1905 when you were born in Zurich, because the transcripts of those earlier interviews are available for qualified scholars at the Center for History of Physics. So weíll have the luxury of focusing on a few key issues relating to your work in solid state physics between 1928 and about 1933 or Ď34.
Hoddeson:Iíd like to actually begin approximately two years before the completion of your doctorate in Leipzig, with a point that you mentioned in your interview with Kuhn and I suppose also in your memoir, concerning Heitler and London.
Hoddeson:You said you got to know them in 1926 when they came to Zurich.
Bloch:Did you read my article on ďReminiscences of Quantum MechanicsĒ? I mention something there.
Hoddeson:You mentioned that you took walks with them.
Hoddeson:And that you discussed the covalent bond theory.
Bloch:I donít know that I discussed it with them. I knew they worked on that. We talked about all kinds of things, mostly about quantum mechanics. Maybe they spoke about that to me. I really donít remember. I couldnít say that I was engaged in that. I mean, I knew their work very well and it impressed me, as I said in my article which I sent to Mott, that in their model also there were (crosstalk)
Hoddeson:Yes, yes, I see, because this theme comes up again and again then in your work later.
Hoddeson:The electron topping from —
Bloch:Yes, I made a, of course, I made a whole point of, yes (crosstalk knocks out a lot of this...)
Hoddeson:OK, we wonít make too much about the impression.
Bloch:Well, of course, I knew them very well, and it was for me very important to talk to people who were much more advanced than I was.
Bloch:At that time. But I would not claim any credit on the theory of covalent bonds, no.
Hoddeson:OK. Now, in 1927, Debuye suggested you move on to Leipzig, were Heisenberg was, because all the people you might have wanted to work with in Zurich, Weil, Debuye, Schrodinger, were leaving just about this time, or had left.
Bloch:Well, actually, there was a slight mistake, Peierls told me afterwards that my memory lapsed. I think Weil had already accepted the post atÖ but he did not leave already in Ď27. He left a few years later. Peierls wrote to me because he was assistant of Pauli, after I was, during that time, he is still went to a lecture of Weilís, so that is not quite correct. Weil did not leave in Ď27 but it was known that he was going to leave, and he left shortly afterwards. Furthermore I would tot have worked with Weil because Iím not a mathematician.
Hoddeson:Why did you choose Heisenberg rather than say Sommerfeld?
Bloch:On Debuyeís recommendation. Debuye told me that Heisenberg was coming to Leipzig, and said, ďIf I were you, I would work with Heisenberg.Ē
Bloch:That was decisive. I trusted Debuyeís judgment in this.
Hoddeson:I see. All right, in your PHYSICS TODAY article you describe your first impressions in Leipzig very well, also in the Mott memoir, but I wonder if you could expand a little bit more and give us just a little bit more detail about the research environment in Leipzig? As I see it, the quantum theory of solids, which Iím particularly interested in, emerged in Munich, and then within a very few years was dispersed first to Leipzig, Zurich, then to many other countries.
Bloch:YesÖYes. And itís quite possible that Heisenberg, who of course also was in contact with Sommerfeld, got this idea that one should — or was interested in that because of Sommerfeldís work. Thatís very possible.
Hoddeson:Right. Right. Now, Bethe went into quite a bit of detail about Munich.
Hoddeson:He told us about the rooms and about the courses, for example, and Iím wondering whether you could perhaps tell us a little bit about Leipzig as you saw it at that time ? For example, the courses that were of interest. Heisenberg gave a solid state course.
Bloch:No, No, no, no, no. He did not give a solid state course. No, He gave the (crosstalk ) .... poh, sure. No, he gave the introductory course on — well, he gave a course on quantum mechanics. I remember I sat in on it. I did not attend particular lectures at that time. I mean, I worked for my thesis. But I did go to a lecture of his, and I remember I was very impressed about the way, how he introduced quantum mechanics, told us how he had introduced it, by deriving it from the Krammers-Heisenberg dispersion formula. You know, his way of getting at it was quite different from Schrodingerís or Debuyeís (DeBroglieís?) and I remember that very well, how very logically he showed us that the matrix formulation followed almost necessarily from the Krammers-Heisenberg dispersion formula. So that is one of my very vivid memories I have. But then of course it was a very good department. Wenzel ? was already there. You know that.
Hoddeson:He was there already.
Bloch:Oh yes. A year later he became Schrodingerís successor at Zurich, but for the first year I was there, he was also there. And Hundt was there, and Debuye was there and Heisenberg was there, so you see that was a collection of really first rate people. It was a large department.
Hoddeson:Now, were the seminars in the same style that one finds here in this country?
Bloch:Very informal. Very informal.
Hoddeson:Did graduate students participate?
Bloch:Well, of course, graduate students — that concept didnít exist in Europe. Everyone was a student. But, yes, there were a few. The number was very small, Iíd say about six people. And Heisenberg was always very nice and ordered cakes. So, he paid for the cakes, for the seminar. It was very informal but on a very high level.
Hoddeson:I see. You say there were about six people. Were these six people —
Bloch:These were students and assistants. I donít quite remember them. Now, for example, Teller, who came later, was not there, yet? Right. Weisacker was not there either yet. Weíre talking about the first years. Very small. I think I was the only student of Heisenberg. And this was an extremely, shall I say, releation of almost friendship. I mean, he and I would take our skis and go over weekend skiing on the southern border of Germany, and we went together to lunch and so forth.
Hoddeson:Did you talk about physics when you went skiing?
Bloch:Oh yes. Oh yes, sure. Well, you see, physics and everyday life were not separate in our lives. It was all the same.
Hoddeson:I see. And these seminars, did they meet on a regular basis once a week, or were they?
Bloch:Iím not quite sure how often. Yes, they were fairly regular. Whether it was once a week, or — I think so, yes. My memory isnít quite good. I think so, yes. Now, for example, Heisenberg wanted me, shortly after I came, he wanted me to report on something I had read on, oh, this five dimensional Geometry fellow. I donít know whether you know that. Oh, there was an attempt by Klein and other people to bring electricity and gravitation together, in a five dimensional geometry. It was interesting, at that time. Heisenberg wanted to know about it. I hadnít looked at that before, so I gave for example a seminar on that. It had nothing whatsoever to do either with Heisenberg or what I was doing. It was all just general information. A paper he didnít have time to read, so he said, ďWould you tell us about it?Ē
Hoddeson:It was a combination of seminar and journal club?
Bloch:You might call it that way, yes.
Hoddeson:But Heisenberg selecting the particular articles that?
Bloch:Yes. Yes. I think I told him that I had read these papers and he said, ďWhy donít you tell us about it at the seminar?Ē or something like that.
Hoddeson:I see, and these seminars would run an hour, hour and a half?
Bloch:Yes, about that. Then we sat together or we started to play ping pong. It was all very informal, very very nice.
Hoddeson:I see. Now, I did come across some notes of a course that Heisenberg at least wrote in note form. I donít know whether the course was ever held. In these papers that were collected by the Kuhn Project in the early sixties, I think it wasÖ Iím not sure. I donít have it written down.
Bloch:Well, thatís possible. I rather doubt that he gave it during that year that I was there.
Hoddeson:Not during that year. It looked as though it might have been dated Ď29, Ď30.
Bloch:Ď29, Ď30, yes, I wasnít there then.
Hoddeson:Only because Peierlsí processes are mentioned, but thereís nothing about bands or diamagnetism or anything like that, so that would date it in that year.
Bloch:I wish Iíd seen that. Do copies of that exist?
Hoddeson:Well, I made a very bad copy. Itís quite thick. I can send it to you.
Bloch:No, no, thatís too much.
Hoddeson:No, no, itís — a Xerox copy — Iíll send you a Xerox, well, I only have a Xerox copy.
Bloch:Yes, but thatís yours. And you could have one made? I mean, could I keep it?
Hoddeson:Iíll send one to you. No, thereís no problem with that. Theyíre not expensive.
Bloch:That would be very nice. I didnít know that, you see.
Hoddeson:Well, you might see it, and might have a comment to make, That would be of interest to us, so — OK, well, yes, the interaction between theorists and experimentalists in Leipzig at that time. Later Wilson talks in his memoir about Gooden coming to tell them about semiconductors, and I was wondering whether in general, there was some contact.
Bloch:Well, we knew them of course very well. I mean, for example, Debuyeís assistant, Sach, he was a friend of mine. We came together from Zurich, so I knew very well what he was doing. I knew he was working on these electrolytes and that kind of stuff that Debuye was interested in. I think so at least. Well, we knew what they were doing, but I do not recall experiments done in this particular field, at least at Leizpzig. Could be that I donít remember. Now that you mention a name it sounds familiar to me, but —
Hoddeson:There wasnít that much. What about the work that was being done in some of the other places, for example Greenhausen in Gottingen.
Bloch:Greenhausen in Gottingen. Well, of course, I think I probably met him before, but it was just — of course he wrote to me for example, when I got finally the key to the fifth — he said, ah, now that makes sense. So from time to time you know he wrote sort of interpolation formulae that contained both. I think I knew probably Owell and there was work done at Seamanís, andÖ and so on. I knew there was a lot of work going on in solid state physics, but my contacts were not terribly good.
Hoddeson:They werenít, on a personal level?
Bloch:Well, personal, yes, but not scientific particularly. Well, you know, there was a certain attitude among the theorists at that time, so to say, we donít really need the experimentalists, we know better.
Hoddeson:There was, really?
Bloch:Well, Iím making a little bit of a joke.
Bloch:The separation between theory and experiment was much greater in Europe than it is here.
Hoddeson:Than in America.
Bloch:Yes. Yes. I knew them, I knew them personally. But you didnít just go and ask an experimenter some curve, and then you say, aha, now I have to explain this curve. It was the other way around. You made your theory, and then you came to him, and if his experiments agreed with it, fine, if they didnít, you said he made a mistake. I mean, Iím a little facetious, youíll remember.
Hoddeson:Yes, Yes. That fits with what Iíve heard from other people. About the visitors passing through, I gather there were a great many.
Bloch:Yes. Yes. And I pointed that out to Peierls. Wait a minute, where did Peierls? Oh no, that was another article— yes, Peierls wrote an obituary for Heisenberg. Do you know that?
Bloch:He mentioned several visitors there, and I pointed out a few more to him. For example —
Hoddeson:— who are they now? I donít remember the obituary. You donít have to go back and look it up.
Bloch:Oh, there were an awful lot.
Hoddeson:Hausman? Slater, of course.
Bloch:Slater? Not in my time. People kept on dropping in. Now, wait a minute, are we still talking about Ď27, Ď28?
Bloch:During that time there were not so many yet.
Hoddeson:Not so many.
Bloch:No, No. It was later. When I was Heisenbergís assistant there were many more.
Hoddeson:Well, since weíre on visitors and we may get off on other subjects, letís talk about the visitors in general.
Bloch:Well, of course, a very important visitor for me was Hausman, because he had worked with Sommerfeld on the same problem.
Hoddeson:Right. Right. Right.
Bloch:But of course he made a perturbation calculation. You know.
Hoddeson:Yes, I know, also he —
Bloch:—but I think us basic idea was quite right. The interaction between electrons and sound waves, he treated very well.
Hoddeson:Right, and he also had the idea of the infinite conductivity.
Bloch:Well, about the periodic potential, that I donít know, and I donít think so. He just improved on Sommerfeldís theory by saying, OK, there are — thereís not just the Fermi gas but there are also sound waves, and the electrons scatter on them. Iím not too positive. One can look that up in the literature. I donít think that he realized that one can have a strong interaction of the ions, if the ions are addressed. I mean, he sort of took it for granted that the electrons somehow travel through a lattice, if it is uniform.
Hoddeson:I think he had the idea that the deviations in periodicity had something to do with the resistance.
Bloch:Of course. That is correct. That means, deviation of periodicity means sound waves.
Bloch:But he treated them I think more or less as a continuum, and, well, and I think — I would say, his is really an approximation. Itís an approximation of weak, weak coupling between the electrons and the ions. And within that framework, it was identical with mine, and actually, after I wrote my T to the fifth paper, he had trouble with it —
Hoddeson:Yes, I know.
Bloch:But he came and said, ďYes, thatís right, I got that too.Ē Thatís obvious.
Hoddeson:Well, he got a T to the third.
Bloch:Well, in the beginning, we both got it wrong.
Hoddeson:You both got very different errors he corrected and said (crosstalk) then you agreed—
Bloch:Yes, but the ó by the time I showed the correction, he said, ďThatís right,Ē and that didnít surprise me at all, because, I mean, the coefficients would not be the same, but his approach was quite the same as mine was. That the resistance was due to collisions with sound waves, or as you say, phonons nowadays.
Hoddeson:Let me just mention some of the people, other visitors. Just to find out whether you interacted with them, not necessarily in this year but perhaps later.
Bloch:Heller, of course. Well, Keller was not a visitor. Heller was also assistant in Leipzig. He was a student of Wundt We worked together for a year.
Hoddeson:Oh, he was assistant to Wundt?
Bloch:Yes. We corrected together.
Hoddeson:This is the year?
Bloch:That was, I think, Ď32. Ď30, Ď31. Yes.
Hoddeson:OK, Slater I guess just in Ď29.
Bloch:Well, I am not sure that I remember Slater. I mean, I knew him later, of course, here.
Hoddeson:Well, thereís a reference in your paper on the first ferromagnetism paper.
Bloch:Oh yes, the determinant formulation.
Hoddeson:He saw a preprint of his paper apparently that he shows you —
Bloch:Oh, that is quite possible — oh, he showed it to me? (crosstalk here)
Hoddeson:— in Leipzig —
Hoddeson:At least thereís that much of an interaction but maybe not a great deal —
Bloch:Yes — well, no, but you see, all these people stayed in Leipzig maybe a week or so, and during that time, we talked to each other. Now that you say it, itís probably true, that I was very grateful for that to Slater, because it was a very nice formulation.
Hoddeson:OK, what about Landau?
Bloch:Landau I met later. Well, let me think. I met Landau during the time that I was in Copenhagen. No, no, just a minute, let me correct that. Just a momentÖ I met Landau, well, Iím sure I met him the first time in Copenhagen. Now, whether that was during the time I was — it could have been during the time that I was Pauliís assistant and just visited Copenhagen. You know, we went there quite often. So Iím not quite sure. I know for sure that I met him some time between 1927 and Ď28 and Ď30. In Copenhagen. And then, I spent a year in Holland. That was in 1929-30, and he visited me in Holland, and there we were —
Bloch:In Haarlem, yes. And then we talked about all kinds — we were very good friends. At that point he told me that I must visit him in Russia, which I did later. And we talked, oh yes, we talked about physics, again like everything, went to museums, he liked paintings.
Hoddeson:You were really very close to him.
Bloch:Oh yes. Oh yes.
Hoddeson:Because thereís quite a lot of contact with Landau and the kinds of ideas that Landau was working on in the later papers we will get to.
Bloch:Yes. I wouldnít be at all surprised if as didnít come up casually, so, in our conversation, you know.
Hoddeson:What about Nordheim?
Bloch:Now, Nordheim, where did I meet Nordheim first? I think it was on a visit in Gottingen. I think he was Bornís assistant, if I remember correctly. Anyhow he was in Gottingen, and we went to Gottingen occasionally for seminar. I think it was even during the time that I was Heisenbergís student, that I went to Gottingen, where I met Max Born the first time, and Nordheim, Iím pretty sure.
Bloch:Frankl, which Frankl do you mean?
Bloch:The Russian. Oh, ja, I thought you were talking about Gottingen. Yes, I believe I met him later on my visit to Russia.
Bloch:Well, I told you, Landau invited me and I went to Russia.
Hoddeson:At that time 1932, Ď33?
Hoddeson:Ď31, OK. All right, Peierls I wonít ask you about because there was so much interaction along the way. Briant?
Bloch:Briant, yes, I, that was somewhat later. After I left Germany in 1933, I spent a few weeks in Paris.
Hoddeson:At the Institute.
Bloch:Right. In fact, I lived there with Lengema. Yes, I knew him very well, I knew his daughter very well there and then I stayed at their house. And Briant — well, Iím not quite sure if I had known — and then Briant, we went out and talked. That was in Paris.
Hoddeson:Because I noticed —
Bloch:— then I later met him when he was in his country, during the war.
Hoddeson:I donít quite know how to fit his work in.
Bloch:Oh, the Briant zones.
Hoddeson:Yes, thereís a lot of, it doesnít —
Bloch:Well, Iím sorry to say, my memoryís not too good on that. Do you remember when he wrote his paper on the zones?
Hoddeson:1930. Then he wrote his second book, French textbook, I canít remember if it was Ď30 or Ď31, but the German one which was much more widely used was 1931.
Bloch:Well, I rather think that I met him personally and again we became good friends, after that, I somehow do not associate Briant very much in my memory until 1933.
Hoddeson:Was he the kind of person who interacted much?
Bloch:Oh, very much. Very much so. Oh yes definitely. A very lively person.
Hoddeson:I see. Thereís not much information about Briand.
Bloch:Well, you know, he had a good deal of engineering background. In fact, he worked for the radio stations of France for a while, and he had a very very sound grasp on reality. You see, his knowledge of waves, in which he was a great expert, derived partly from his work with radio waves.
Bloch:He knew about that — until the war, when we worked together, in fact we published a paper on that.
Hoddeson:Thatís very interesting.
Bloch:Briand was a very very fine man to talk to.
Hoddeson:Somehow in his book, there are places, he references you all the time and talks about your thesis and your work in general. Peierls, he seems to have had a running argument within his footnotes.
Hoddeson:Briand with Peierls, not the other way around. But he doesnít like the processes, and he doesnít mention Peierlsí work on the band gaps.
Hoddeson:He may have mentioned it, but he doesnít —
Bloch:— I donít know about that —
Hoddeson:— give him as much credit —
Bloch:— well, I donít know, frankly, I donít know, it may be that maybe they didnít hit it together personally so well, maybe — well, I got along with Briand very well.
Hoddeson:I donít have good information on that. Itís just a feeling I had. Maybe it isnít even right.
Bloch:Well, it may of course have been that perhaps he knew me better personally and therefore gave me more credit. I donít know. I know Briand and I got along together very well, I mean, our thoughts running really parallel. Peierlsí way of thinking was perhaps more strange to Briand. Maybe it was. Actually I spoke French and Peierls didnít.
Hoddeson:Oh, that might have had something... well, letís look at the work now. The first problem that Heisenberg suggested to you, I guess it was in the fall of 1927, concerned ferromagnetism.
Hoddeson:That is, to compute the Weiss field of 1 to 1907.
Bloch:Well, that he already knew.
Hoddeson:He already knew, OK.
Bloch:He didnít work out the details, but he said, ďLook, there are these strong forces in the Weiss field, there must be something else, there must be an electric force, there must be exchange,Ē and then he just sort of sketched it to me briefly. Thatís it. He knew it.
Hoddeson:Did you feel that this was really opening up a whole new era in magnetism?
Bloch:Oh yes, of course — well, no, it is not such a great prospectus as you have now, but I said, ďOh ja.Ē You see, everything, all of a sudden people explain things. I said, ďAha, that is the explanation of ferromagnetism.Ē I wasnít terribly, I didnít think of ferromagnetism as the most important thing in the world. Neither did Heisenberg. It was a nice little byproduct.
Hoddeson:It was a way of cleaning up the problems that — quantum mechanicsÖ
Bloch:— well, yes, thatís right, and Heisenberg evidently felt that there are open things in solid state and the Weiss field was one. And he probably just thought about it, that must be it, and mentioned to me, ďThereís no point in my going into it,Ē although he didnít do it, his mathematics wasnít very good.
Hoddeson:Then you did something about it. But you didnít expect to go into it all necessarily yourself? at that time?
Bloch:Not at that point, no. I mean, I felt, Iím not going to compete with Heisenberg.
Hoddeson:Did he talk about the problem with you while he was finishing up the details?
Bloch:Well, I think he showed me the manuscript.
Hoddeson:But you didnít criticize it?
Bloch:Oh, well, no. I would not have done better at that time either. That came later.
Hoddeson:Your main work was your thesis, which, you discussed it at length in several articles.
Bloch:Oh yes. Oh yes. Oh yes. There I was in constant contact. Now, in one way Mott and Peierls are not fair, in their obituary. They say that Heisenberg suggested to me to study electrons in the periodic potential. Thatís not true.
Hoddeson:What is true?
Bloch:What is true— he said, ďLook, thereís all that work of Sommerfeld and Pauli on the free electrons, and I think it would be interesting to look into that.Ē I mean, it was sort of understood that electrons, theÖ is not a vacuum in which electrons simply run. That was about it.
Hoddeson:I see, and then you decided to just take the ions at that point, not to worry about the electron interactions with each other?
Bloch:First — well, as I wrote in my article, the first thing was, why is it, how come they say that electrons donít run free through a metal? And as I say, that was the — now, you ask, if it is periodic, they practically run through free. That was I think the main step. I mean, as I said, I said, ďThatís it,Ē because you see, somehow, people must have been waiting for that. I was not the only one who said it was absurd to think that electrons just run freely through a metal, and so, the moment I said, ďWell, of course the ions are there, but they can run anyway,Ē then most people felt I got so much applause after that, that I was quite surprised, you know, but it shows that people had been waiting for that. I was not the only one who thought that was crazy. And so.
Hoddeson:Sure. When did you learn about the work of Wismer and Rosenfeld?
Bloch:I think only (crosstalk)
Hoddeson:You mention it in a footnote.
Bloch:Yes, thatís right. I think, well, maybe somebody pointed out to me, it might even be Heisenberg, somebody else, that there was a paper on periodic potential before, then I read it. But as you just pointed out, they found that out but they didnít attempt to follow it further. You read that paper?
Hoddeson:Yes, some time ago, not recently.
Bloch:I read it 50 years ago so you know it better than I.
Hoddeson:When did you learn about Flokayís theorem?
Bloch:Oh, that I donít know.
Bloch:Oh yes, after.
Hoddeson:Much after, I see, so you, independently —
Bloch:But you know, Flokay, that came from celestial mechanics. There was a question of periodic perturbation of planetary motion. Well, it was, mathematically it was the same problem. I think his periodicity was entirely out in space.
Hoddeson:I havenít looked at his work.
Bloch:Well, I think it was in reference to celestial mechanics.
Hoddeson:If I have a chance it would be fun to look at that. Some time. See what he does do. You know, itís probably different from what people say he does.
Bloch:No, no, I think, frankly I donít quite know myself, exactly what Flokayís theorem is, whether it has this band structure, or whether it shows these modulated motions, I do not know.
Hoddeson:I donít either.
Bloch:But I know it has to do with the effect of a periodic potential (perturbation?).
Hoddeson:Did you have any idea at this time of other bands besides the first one?
Bloch:I think so. I think so.
Hoddeson:Bands in general?
Bloch:Well, yes, because you see, I started, I studied the opposite limit. I said, ďSuppose electrons are in first approximation simply bound to each state.Ē Then I said, ďBut then you must make a combination.Ē But then it was clear that since the atom has excited states, to each excited state there would belong a band, and this (crosstalk)
Hoddeson:— you felt that was clear to you at that time?
Bloch:Oh yes, sure.
Hoddeson:But it doesnít appear in —
Bloch:— no, no, no, I wasnít interested in higher bands.
Hoddeson:No, no, but you had the idea.
Bloch:Oh yes. Oh yes. There can be no doubt about it, because when you start from the isolated atom, then of course, I said something about the ground state band. It was clear that the upper states would be equally spread.
Hoddeson:Did you have any idea at that time that the idea of explaining the difference between a metal and an insulator?
Bloch:No — well, yes, but you see, there I made mistake. Itís even mentioned in my thesis. I thought that the difference is simply a question of binding. It was stupid of me, not to realize, at that point — Peierls said to me later, ďIt seems rather obvious that a filled shell must be...Ē I said, ďYes, but I missed it.Ē And Wilson pointed that out to me.
Hoddeson:Was that the first time you came across that idea, in 1931, when Wilson?
Bloch:I didnít come across that idea. No, I had the wrong idea, and then Wilson correctly described some of these Products. Stubborn, I said, ďI donít believe it.Ē Iíd been working quite wrong. I said, ďIn that case, a divalent metal would have to be an insulator, ďBut then he pointed out to me that other bands can cross in.
Bloch:You see, if it were a single band, then with two electrons per electron, would be filled. And then I said, ďHow can that be? There are metals who give off two electrons per atom, theyíre still metals, theyíre not insulators.Ē
Bloch:But then he was very smart, and said, ďWell, you have to think about it a little bit.Ē But then he came back and said, ďYes, but you see, the bands may cross.Ē
Hoddeson:What about the idea of gaps between the bands?
Bloch:Well, it is also quite obvious, when you start from the opposite limit. You see, the excited states are widely separated, and if they jump only occasionally, each is slightly splits on the [?] cross diagrams.
Hoddeson:So the fact was — that was obvious?
Bloch:You see, the fact — oh yes — well, obvious, look, you have to be careful. The fact that there were bands, several bands, and there were gaps was to me totally trivial. I mean, that, I said, so what? I understand that. But what I missed was a, that this makes a difference, the difference between an Emulator and a metal — incredibly stupid. And also, and then, how semiconductors, all that sort of thing, that by excitation, you can create electrons. I just didnít think of it. But as I say, the existence of bands and the gaps between the bands, that was, I wasnít moved by that, not at all.
Hoddeson:I see. Did you read Betheís thesis at that time, by the way, on the Davidson and Wermer experiment?
Bloch:I donít remember it. I probably did. I read pretty much everything at that time.
Hoddeson:Because I read that recently, I just brought it along, and he actually has the band gap in a peculiar way. He doesnít say, because what heís concerned with is electrons,
Bloch:— scattering —
Hoddeson:— scattering, yes, and he has regions Ė-
Bloch:Yes, where they cannot go in.
Hoddeson:Where they are totally reflected.
Bloch:Yes. Well, you know, this is of course something that already Avard knew, on the scattering of X-rays.
Hoddeson:It certainly is in here. He calls them regions.
Bloch:Does he quote Avardís theory? Dynamical theory of —
Hoddeson:— he has reference to Ė-
Bloch:— ja, ja, ja —
Hoddeson:— Avard —
Bloch:— yes, because, you see, that was of course like Haaston too. They considered the analogy between X-rays and electron waves. I mean, we know already something about rays going through — of course, mainly from the X-rays — and so therefore I think it is quite clear, he wasnít particularly proud of that either, because that is probably in Avardís paper. On the scattering of X-rays.
Hoddeson:I see. Iíll have to check that. But in a way, that corresponds to the gap.
Bloch:Ja, I wouldnít be surprised if that same thing appeared in Avardís paper.
Hoddeson:Avard is a good deal earlier. OK, Iím wondering whether the picture of the Fermi energy surface was apparent to you at that time.
Bloch:No, not particularly in those terms. I made my life immediately very simple, because then I took it as a sphere. I made a quadrilateral approximation on that. I realized that it is not a sphere, but it wasnít terribly important to me. I mean, all these things later by Schoenberg and so on, on Fermi surfaces, were not —
Hoddeson:— not of interest to you.
Hoddeson:But in a way, you have some expressions for the energy which then are plotted out in the SommerfeldóBethe article Ė-
Bloch:That may be the cosine dependence, the sum of cosines, yes. Yes, itís true, that gives you a kind of a Fermi state, but I wasnít interested in that. I wanted to calculate the resistance. It was too complicated for me, so I approximated it by the quadratic minimum. That then is a sphere.
Hoddeson:Right. Now, I gather you did talk to Peierls about the hole effect while he was working, on it, or...
Bloch:— probably. Probably.
Hoddeson:And there he also works with almost filled bands. The first bands.
Bloch:That is because he has these holes, the matrix hole effect? I thought that was ingenious, but I didnít think of that. You see, by that time, of course, in Peierlsí work on the hole effect, the idea of filled bands and holes in bands was already much more obvious.
Hoddeson:A question on the hole. Peierls doesnít mention it very clearly, but the idea is really in his hole effect paper —
Bloch:That holes have actually positive charges.
Hoddeson:Yes, although he doesnít come out and say that.
Hoddeson:You say that in an article in 1931 which I probably have here, in the — You do have a picture of holes here.
Bloch:Oh, that was a talk I have.
Hoddeson:This is a talk... I think itís here...
Bloch:Is that mine?
Hoddeson:This is yours. The question is, where is?
Bloch:Here, you have something Ė- no, no.
Hoddeson:But thatís not Ė- Iím looking for the hole. I was sure the hole was in here.
Bloch:Ja, well, I probably had that. That was much later, wasnít it?
Hoddeson:This is 1930-í31?
Bloch:Ja, well, I must have mentioned Peierlsí work there too. That was not my idea.
Hoddeson:OK, anyway, the physical picture of the hole is here, and also in a paper by Heisenberg.
Bloch:Yes. Now, look, this is an interesting thing Ė- because Heisenberg had for a while the idea, and I think there are two, that the protons Ė- not the positive electrons, but that the protons could be holes, in an electron sea. And I remember Heisenberg once gave a talk in Copenhagen on this idea. But —
Hoddeson:These are the holes in quantum electrodynamics.
Hoddeson:Not the holes in solids.
Bloch:Had nothing to do with Ė- well, it had to do with Ė but I think itís quite true that the idea that a hole could be a positive electron came already from Diracís hole theory. I mean, that was not a terribly original thought of Peierls. He probably didnít mention it because he said, ďThatís well known.Ē I think so.
Hoddeson:So you think there was a connection between the hole in quantum electrodynamics and the hole in semiconductors?
Bloch:I think so. I think so.
Hoddeson:In Heisenbergís mind certainly.
Hoddeson:Maybe, we donít know whether it was in anybody elseís mind.
Bloch:Well, now, I must confess, at this point, my chronology is a little bit mixed up.
Hoddeson:Well, the hole appears in Diracís paper first, clearly in 1931. I think so, and the analogy —
Bloch:Oh, Peierlsí was before that.
Bloch:Ah, then it may even be that he had learned from Peierls.
Hoddeson:If he read that. But Peierls doesnít describe the hole very clearly. He has a mathematical equation.
Bloch:Ja, well, he showed that you get the negative hole effect, this is an electron missing. Thatís all.
Hoddeson:Yes. But he doesnít talk about it. He doesnít use the word. The word ďholeĒÖHe just talks about it —
Bloch:Ja, well, that may very well be.
Hoddeson:It doesnít seem to be clearly described until 1931 in, through Ė-
Bloch:— probably through (crosstalk)
Hoddeson:— Heisenberg Ė-
Bloch:Now, then Dirac came. You see, I think this is rather a leapfrog thing. Heisenberg or Peierls showed the negative hole effect, the presence of holes, and then Dirac must have known something about solid state physics, — well, he also thought it was Ė- he also thought at first they were protons. Dirac.
Hoddeson:Dirac thought, in quantum electrodynamics, that they were protons.
Bloch:Yes. But it was clear, these were positrons. And then by that time, the idea that a hole in a filled sea acts like a positive charge was known, and that probably, Peierls realized that his hole effect could be explained by simply assuming that they were electrons with positive charge. I think so. Iím guessing here. Iím sorry, I —
Hoddeson:It would be fun to work.
Bloch:Iím not a reliable historical source on that.
Hoddeson:It would be fun to work that out.
Bloch:There was so much interplay, you know, between all the physicists at that time, that as soon as somebody had an idea, another one took it up and put it in a different form, used it somewhere else, and so forth.
Hoddeson:Right. Oh, I had another question about your thesis: the application of group theory was only an elegant way of writing it up?
Bloch:Yes. Yes, I called it —
Hoddeson:You found the Bloch waves using 48 analysis (crosstalk)
Bloch:Yes, sure, sure, sure. I mean, as I say, I followed the fashion at that time. Maybe it is, it probably is more general there. I think itís the only time in my life that I made use of group theory. And it wasnít even necessary.
Hoddeson:OK, letís go on to Zurich, where you spent the next year, 1928 —
Bloch:Pauli, no, Iím sorry, Heisenberg was so impressed by my group theory thing that he said, ďOh, you have a way of determining group characters.Ē I just barely knew what group characters are! So he looked at it and said, ďOh, thatís wonderful, thatís a new method in group theory.Ē Of course it wasnít new at all, groups are very well known.
Hoddeson:I see, Heisenberg was not up on group theory?
Bloch:Well, Heisenberg, he took the same casual attitude as I took, on group theory, it might be used once in a while.
Hoddeson:Did Slaterís method really play this role that itís reputed to have played, in slaying —
Bloch:— ja, well, you see, — oh, I see. Slater. That was group very well afterwards.
Hoddeson:Afterwards, well, thatís what — (crosstalk all in here)
Bloch:— slay — yes, I think, we all were very relieved that one had a much more familiar way of expressing the content, than all this general highbrow group theoretical arguments. Well, Ehrenfest put it very well, he talked about group pest. But he was wrong and so was I. We underestimated the power of group theory at that time.
Hoddeson:You, in your Kuhn interview you described going from Leipzig to Zurich as going from optimism to pessimism. And mentioned that this partly had to do with the people involved, and the politics.
Bloch:— well, and the politics.
Hoddeson:Well, it was mainly Pauli?
Bloch:Well, you see, the difference between optimism and pessimism is the difference between Heisenberg and Pauli. Heisenberg was a very optimistic man and Pauli was a very pessimistic man, by nature. Thatís why Pauli was so effective as a critic. Whereas Heisenbergís critical abilities were not very great. Although Pauli made mistakes. I mean, he fell for my wrong theory of superconductivity: there, he was not critical where he should have been.
Hoddeson:In your piece for the Mott volume, you portray Heisenbergís feeling about solid stateÖ You describe in the Mott volume article Heisenbergís feeling about solid state in 1928 as ďa field to which quantum mechanics could fruitfully be applied.Ē
Hoddeson:Now, was this also true of Pauli?
Bloch:Pauli— Weisskopf told, Pauli said to him once, ďI donít like solid state physics, although I started it.Ē
Hoddeson:Yes. And yet he always had his hand in somehow, though, in some way.
Bloch:Well, yes, his application of Fermi statistics was —
Hoddeson:— was absolutely — the first —
Bloch:— very important, and then he —
Hoddeson:And then he was —
Bloch:Thatís what he meant when he said he started it.
Hoddeson:Right, but he was always interested in superconductivity.
Bloch:Well, I think Iíve described it properly. He just felt, letís get done with this, letís explain it all and then go to more important things.
Bloch:Yes, he was interested in it. He was interested in it. Much much later, for example, this was not at Zurich, I was already in America, he pointed out to me a paper by which he said was important, and he was right, it was.
Bloch:Well, I mean, Pauli, after all, he was a physicist, and as much he could not entirely ignore interest in problems in solid state physics, but he didnít really have his heart it.
Hoddeson:You also say in the Mott volume that you started to work on superconductivity before you went to Pauli.
Bloch:Well, thatís saying a little bit too much — I certainly thought about it, yes.
Hoddeson:And how far had you gotten?
Bloch:Not beyond the first idea, to say, that must be a minimum in the energy.
Hoddeson:— and that the electron —
Bloch:You see, that was not a trivial thought, because, one might have said, why do currents persist? Maybe it has some strong selection rules, that, and then I said, that cannot be, because the superconductors, theyíre not even particularly pure, and all that. So therefore, there must be insensitive toÖ and that can only be anÖ and that was the only idea I had. I seem to remember that Landau told me something once similar, probably, some time — I said, ďThat must be it.Ē
Hoddeson:This idea that you and Landau had about the same time and independently, I guess, of the energy minimum —
Bloch:Iím not sure but I do not think that Landau ever published anything about it. Neither did I. We never published.
Hoddeson:No, I learned this front the Sommerfeld Bethe, where he discusses this idea that you and Landau had. That is not so bad.
Bloch:No, itís not bad at all, itís — proves, in fact, only how difficult it is to get that, we didnít realize. But, well, I think, I probably told it to Bethe. I have a feeling that it came up in talks with Landau. Either I mentioned it to him or vice versa.
Hoddeson:But, when it might have come up?
Bloch:Probably in Holland. Probably in Holland. No, no, excuse me, no, before that. Before that. Oh, no, wait a minute, it might have been quite different. It might have been that I had this idea, but heard from Landau only when I met him, which was later, that he had the idea too, because I had this idea before I went to Pauli. I remember. In fact, maybe I mentioned it to Pauli and he said, ďYes, OK, go ahead and finish that work.
Hoddeson:Do you have old notebooks in which you might have scribbled something?
Hoddeson:You donít keep those things. Too bad. One could trace things like that. Well, itís not always possible.
Bloch:See, I didnít predict at that time that this would be the real later.
Hoddeson:OK, thatís very interesting. At what time did you state the theorem that Laundau quotes you, that all theories of super conductivity can be disproved?
Bloch:I donít know. Itís one of those jokes, you know. Laundauís formulation? That might well be.
Hoddeson:In London. In Londonís book.
Bloch:Oh, I see.
Hoddeson:I was just wondering whether that dated back to the time when you were with Pauli, or?
Bloch:Yes, probably after that — when I realized why I failed, then I made that general statement.
Hoddeson:Were you up on other peopleís theories of superconductivity?
Bloch:Well, not up, but I know that Heisenberg for example tried it at one point. He had some idea of condensation in space, as he called it. Yes. He may not have published on it. I think, the way he put it, his theory would have failed for the same reason for which mine failed. You see, in the long range, all of us, that we missed. That was the point.
Hoddeson:What about other people working on it as well, Frankl, Kronig, Elssasser?
Bloch:Yes, that is possible. I donít remember much about that. I mean, of course, superconductivity was a problem to be thought about. Well, it was common knowledge.
Hoddeson:You wrote your 1939 paper on, ďThe susceptibility and resistance of metals in magnetic fieldsĒ while you were in Zurich?
Bloch:In Zurich, yes.
Hoddeson:Yes, and you thank Pauli for stimulating this work, and for clarifying remarks.
Hoddeson:I was wondering how he felt about this problem. Here you were correcting him, in a way.
Bloch:Well, he felt good about that, because it was a sort of extrapolation of his own work on the paramagnetism ferromagnetism.
Bloch:He liked that. You know, one always likes oneís children to do it, something else —
Hoddeson:The other paper you wrote in this period was on the ferromagnetism possibility of paramagnetism by conduction of electrons.
Bloch:Ja, and, well, — right — when electrons are not low? cost? current?
Hoddeson:You say this is an outgrowth of Pauliís paramagnetism paper, and I didnít understand that.
Bloch:Well, I think it was — again, Iím not too sure. It was insofar as one talks about magnetic properties of conduction of electrons, I mean, insofar, it is of course an outgrowth. I mean, Pauli was the one who pointed out in his paper, on paramagnetism, I mean, he was the properties of conduction electrons. So that, to that extent it was. In the meantime, Heisenberg had his paper on ferromagnetism, so then there was the question, could one have only paramagnetism, with free electrons, or couldnít one also have ferromagnetism? I think in that sense, it grew out of it. I mean, it was rather obvious, considering Pauliís paper and Heisenbergís paper, this problem was not too far fetched.
Hoddeson:Now, a key achievement in this paper is your, the first calculation of the interaction energy of the electron gas.
Bloch:Yes, I think thatís probably true.
Hoddeson:You have a calculation, and, itís one level less than Wignerís later calculation in 1934.
Hoddeson:In fact, youíre quoted in a footnote in the Wigner paper. He cites you for this. So this is, I didnít realize, this is where that was done first.
Bloch:Yes. But again I had no such general ideas in mind. I mean, itís — the simple question was, after all, is it necessary for ferromagnetism, to think that the electrons are localized like Heisenberg did? And then I realized, no.
Hoddeson:They have to be very far apart.
Bloch:Yes, well, I gave some conditions on it.
Bloch:Whether it would or whether it would not (crosstalk)
Hoddeson:— yes, yes, yes. Conditions, I guess itís this...
Bloch:Iíd like to talk more. You remind me about all kinds of things.
Hoddeson:Well, tell me. Itís OK. We donít have to stick to my questions necessarily. They may not be the best ones.
Bloch:No. Go ahead.
Hoddeson:But then, once you find you get this result, then itís clear that the Heitler-London method is probably better, isnít that true?
Bloch:I donít know that I particularly thought that way. Did I say that?
Bloch:Oh, no, no, no.
Hoddeson:That was a comment I took from, I guess Slater mentioned that in his autobiography. You know, I guess itís this paper in which you thank Slater somewhere for the —
Bloch:— I donít think anybody read my papers as carefully as you do.
Hoddeson:Yes, here it is. Well, this doesnít necessarily mean you talked to Slater about it. On page 568, you thank him for a preprint.
Bloch:Well, that was later. That was not a lapse of [?] research. This paper was written in Zurich.
Hoddeson:This paper was written in Zurich?
Bloch:Yes, and I donít think I — it might well be that I met Slater only in Zurich. It could well be.
Hoddeson:I see. Could be, yes. In fact I think Slater did mention that it was in Zurich, so, I was wrong before. But then in discussing this paper in his book, Slater goes on to say that, then he got a preprint of your paper, and while he was in Leipzig after that — he must have stopped in Zurich and then gone to Leipzig —
Bloch:— while I was at Zurich, he was at Leipzig, that could well be, yes.
Hoddeson:And he was at that time exploring the relationship of the Heitler-London method to the Hundt-Mulligan method, and he noticed that your paper, your approach and Heisenbergís to ferromagnetism —
Bloch:— corresponded to —
Hoddeson:— corresponded to those two approaches. And then he went on with this idea, and did more on ferromagnetism.
Hoddeson:Stimulated in that way. Have you studied Franklís theory of ferromagnetism?
Bloch:Apparently not, because really at this point it doesnít ring a bell at all. I may have forgotten.
Hoddeson:Letís move on to the Netherlands, for 1929 to Ď30. You were there on a Lorenz Foundation Fellowship.
Bloch:Right, in Utrecht. And later I had something else, kind of an assistantship in Haarlem.
Hoddeson:I see, and how was this arranged, with — the fellowship, the assistantship?
Bloch:I donít remember. While I was still in Zurich, I think. I donít remember whether I wrote to Krammers, could I come to you?, or they wrote to me, will you come to me? I had also an offer as an assistant to Max Born at that time, and Pauli said, ďDonít go.Ē ďHeís a mathematician.Ē
Hoddeson:I see. I gather you got along very well with Krammers.
Bloch:Oh yes, very well.
Hoddeson:Where was the meeting of minds?
Bloch:First of all, I think I met him before already, and then, well, he was just very pleased that I would come, on a fellowship, and decide to spend it in Utrecht.
Hoddeson:I see. Now, what was he working on then, at that time?
Bloch:Oh, all kinds of things. Among other things, on the comintization (?) of the spinning top. I know he had a student by the name of Ickmann. He was not particularly working on these things during this time. He was just a wonderful person. He was an excellent musician. He played cello very well, talked about poetry, and of course physics. Again, it was the same thing.
Hoddeson:It was here that you found this error in your thesis and solved the problems.
Bloch:Yes. But that was not because of Krammers.
Hoddeson:He just happened to find it? Or no?
Bloch:Well, you see, this was a great time for me, because I had no obligation. All I had to do was think and so on. I thought, and I realized there was something fishy in my thesis. But I realized simply that the homogenous equation has a solution, if oneÖand then I realized, how one has to be careful.
Hoddeson:Then you had your big paper on the Heisenberg theory at low temperatures.
Bloch:Thatís right. It was a spin base.
Hoddeson:Yes, yes, and —
Bloch:Well, that was another thing. Then I said, well, now, let me think of it Ė- Heisenberg, thatís very complicated, maybe at low temperatures Ė- well, the first important thing was, to realize that the absolute minimum of the state is, all spins parallel. Thereís only one state.
Bloch:Then I said, well, OK, thatís rigorous. Now, suppose we flip one spin around, or two, or three. Then I didnít go too far, because then it became dangerous.
Hoddeson:Were you surprised when you found the spin wages? You donít call them spin waves yet in this paper.
Bloch:No. No, no.
Hoddeson:But very soon after that, you do call them spin waves.
Hoddeson:I guess it was at a talk you gave in Leipzig.
Bloch:Yes. I didnít call them spin waves till then?
Hoddeson:Maybe I missed it.
Bloch:I thought so, IÖ
Hoddeson:Well, maybe you do.
Bloch:I think so, spin waves, didnít I say that already in the —
Hoddeson:I think this is where it appears first in the paper.
Bloch:That is this paper, yes. Well, I donít know. I mean, that they were spin waves is obvious. Whether I used the word or not.
Hoddeson:Is there a story behind how you found the spin waves?
Bloch:No, not particularly. You see, of course, in a way it was very much the same thing that I was familiar with from the hopping of electrons in a metal. The way I started my thesis. I said, well, if electrons can hop, spins can also hop.
Hoddeson:I see. But thatís (crosstalk) Ė a somewhat different idea.
Bloch:Well, it was related. It was related. First of all, I was not impressed by Heisenbergís mathematics, though the fundamental idea was clearly right. And then I said, maybe one can do something more rigorous. And then I said, well, if at all, maybe one can do it at low temperatures. And I really donít know more about that. Thatís why I wrote the Ė wrote to Mott and said, ďI donít think I can write a great deal about ferromagnetism, because I donít really, I mean, it was just all kinds -Ė you see, itís this way, yes. When I came to Krammers, I had before been Pauliís assistant and so forth, and then also there was a time when I was quite free. I could do what I want. And then, you sort of pick up old things which you have in the back of your mind, where you had doubts before, that maybe I can do something — just pick them out. One was my doubts about the integral equation. And the other was that I felt, well, yes, Heisenberg is basically right, but his mathematics isnít very reliable. And then, the only way to do anything more rigorous to me, seemed to me, at low temperatures, and thatís true. If you had been able to go further, I would even have written a paper on phase conditions.
Hoddeson:Well, you do almost, a few years later. I guess you discussed this with Krammers?
Bloch:Oh yes, of course.
Hoddeson:Was he happy?
Bloch:Yes, indeed. Iím sure I wrote something about it. Donít I have at the end?
Hoddeson:Yes, you thank him at the end.
Bloch:Oh sure. Oh yes, of course. Oh yes. I told Krammers.
Hoddeson:Later Krammers wrote a paper with Heller in 1934, attempting to develop the spin wave theory on a phenomenological basis.
Bloch:Who did that?
Bloch:Ja, well, but that was not the super-exchange. That was something else I talked with him about later.
Hoddeson:I see. You make a point about dropping the spin orbit coupling here. A big point of it, and I was wondering whether that was something that was particularly in the air at the time. The fact that you bother to make this big point about dropping it.
Bloch:I donít remember that. That was of course much more important later, when I talked about the main walls, you know, Well, —
Hoddeson:OK, this you did in Utrecht, and then you gave a talk in Leipzig, but while you were in Holland —
Bloch:That was a conference, I believe, yes. I went to Leipzig for a conference.
Hoddeson:A conference organized by Debuye.
Bloch:Right. Right. Right.
Hoddeson:Leipzig, ? I gather this was the third one, so it sounds as if there was a regular series of conferences.
Hoddeson:Organized by Debuye, is that correct?
Bloch:Yes, that is correct.
Hoddeson:And there was one in 1930?
Hoddeson:Mott was there?
Bloch:Yes, I think that was the only one I was — Yes, this was on solid state. Mott and Mark, oh yes.
Hoddeson:Your paper, I just have a little more Xeroxed on your paper.
Bloch:I donít think that was particularly new. That was more a summary.
Hoddeson:Yes, it was a summary, but here you actually give the physical picture of the spin wave, and you call them spin waves.
Hoddeson:In this paper.
Bloch:Ja, ja, because you see, that was sort of a more popular talk, so —
Hoddeson:And you talk a lot about this interaction between the electrons, the need to take these into account.
Bloch:Yes. Well, in a general talk, one tries to line out the essential problems.
Hoddeson:Were other people worrying about the electron-electron interaction betides you? I mean, itís not very much in the literature, that I see.
Bloch:But I spoke about, not in context with superconductivity, it was ferromagnetism.
Hoddeson:Yes, but later when you wrote your review, you say, the big review —
Bloch:Oh, the book article.
Hoddeson:Yes, that, at the very end, everything looks like itís going to be solved, except superconductivity and the transition to ferromagnetism, and you say, well, but we still havenít taken the electron-electron interaction into account, and maybe the —
Bloch:— thatís right. Well, at that time, I had no doubt realized already that for superconductivity, you would have to take it into account.
Hoddeson:Do you know when you realized that it would have to be taken into account for superconductivity? Was that quite early or did that come out of your work with ferromagnetism? You started on superconductivity before ferromagnetism, right?
Bloch:Yes. Yes, yes, well, started, yes, sure. Thatís right. Iím sorry, what is your question?
Hoddeson:The question is, when you realized that the electron-electron interaction would be important for superconductivity?
Bloch:Well, as I say, I think I had that idea already sort of in my mind when I came to Pauli. It must have been at Leipzig. Because, well, I mean, as I told you, when Pauli said, or maybe I mentioned to Pauli something and he said, ďYes, yes — yes, go ahead and experiment on superconductivity, ďIím sure that the electron-electron, because that was the main thing that was missing in my paper. And then also, as I said, here, there was this analogy between ferromagnetism, between kurie temperature andÖ temperature. I had that feeling, there must be something in common, or between ferromagnetism and persistent currents, and both persisted. She magnetism persisted and the currents persisted. So how could that be? And I felt that must be a common cause.
Hoddeson:Then you went back to Leipzig, 1930-31.
Hoddeson:As Heisenbergís assistant.
Hoddeson:And then you, in the current review, you say at this point you wrote a long and learned paper on ferromagnetism.
Bloch:Yes, sure, that was for
Hoddeson:Now, which one is that?
Bloch:Of course, this was still, theÖ was coming.
Hoddeson:Thatís theÖ but thatís later. (crosstalk)
Bloch:No, it wasnít later, I worked on it in the summer, in the summer of Ď31. Yes. No, you see, as a matter of fact, Heisenberg said to me I shouldÖ myself, or become a — you know the procedure — and so I wrote something for them. I said, ďWill that do?Ē and he said, ďYes.Ē But then it was probably published later. I went to Copenhagen then, you see.
Hoddeson:Right. It was published in Ď32.
Bloch:And my former — in Ď32, right. It was written in Ď31.
Hoddeson:Yes, you submitted it in September, Ď31.
Bloch:There you are, see?
Hoddeson:Right. You also wrote a paper with G Ė-
Bloch:Gentia, oh yes. Yes.
Hoddeson:No, who is he?
Bloch:Well, heís an Italian. His father was a Senator, very important man in Italy, and he spent a year in Leipzig, and he was a nice fellow, but helpless and didnít quite know what to do, and also again I spoke a little Italian, so took him under my protection, and then we worked together on this.
Hoddeson:On this :
Bloch:Right. That was of course very important, because that came in then here. That was, I think, ja, right.
Hoddeson:OK. I think we should move on to the big paper because weíre running out of —
Bloch:By the way, there was a man by the name of Becker. You know Becker. And we had discussions on that quite frequently, on ferromagnetism, and this question of in ferromagnets, I think he made me pay attention to that, and in Heisenbergís spin, it would be entirely isotropic, and then we said, that must be spin orbit interaction, and I think that stimulated then my paper with Gentia. But that paper played a very important role. We discussed it a great deal.
Hoddeson:I see. Letís move on to this paper —
Bloch:We had meetings. You know, Eckles in Berlin and we were in Leipzig, and we met in a little village in between, just to talk about ferromagnetism. Heisner was there too.
Hoddeson:In that way you learned about the experiments by Stictus and Tonks?
Bloch:Probably. Very probably. Yes.
Hoddeson:They seem to be the takeoff place for this.
Bloch:Ja, ja, right. Problem for the walls.
Hoddeson:For the walls?
Bloch:Yes, yes. Yes, thatís very possible. I think thatís probably true.
Hoddeson:You would have been in touch with?
Bloch:Ja, ja, Becker was, wasnít thatÖHe was much more in contact with industrial research than I was. And I think he pointed that out to me. Iím sure. Yes.
Hoddeson:They use a very elegant Dirac transformation theory. In this paper, to get your equation for the spin wave function. There are references to Dirac. Hereís one, and well, itís quite Ė- (crosstalk) quite an elegant, you know —
Bloch:Well, you see.
Hoddeson:It is a very learned paper.
Bloch:Well, one had to, sort of. When you wanted to become a teacher at the university, you had to show that you knew something. So, thatís probably why I put that in.
Hoddeson:I see. Let me just quickly ask a few of the things that struck me.
Bloch:You know, there are all kinds of funny things also about the partition function, sort of little jokes I played with then did you look this up? This is rather cute.
Hoddeson:Iíll have to look more carefully.
Bloch:That is solution, that one gets the free energy fromÖ equation, but itís not — it sort of replaces theÖ of temperature.
Hoddeson:(to visitor) This is a good time for you to show up. Weíre just on the last paper, the one that you read. (crosstalk) He read this paper more carefully than I did.
Man:I was very impressed by that paper.
Bloch:Which one? That one?
Man:This one. There are things in this paper that have become ordinary knowledge.
Bloch:I told your wife that in order to become aÖ in Germany, you have to write a schritft something, you have to display what a scholar you were, so, everything I knew came into there. You know, this idea, the connection between the partition function, that was a little joke in this paper. How your differential equation — the free energy? Iím proud of that. You see, the idea was very simple. In a Schrodinger equation, everything depends equally, i.e., t. And in the free energy itís equal, that e minus e over k t, so that if you place i e by i times i over k —
Man:We know the trick, yes.
Bloch:You know that? I point it out here.
Man:I see. That is tremendously important in the whole.
Hoddeson:Hereís the question you had about the Wigner —
Man:The representation —
Bloch:— let me just get this little joke here. Here, (German)
Hoddeson:This is page 305.
Bloch:It had very little to do with it. See, here. Itís just one of those things. You had another question?
Man:Here, in this — representation —
Bloch:Oh, ja, ja, ja.
Man:Had you invented that?
Bloch:Yes, I think I did.
Man:I see, because that is known as Wignerís trick, I think.
Bloch:Maybe so, I guess so. Wigner probably saw it in a much more deep connection. You see, I was well prepared to write this paper, because without publishing anything — I had been in the hospital, I was with a broken leg — the summer before — and during that time I just, in order to get not too much bored, thought about all kinds of ways of representation, also polynomials and so forth, so I had that all in the back of my mind when I wrote that paper.
Hoddeson:On page 323 thereís a — equation that is almost the Ginsberg-Landau.
Bloch:Oh yes, yes, yes. I mean, thatís another thing. He was a little angry, Landau. He never gave me any credit for that. That was the first I said, letís use the psi? as a continuous function. I was really annoyed about Landau, because, the first time I told about it was on my visit to Russia, and there — that was in Kharkov and Tom was there, Landau was there, and I told them —
Hoddeson:This was in Ď34?
Bloch:Ď31. Ď31. And Landau was there. Well, he published that, so — thatís true. I mean, this is the first time, parameter angles.
Hoddeson:You certainly talked to Landau, because on page 321 you have a long footnote about Landau pointing, you must have been in contact with Landau in this period.
Hoddeson:You contact — that the spin orbit coupling was more important than the dipole —
Bloch:Yes. I donít how that, whether he told me that verbally or whether he wrote to me. I would have to think about that a little more. Yes.
Hoddeson:Well, then Ė-
Bloch:— as I told you, Landau and I knew each other very well, and in many ways our thoughts were in parallel, not entirely without — I mean, there was a good deal of coupling.
Hoddeson:You have the calculation for the wall in here and everything. (crosstalk) Ö You were in Leipzig. I think we may write to you with more questions about this paper.
Bloch:Well, I hope I can answer them. Well, actually, you know, I would almost prefer, I donít like very much to write.
Hoddeson:That or another visit?
Bloch:Is it that much.
Hoddeson:I donít know.
Bloch:If we could finish it today, Iíd just as well.
Hoddeson:OK, let me go on to some other.
Bloch:I have another half hour. Unless you are?
Hoddeson:I donít know, what time do you? Youíre not constrained either. OK, then, let me go back and see — this is the main wall — How was this paper received by other people who were interested in ferromagnetism? Did you get a lot of feedback immediately? Do you remember any?
Bloch:Oh, I think so.
Hoddeson:It was really quite a — you donít remember, OK. I would like to go back to, now weíre up to, weíre in Leipzig in the spring of 1931.
Bloch:In the spring of 1931, in the spring of Ď31? I came already in the fall of Ď30. I didnít do much in the winter but in the summer I got busy, yes.
Hoddeson:Yes. Now, during this period, Wilson was there. And this was —
Bloch:Was it that late?
Hoddeson:Well, this is the first of his papers. It was sent in June, 1931, so he must have —
Bloch:Ah, I was already Pauliís assistant. I didnít realize it was so late. Yes. Yes.
Hoddeson:You were in Leipzig at this time.
Bloch:Yes, I was Pauliís assistant.
Hoddeson:He gave his two colloquia on semiconductors and the electron theory in general.
Bloch:Iím sure I was there. Yes.
Hoddeson:I was wondering if you remembered anything other than that argument you had.
Bloch:Well, you see, I had pretty much forgotten about it, until somebody from England wrote to me and he had spoken to Wilson, and then the whole thing came back to me, and then I read my old thesis again and all of a sudden I realized — but at the time I wasnít very clear about it. That was much much later, really. I mean, I thought that Wilson had had this idea right away, but apparently it was not until Ď31.
Hoddeson:No, itís not, itís quite late Ď31.
Bloch:It seems so obvious now that people thought — well, anyway.
Hoddeson:So this idea of metals as solids with unfilled bands, insulators with —
Bloch:That was Wilsonís.
Hoddeson:Not until 1931?
Hoddeson:And the credit is really to Wilson.
Man:Did you discuss Betheís thesis?
Hoddeson:Yes, and Bloch suggested that that diagram is probably in Avardís work already.
Bloch:Yes, Avard had in fact a book on the dynamic of theory of X-ray scattering, this same problem, what happens if a wave hits a lattice, here, and comes back, and I think, Bethe alsoÖís paper.
Man:The question was, I donít know if you covered it, had you read Betheís thesis at the time you were writing your own?
Bloch:I really donít remember. I really donít remember. Anyhow, I donít know that either Bethe or I would have thought that the presence of a gap and reflection of X-rays were Ė-
Man:Because he practically invents the band gap. But he doesnít quite say it.
Bloch:Well, thatís right. Thatís right.
Hoddeson:OK, then you have this article which must have been a talk in theÖWe already looked at this briefly. This is just in the same period.
Hoddeson:After the summer of Ė-
Bloch:Yes, yes, yes.
Hoddeson:Yes, now, thereís a reference to —
Bloch:You showed that to me before.
Hoddeson:Yes, I did, thatís right. The subject is band theory as well as photoelectric emission and things like that. In this paper, this is actually the first really clear discussion of the bands idea that Iíve come across.
Bloch:Well, I also had this sort of thing. I mean, this break —
Hoddeson:No, here I meant the metal —
Bloch:— oh, metal insulator —
Hoddeson:— insulator here, in terms of these pictures that are due to —
Bloch:— yes, well, that was just a popularization of Wilsonís ideas, I would call it.
Hoddeson:I see. OK. (crosstalk)
Bloch:You see,Ö sort of more general audience, and so you sort of tried to make it simple.
Hoddeson:OK, thatís what I wanted to know, what Wilsonís input was, because the only place you mention Wilson is when you discuss the semiconductor.
Bloch:Oh, thatís not decent of me. I should have ... before.
Hoddeson:OK, thatís what I wanted to know. Thereís no reference to Briand?
Hoddeson:Peierls of course for the weak potential case.
Bloch:Well, I told you, I really only me Briand two years later. Personally. And already you say, hisÖwell, when I wrote this paper, I didnít pay much attention.
Hoddeson:That suggests that his work wasnít being circulated or talked about very much in Germany.
Bloch:I would think so or I probably would have heard about it, yes.
Hoddeson:OK, thatís interesting.
Bloch:Well, but you see, so often with these things, the importance of what one did often came out much later. Probably Briand didnít think either that hisÖwas important.
Hoddeson:Then you go to Copenhagen in 1931 via Soviet Union, and Landauís invitation.
Man:Was Landau in Copenhagen or Soviet Union then?
Bloch:No, no, he was in Russia then. I men him before in Copenhagen, and at that time he and Ė and I went to Russia, saw him in Kharkov.
Hoddeson:Did you discuss—
Bloch:— then also in Leningrad —
Hoddeson:— discuss ferromagnetism and superconductivity with Landau, also with Niels Bohr?
Bloch:Ja, with Niels Bohr later, yes yes. Bohr was very much interested in superconductivity. But Landau, I donít really know. I mean I had pretty much given up on superconductivity at that time, I mean, when I was in Russia, but then Bohr had gone into this thing and he was very much interested, had all kinds of great ideas.
Man:What was the role of superconductivity? Was there a feeling that the theory may be wrong because you didnít?
Bloch:No, it was, I told it to — it was one of the puzzles, that physics knew it was a very strange thing.
Man:Yes, but the feeling was that it was solvable within the framework?
Bloch:We tried it certainly, yes. But it was harder than before.
Hoddeson:Are there any other strong memories of your visit to the Soviet Union? You visited Kharkov, Moscow and Leningrad?
Bloch:Yes. I met Thom at that time and was very much impressed. In fact I stayed at his home in Moscow with him.
Hoddeson:In what way did he influence your work?
Bloch:Well, I donít know that he influenced my work directly, but there was again a feeling of a man who was very much, as I said, very good thinker. Thom and I became very good friends at that time.
Hoddeson:While you were in Copenhagen I guess you wrote your review on magnetism, is that correct?
Bloch:Yes, I think so. Yes. In fact I mentioned that in my paper here. Peierls (crosstalk) reference to that. Thatís really amusing. Here.
Hoddeson:This is the letter —
Bloch:— while I waited for Bohr, I, like, you know what Born says ? Born says one of his priests, his high priests, like the old high priests, and I say, I donít mean to hit Pauli by that, Iím writing my handbook article and at the moment Iím annoyed that there is something so stupid as thermodynamics. Amusing. I also say, ďBohr hasnít come yet. I hope that God will give me long enough life that I will still see him some day.Ē Amusing. It is so typical of the way we wrote at that time. The jokes of physics were not separated from each other.
Hoddeson:Was that your main work in Copenhagen, writing the review on —
Bloch:— oh no, no, no. My main work was on the stopping [?] power of particles in matter.
Hoddeson:And did you discuss that with Bohr?
Bloch:Oh yes. Have you seen my reminiscences of Niels Bohr?
Bloch:You have that. I mentioned that there, at that time. I knew nothing about the problem. It was an old problem that Bohr had done in 1930. Then I picked it. That was my main work in Copenhagen, and Bohr was very much interested in that. I mean, superconductivity came out only marginally, I would say.
Hoddeson:I see. OK. But then you went back to superconductivity, didnít you, in about 1932 or so?
Bloch:Not really. Not really. I came back to it very much later, in connection with the effect, but that was over ten years.
Hoddeson:Were you still in Leipzig, at the 1933 conference that was held there on ferromagnetism?
Bloch:I donít think so. That was already after Hitler was in. No, I didnít go.
Hoddeson:You left —
Bloch:— in the spring of Ď33.
Hoddeson:In the spring of Ď33, not because you were thrown out, but because you thought that, the writing was on the wall.
Bloch:Of course, and then, with my Jewish name — besides, I — after all, I despised those guys from the beginning. But you said, I saw the writing on the wall. This is true. InÖ before Hitler came to power, already I applied for a Rockefeller fellowship. So, when he came to power, he came to power a little earlier than I thought, but I had already my Rockefeller Fellowship, so there was only the summer between — and thatís when I went to Paris. And to Copenhagen.
Hoddeson:You went home to Switzerland for a while.
Hoddeson:Not for very long though?
Bloch:Well, yes, I spent more or less the whole summer there. Except for Paris in between. And Italy.
Hoddeson:You told Weiner that you got an ironic letter from the dean of the university asking you to come back and teach your courses, and that they would provide guards. Incredible story.
Hoddeson:Then you went to Paris, where you taught briefly.
Hoddeson:You mentioned that already.
Bloch:Well, I gave some lectures at the — yes. I stayed at the —
Hoddeson:Right, and you mentioned that you met Briand there.
Hoddeson:Did you see him very frequently?
Bloch:Well, I donít remember.
Hoddeson:But you did talk about physics? and everything else.
Bloch:Oh sure, of course. Look, that was again a very close knit group. There wasÖ and Briand and well, all these people there,Ö We worked together all the time.
Man:Did you ask the Peierls question?
Hoddeson:I brought it up.
Bloch:Ja, somehow, I donít quite know why it is that Briand — I said to your wife, it may have been that I spoke French. My communication with Briand was simply better than Peierlsí. Peierls I donít think spoke French.
Hoddeson:In Italy in 1933, you say in your interview with Weiner that you discussed second quantization, with Fermi, in connection with his theory of the neutrino. Had you any ideas of applying some of this field theory to solid state physics might help with some of the problems?
Bloch:Well, not solid state physics so much. You see, I used second quantization in a different way, because I had before written a paper on stopping power, where I considered the atom as a gas, you know, and there I felt that again, it was a sort of idea, that could no longer use as a parameter. I know, it was, and then, I hit on the second quantization. I told Fermi about it, and I think I told that too — Fermi said, ďI donít understand a word,Ē and then FermiÖ I did think of that application. No, but you see, I wrote, if I may say so, I wrote a paper when I was in Rome on the propagation of sound waves in a Fermi gas. I mean, people say nowadays, how I created bozons out of fermions. Yes, well... I can give it to you, the reference, if you want it.
Hoddeson:Yes, I want it. Because you mention in the Kuhn interview that while you were in Rome, you wanted to understand how plasma oscillations, and the oscillations of an electron gas could be explained in terms of the quantized amplitudes.
Hoddeson:I didnít realize that plasma oscillations were in quantum systems, where were thought of before the fifties?
Bloch:Here, this is it. (German)
Hoddeson:Thatís 1934, OK. 1934, this is.
Bloch:Maybe the next volume. Oh, here it comes, I think. Yes, thatís it. Here we are.
Hoddeson:Oh, there, itís inÖ I can just dictate the reference, then I can get it. OK. So this is incoherent —
Bloch:— incoherent scattering of X-rays and propensity for fluctuations in a degenerate Fermi gas.
Hoddeson:Right, OK, and the volume here is — volume 7, OK, 1934. And the page number is 385. OK, I will get that.
Bloch:That was the sort of, in that context, OK?... came up the propagation of fields and the, I discussed it with Fermi and Fermi didnít understand a word, up until this paper on neutrinos. That was the great wonderful thing of Fermi. Not that he was in that sense, but that if somebody was not completely clear heíd say, ďStop.Ē Then I say, ďThank God, I have to think about it.Ē If even Fermi cannot understand it.
Man:This is entirely the charged Fermi gas?
Bloch:Well, I thought of electrons, yes. I mean, the charged growing out of — again, growing out of going to a paper of Heisenberg, Heisenberg had a paper onÖ but I looked at it in an additional way. Itís the same thing as the scattering of light in the sky. Thereís the radio approach, where you consider scattering of individual electrons and fluctuations, and then thereís Einsteinís approach, where you talk about the scattering of light on sound waves, and mine is sort of an application of that to the fluctuation of a Fermi gas.
Hoddeson:Did you use the words plasma oscillations in that period?
Hoddeson:That was invented later.
Bloch:Yes, Iím sure. Again, it didnít surprise me very much I just followed a Fermi gas as such, and I said, well, after all, there ought to be sound waves in it, how are they excited?
Hoddeson:Right. Very interesting. Once you got to Stanford in 1933, your interest shifted to non-solid state subjects.
Bloch:Yes, because people here worked on other things. I did some X-ray work.
Hoddeson:Right. You did give a seminar, you told Weiner, on Fritz Londonís theory of superconductivity in Ď34 or Ď35.
Bloch:Yes, that was a joint seminar we had with Oppenheimer, yes.
Hoddeson:Yes. You must have been still following it closely.
Bloch:Yes, sure, sure.
Hoddeson:Had you heard about the Meitner effect? at that time?
Bloch:Yes. Yes, I think so, because thatís what — (crosstalk)
Hoddeson:— London theory was all about, of course.
Hoddeson:OK, what I would like to do now, just in the last five or ten minutes, is generalize our discussion just a little bit, and talk about the period l926-33. As historians, weíre all obsessed with delineating and characterizing historical periods.
Bloch:Why is Ď33 for you a — ?
Hoddeson:Well, it seems to me that, at that period, first of all, there was the emigration, and many people switched from solid state at that time who were working in it, including you and Peierls and Bethe. And while other people entered who hadnít been working in solid state at that time, people like Mott and Wigner.
Bloch:Thatís true, yes.
Hoddeson:So there was a change of people.
Bloch:Of personnel, yes.
Hoddeson:A change of personnel. There was also a change in approach, it seems to me, the Wigner-Seitz paper came out in Ď33, and that seems to mark the beginning of more quantitative calculations, focused on particular —
Bloch:— yes, quite, quite...
Hoddeson:— solids, where before they were general.
Bloch:Right. That is very true.
Hoddeson:Also at that time there was a surge of review articles, not only yours, but Sommerfeld-Bethe, I mean, Ď32, Ď34, there was a big surge —
Bloch:— I realize, it was sort of the end of a certain period —
Hoddeson:— summing up the time. And also it was a time when — well, this was the time when the textbooks were beginning, the subject was being summarized, so thatís why I call it a breaking point, to a degree.
Bloch:Well, now what you say, I see. I didnít quite understand before, but now you explain it, yes. Itís true, yes.
Hoddeson:And that period seems to begin about 1926 with the Pauli work.
Bloch:Yes, I think so. Pauli was a start, yes.
Hoddeson:So that seems to be a period. But there seems to be a break. Itís not just Ď26 to Ď33. At about Ď28, there seems to be a change also. Up till Ď28 it seems to be a kind of developing the tools, whereas from Ď28 to Ď33, it looks to me as thought it were a time when things were being worked out.
Hoddeson:Do you agree with that?
Bloch:Yes, I think thatís true.
Hoddeson:OK — did the political events in Ď33 have a direct impact on your changing fields?
Bloch:Of course. Oh yes, definitely, yes.
Hoddeson:You probably would have stayed in solid state.
Bloch:More by the fact that we were put in a new environment. For example, my paper on the theory of different times came because I spent a summer in Zurich, as you remember, and there, I think Pauli told us about Diracís theory where he introduces different times, poly [?] for each particle. I wrote a paper because I was there already by then.
Hoddeson:Right. Partly it was that they were new and exciting discoveries in quantum —
Bloch:I would say it was more the new environment to which we were, you might say, forced. Hitler did me a great deal of favor, but he didnít intend it.
Hoddeson:In retrospect, you were writing two big reviews at that time, magnetism and (crosstalk)
Hoddeson:And in this period, did it feel as though this were the end and the beginning of an era? While you were writing those?
Bloch:Well, I donít know that I thought in these terms one way or the other. I mean, I was getting paid for writing up articles. So I wrote articles.
Hoddeson:Lots of people were being asked at that time to write articles. Do you remember?
Bloch:Ja. Well, I think that was more the job of the editors. They were the smart people who said ďThis is time.Ē
Hoddeson:ďThis is the end of an era.Ē
Bloch:And we — being very mercenary people they offered money for our articles and we wrote articles.
Hoddeson:Who was the editor at that time of the Ė-
Bloch:The HANDBOOK OF RADIOLOGY.
Hoddeson:At that time, I see.
Bloch:That was Marx, I think. Not the founder of Socialism.
Hoddeson:No. Well, weíve run out of my questions.
Bloch:Well, fine, I think that timing was very good. Yes, I mean, if you have more questions, but I would rather, as I say, I find it very difficult in writing because I try to be too complete and so on.
Hoddeson:May I call you on the telephone with questions?
Bloch:Yes. You can also write to me. If my answer doesnít come soon or so, I wish, I didnít have to write too much. Iím just lazy.
Hoddeson:Weíre the same way.
Bloch:Just a lazy man, thatís all.
Man:Another alternative is what we did with Peierls, to ask questions on a cassette tape. She sent to Peierls questions on a cassette tape and he just dictated answers back on the tape.
Bloch:Oh, ja. Well, that — I donít particularly like.
Hoddeson:OK. Each person has his or her own way of dealing with things.
Bloch:Well, Peierls is not as lazy as I am.
Well, thank you very very much.