Oral History Transcript — Dr. Pierre Aigrain
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. Pierre Aigrain
Dr. Pierre Aigrain; May 22, 1981
Abstract: Autobiographical highlights relevant to solid state physics history in France. Transport of interests from Carnegie Institute of Technology (centering around Frederich Seitz) to Paris (by Aigrain) and start of semi-conductor research in France at College de France and its satellites. Contribution of his students and co-workers.
Hoddeson:This is Lillian Hoddeson and today Iím with Leonard Dobrzynski in Paris and weíre speaking with Pierre Aigrain, and itís the 22nd of May, 1981, and this is really a preliminary interview to help us get oriented, and perhaps there will be a more detailed session later on specific papers. I suppose the main subject of the interview will be the emergence of semiconductor research in France, in which you played such a large role, and I hope that you will try to follow your life chronologically, but I hope that you will take the opportunity in between to act as an observer, and tell us what was happening around you of importance, so that we can go back -- So we will start with your birth. You were born in 1924?
Aigrain:I was born in l924.
Hoddeson:Your father was an engineer?
Aigrain:My father was an engineer.
Hoddeson:Was that where your scientific interests came from?
Aigrain:To some extent. But even so, my initial line in life was not scientific. However, for reasons which were connected with my family, my father was an engineer -- my mother was a daughter of an ex-bricklayer who created his own company, and had been very successful -- my father took over, on my grandfatherís death, before I was born, and turned out to be a good engineer but a very poor manager, so the company failed rather rapidly, and my parentsí lives started going on the rocks relatively early. So that by the time I was finishing my secondary education, my parents were already partly separated. My mother was a very strong personality. She had decided that I would not become an engineer. It was a kind of a generalization -- one engineer -- in math and [???] Now, it seems indeed I was getting better grades, marks in physics, as I know the subjects. She pushed me in one direction, where math and physics would be useful to get in, but it should not be engineering or teaching, and so I became a naval officer. I started in life as a naval officer. I entered --
Hoddeson:Do you know about when --?
Aigrain:-- in Ď42.
Hoddeson:In Ď42, during the war.
Aigrain:During the war. During the war, in 1942, the south part of France was not occupied, yet. I entered the Naval Academy in 8th of October l942. But it turns out that less than one month later, the south of France was occupied, the French Navy was scattered in Toulon, I remember, and the Naval Academy was momentary closed. It was later re-opened as a civilian naval academy, for civilian naval officers, and I joined it again at that time, especially since my mother had just been arrested by the Germans, in between being arrested by the Germans, was sent to Ravensbrück, and came back only [???] And then of course there was the liberation of France. I was still in French Navy, and I applied then to become a naval pilot, naval flier, and was sent to the United States.
Hoddeson:At this stage, did you have any scientific training at all?
Aigrain:Not really. Iíd always -- well, we had some scientific courses at the Naval Academy, but they were certainly not very [???]. I had a personal interest in science, and kept reading, but I would say, I was much more a self-taught scientist, and there was a lot to learn.
Hoddeson:Before you got to the United States, were there any particular books that you read that were very important to you that you remember?
Aigrain:If I had to cite one, well, I would cite first some of the book of vulgarization written by de Broglie, and the famous Dirac treatise on quantum mechanics.
Hoddeson:Well, then you were really quite well along in physics.
Aigrain:Well, I was interested in such. I was reading it and trying to understand, alone; this subject of course was not even taught in France at that time.
Hoddeson:Well, thatís quite a book to read without any training.
Aigrain:It takes a little longer time, thatís all.
Aigrain:You have to read it yourself a number of times before you really understand. Now, I was sent to the United States, first to Chapel Hill, North Carolina, for preflight training, then to the Naval Air Station at Millington, Tennessee, and that was the happiest thing in my life, flight, flight training. So, the Navy decided to take me back to Paris. But as I was waiting in New York for a Ship, to get ready to take me back to Paris, another naval officer, older than I was ó- I mean, I was not the only one with flight training, there were several -- people in the same condition -- met us in New York. And he said, ďBut I donít understand the Navy, because of course they need naval fliers, but they probably donít realize that they probably need electronics people even more.Ē Because after all, radar had developed during -- and radar education, -- during the Second World War, and most French naval officers were untrained in these areas. So he managed to convince the French Navy to let those of us who could get in take a masterís science course in a university -- it was Carnegie Tech in electrical engineering at this point.
Hoddeson:Now, let me see. You were in New York.
Aigrain:I was in New York.
Hoddeson:This manís name is Mathis?
Aigrain:No, Guigonis, who is now retired, not too long ago. He went to industry and now is retired. And so eight of us including myself ended up at Carnegie Tech as graduate students working towards a masterís degree in electrical engineering. And so my real formal scientific training, and my first contact with research which you do as a Master of Science thesis, was in the field of electronic circuitís research, and I got my masterís degree in 1946, and at that time, the university offered me a fellowship to go on for a doctorate. The French Navy, after scratching its head for a long time, -- from the stories, people going back to France, and always by ship at that time , travel was complicated -- and on the day I was supposed to ship out of New York back to France, the French Navy had not yet answered if I was allowed to stay in. And then Guigonis did a second very important thing in my life. He signed a personal order that I was not to proceed to France, instead stay in New York... so indeed the permission to stay arrived three days after the ship left. So I stayed, for two more years at Carnegie Tech. And I finally got the PhD in EE on circuit theory, but this is also the time where I started studying solid state physics.
Hoddeson:Seitz was there then.
Aigrain:Seitz was there. And in fact this was directly connected to Seitz. Now, there comes in a second man who will be very important in my life. There was a student at the Ecole Normale Supérieure named Claude Dugas, and at that time the physics laboratory of the Ecole Normale Supérieure was headed by Yves Rocard. He is the father of the politician who is in the news, these last few days, and Yves Rocard who was himself a physicist and a mathematician, not very well versed in modern physics -- he was a classical physicist -- but with an extraordinary knack for finding and guessing what would be the important subjects in the future.
Hoddeson:What was his position then?
Aigrain:He was director of the physics laboratory of the Ecole Normale Supérieure.
Hoddeson:So he was in France.
Aigrain:He was in France.
Hoddeson:While you were in --
Aigrain:-- he was in France, but he strangely had also connections with the Navy. He had been an important man in the French Resistance. During the war, he found himself in a position to have information abort German work on V-1 missiles, very precise Information, and actual1y sailed from France in this small one person sailboat to join an American destroyer in the Mediterranean, and bring that information to London. And he was incorporated in the French Navy as scientific counselor. When he came back to France, to renew his directorship of the physics laboratory of the Ecole Normale Supérieure, he kept his contact with the Navy. Now, first, during the recon quest of Europe, as scientific counselor of the Navy, he was along with the troops, especially with the Marines, which were fighting, and as a result, he happened to be the first one to spot some German military laboratories, including one on the Lake of Constance, in which the Germans had been studying infrared sensitive cells, based on lead sulfide, lead telluride -- a field in which they had made great progress. They were ahead, of the [???] he theoz1ized, in that field -- not in a position to put the equipment in operation, but they had done a lot of advanced work. And so he convinced young physicists of the Ecole Normale Supérieure that this area was interesting, that these materials, they were semiconductors. He didnít know what semiconductors were, but he was convinced they were about things that were worth studying. Claude Dugas went to work for a few months in this laboratory on Lake Constance, now under French supervision, and he found out that he didnít know enough about solid state physics to really understand the work, so he tried to find a book which told about so1id state physics, and the book which was at that time standard in solid state physics in the world, especially -- in the United States, was Seitzís MODERN THEORY OF SOLIDS, which had come out just at the beginning of the war. And since his English at that time was not very good, he had the brilliant idea that the best way to understand the book would be to translate it. Since on top of it he was very short on money, Dugas is the son of a poor peasant, he thought, as long as he was translating it, he might make a little money about it, and he brought it to a French publisher, and he got an authorization to come out with the French version, giving him a little money. Now, when Seitz, asked for permission on his book, found out that a young Frenchman was translating his book, he invited Claude Dugas to spend one year at Carnegie Tech. And so, during this last year, 1947-48, Claude Dugas arrived at Carnegie Tech to work with Seitz. And since Rocard knew me, because through the scientific counselor of the Navy he had heard about this French naval officer who was working on a PhD out in Pittsburgh, a very manual city as we know, he and I of course immediately established contact. Now, I was taking probably more physics than electronics courses in my study. Here comes this man, who was very enthusiastic about semiconductors, to this remarkable solid state physics department of Carnegie Tech, at this time, with Seitz and others.
Hoddeson:[???], Taylor --
Aigrain:Taylor was one. He had no trouble convincing me this was important. And so I started studying semiconductor physics during my last year, at the same time that I was finishing my thesis on circuit theory.
Hoddeson:Was there any semiconductor work at all in France?
Aigrain:At that time, no semiconductor work in France. Now, if we go back, but of course, before the war I was very young, I didnít know much, but what I could find out is that first, Jaffe -- he was a Russian physicist who had come to France for one year as a visiting professor around 1934, and his lectures had been published in France, so a few people probably had heard about this. Amongst them, Liandrat, who had -- who at that time was -- I say at that time, because he was not very old -- anyway, but at that time he was a rouge Communist, who was not only interested in the physicist but also interested in the Soviet Union that he went to spend one year in the Soviet Union working on semiconductors with Frankl in the same laboratory in Leningrad where Jaffe taught. And so there was this man who had worked in semiconductors. However, he was never able to really start a school, for several reasons -- the first one is the pre-war money for science in France was non-existent -- there was no money. The second part is that he had managed to put himself in a strong fight with both the Communists and the anti-Communists. Because he had left as a strange Communist, and Frankl was important, and Frankl convinced him and he came back fighting the Communist Party, with the result, nobody supported him. And poor Liandrat did not stay in the field. He had made, by the way, a very important theoretical contribution. He was probably one of the first people in the world who thought that minority carrier injection in semiconductor was possible.
Hoddeson:Letís see, when was that?
Hoddeson:Thatís earlier than Davidov.
Hoddeson:Because it appears in Davidov.
Aigrain:Davidovís paper on the p-m injunctions is 1939.
Aigrain:And Davidov does predict the minority carrier injunction.
Hoddeson:I gather people couldnít understand through the mathematics of the --
Aigrain:His mathematics was, not sloppy, but it was difficult.
Hoddeson:So people didnít really know about it, up till the transistor. Thatís very interesting.
Aigrain:Now, Liandrat had not thought about the p-m injunction. He had an idea which worked, but which is not the easiest one, of using the field effect for minority carrier injection. The idea, you know, is to create an inversion layer by field effect, by applying a strong field perpendicular to the surface of the semiconductor, and in this inversion layer, of course, you have minority carriers. If you suddenly remove the field or inverse the field, you inject these carriers inside the semiconductors. [???]
Hoddeson:Thatís very interesting. Did he write anything about that?
Aigrain:He, I think, quoted it -- there is one -- well, one of two papers, which did not attract, as far as I can know, any real attention from the scientific world. In fact, at that time, itís really strange, but the leading idea was that you add, a semiconductor -- nobody knew that -- but that in some materials, the holes were not mobile. In others electrons were not mobile. And the idea that in the same materials, both holes and electrons could move was not accepted.
Hoddeson:Iím particularly interested in the subject. Is there a place where I might be able to look for these?
Aigrain:I think I can find the reference to Liandratís paper. Because I met Liandrat much later, because after my lab was established at the Ecole Normale Supérieure, Liandrat came to work for one year. He came to work, because he wanted to do some experiments, which he had never been able to do, which we indeed managed to make work. I might say also he attracted my attention at that time on the photoelectric magnetic phenomena, which had been discovered about 1935 by a Russian named Kikoim who was also a student of Franklís.
Hoddeson:So there was a little bit in France.
Aigrain:So there was this little bit. But this man, you know, had published two papers, had never been ab1e to do any experimental work -- he didnít start doing it till afterwards. In fact, when he came back, Dugas and me together --
Hoddeson:This was about 1948?
Aigrain:In Ď48, in June, Ď48, we found that nobody was doing anything in semiconductors in France, with the exception of an astronomer named Lallemand who was not really working in semiconductors, but was, had set up a small production line for his famous lead sulfide cells according to the German methods. It was purely empirical. And two Germans which had been hired, possibly in part by force, by the telephone administration laboratory to work in France. These two Germans, one of them very famous, Walker, the other one was a little less known, Mattaray (?), who is now consulting to [???]. Of course went back to Germany, and they were beginning to work on germanium and germanium transistors. Remember, June 1948, when I came back to France, is exactly the time at which the famous bomb paper appeared.
Aigrain:Of course the discovery was six months earlier but thatís the time the paper appeared in print.
Aigrain:So when I came back to France, Yves Rocard, director of the physics lab of the Ecole Normale Supérieure, immediately managed to tell the French Navy, ďYou are not going to use this guy, heís a doctor, a PhD -- what are you going to do with him aboard a ship?Ē He has never been aboard a ship.Ē I mean, in all my life, I was theoretically ten years in the French Navy, most of it on detached duty, and I think Iíd only spent 27 days at sea, as a naval officer, not counting the time I spent as a passenger going to the United States and coming back. So it comes to the French Navy to put me in disposal this to this laboratory, and with Claude Dugas we decided we would start working mostly on germanium, and rectification and transistor effect in germanium. Means and money and people for research were still extraordinarily short. But frankly we did not suffer from it -- because first, as far as people were concerned, we were at the Ecole Normale Supérieure there were a bunch of young people who had to spend one year, according to the rule of the school, one year doing research.
Hoddeson:Now this period is l948-50 roughly?
Hoddeson:You were officially part of the naval --
Aigrain:I was still officially part of the Navy --
Hoddeson:-- naval research --
Aigrain:And working at the Ecole Normale Supérieure. And so, we had young people who were bright and wanted to come to work. They had to do it anyway as part of their curriculum. And Rocard also managed to find money, by ways which may not always have been in agreement with the rules of public accounting but which were efficient. He managed first -- the Navy of course had captured equipment from the Germans. That equipment went to our lab. But Rocard managed to sell some of it and get money to buy others. We used -- at that time; it was easy to buy some equipment very cheap, because of the surpluses. The war surpluses. We used to buy war surpluses in New York. I used to travel back and forth. At that time air travel became more convenient.
Hoddeson:Where? On Canal St.?
Aigrain:And there we used to take this equipment apart and make our own equipment. Of course we had no technicians, it was all our work but it was fun, and also it was a period where everything you tried, in physics work. Any idea was fair game and worked. It required very little imagination to make discoveries and to publish original papers, which were important.
Hoddeson:What were the outstanding ones in that period?
Aigrain:Well, during that period, it turns out, I was working also on the theory of p-m injunctions and the phenomena of carrier multiplication on point contact transistors. I was working with Dugas. We made the first proposals for electron paramagnetic resonance on impurities in semiconductors. We couldnít carry out the experiments. We had no, didnít have no paramagnetic resonance equipment. We didnít have the low temperatures and so on. But all this was first reported I think in l949, at the Reading Conference on Semiconductors, the first international conference on semiconductors. Very few people, very small conference. And so, I would say also, there was still a lot of witchcraft in semiconductor physics, about things like surface treatments. As I say Dugas was the son of a poor peasant. His father was actuality a small grower of cognac, and so he always had some bottles of hundred year old cognac, and when one of the real top guys of the United State came to Paris, we managed to have him to come to our lab, and we served hundred year old cognac.
Aigrain:And that was a tremendous help to our work. (Laughter) The other thing which I think was important in that period, even though we did not know it was so important -- at that time, the French rule -- because of course he was working on his doctorate, and I had decided to get a French doctorate also in solid state physics, which I did in 1950, and at that time, the rule about getting your doctorate is that you had to present an original work youíd done, but three weeks before presenting your thesis, you were given by the president of the jury a subject, which was not your subject, which of course was in the same genera1 field -- you were not given the subject of history or bio1ogy -- but the subject that was supposed to be brought, it was, you know purely bibliographical (ideographical?) -- explaining the state of the art in the subject which had been given you three weeks ahead of time. It was a kind of pedagogical test really more than anything else. The subject I was given was, adsorption and catalysis in heterogamous catalysis, by salts, and as I was studying it, I was struck that a fair proportion of catalysts were semiconductors, mostly oxide or sulfide -- they were semiconductors. And we, Dugas and I got to discussing it, and we came with the idea, back with the idea that the adsorption of impurities on the surface of a semiconductor (was a conductor, that is it didnít bulk) could exp1ain the mechanism of adsorption, explain why the binding energy decreased so fast with surface coverage, because of the formation of the bio-layer, and even explained the kinetics of the process. But since we were not chemists -- I mean, I brought it up during my thesis in France, but in two minutes -- we just left that lying around. And so it was in 1951, one year later that the first scientific conference took place in postwar Germany. You know, after the war of course there were no internationa1 scientific conferences. There was a conference on electro-chemistry, and Germans had invited Alfred Kaestler, who of course was not an electro-chemist, but they invited him because he speaks well in German, as Kaestler is Alsacian.
Hoddeson:-- where was it held? --
Aigrain:It was held in Berlin, in 1951, and Kaestler came to me and said, ďWe1l, I would like to give a paper, but I donít know anything about the subject. Do you have something which I could present in your name?Ē So we quickly wrote, put on paper, that idea that I had one year ago, Dugas and myself. Kaestler translated it in German. I mention that because that was the paper of mine which has had the most references. That paper is in a language I canít understand. I never read it myself. And I did not present it. It was presented by Kaestler. I wrote it, yes.
Hoddeson:And it appeared in the PROCEEDINGS?
Aigrain:It appeared in the Zeitschrift fur Electrochemistry in 1951, and -- well, you know, itís often the case. An idea is in the air, and it developed that two Germans, Weisser -- I donít remember the other one -- that they had had a similar idea which they had not pushed quite that far, and Volkenstein in the Soviet Union had been working on similar lines. He even had started earlier, but you know, at that time the Russian publications were not available. And so, we came out with this contribution, in semiconductor physics, to physical chemistry. A paper which probably would never have been published, if it hadnít been for this chain of circumstances. We laughed a bit. We were afraid to write something about the subject at the time. In 1950, I was 26. It is also about that time that the Liandrat episode took place. He came to my lab. He wrote us orders and we said, ďWell, itís interesting to see that guy.Ē He spent one year in the lab. He was I think teaching physics in Madagascar or something like that by that time. But he couldnít take a sabbatical. And --
Hoddeson:Was, that as a result of something that happened during the war, or because (crosstalk).
Aigrain:Well, he had led a very complicated life, and I think he went on -- he was at the same time a rather complex person. He attracted our attention to the papers of Dunbar and Kikoim in Frankls lab, pre-war, in which they had reported the photoelectric magnetic effect. You know, you take a piece of semiconductor, in magnetic field and if you i1luminate it, a voltage appears, perpendicular to the magnetic field, along -- and Frankl had published a paper, proposing a theory which actually cannot possibly work. And thatís another case, Frankl started with the idea that only one type of carrier could be mobile in a given semiconductor, but as Kikoim had discovered the effect of copper oxide, which is not a very reproducible semiconductor, -- so very little was known about the material, and as a result, Frankl had made a theory -- itís easy to understand if you illuminate one phase, you are creating free carriers, and because of carriers diffusion sign, you are going to create a barrier, you are going to create a potential difference between the surface and the interior. And thatís the Dunbar effect. And the Dunbar effect cannot occur in a semiconductor in which one of the carriers would not be mobile because in fact, if the two carrierís holes and electrons have the same ability, there would be no Dunbar effect, because the diffusion of the electron -- holes pairs carries no current and so there would be no voltage. So Frankl made a theory of the Dunbar effect, and it was all right, as long as one of the mobilityís was zero. But itís also easy to see that if the two carriers are not simultaneously mobile, there can be no photoelectric magnetic effect, as in the steady state. Because once the voltage is established, carrier motion is nil and the magnetic field doesnít do anything on the mobile carriers. So he had tried to compute -- he had assumed there was a Hall Effect, even when you had a voltage and no carrier motion, which is not true. Which in fact is in contradiction with the principle of energy conservation? In that way too you could think of perpetual motion. I was struck by this contradiction, and since at that time the theory of carrier mobility, Hall effect fact, with holes and electrons both mobile, was possible, we just made it, and from the explanation of the photoelectric magnetic effect, and found that the photoelectric magnetic effect was a very convenient way to study many properties of semiconductors which had not been quite as easily accessible to them. So thatís how we started in that area.
Hoddeson:Was it just you two? Or were there others?
Aigrain:Of course I had some students working on the problem. A young man named Buliar who has since died made a thesis on the subject. It was, by the way, some aspects of the theory were worked on by a young student, a Master of Science student, named Philippe Nozières, who has since become one of the best solid state theoreticians in France. But that was just his first.
Hoddeson:He was a student at that time?
Aigrain:Yes. Well, that was the time when we had Master of Science students like Philippe Nozières, Pierre Gilles de Gennes -- he was there two years or one. It is easy when you have students like that. When you have a [???] director who attract -- even though he has no credit, he-manages to give you the equipment and even money when you need it -- and when you are in a field where, as one of my students was saying, all, you have to do is to take any piece of semiconductor, put a magnetic field and an electric field line or a point contact or a junction on it -- you apply any of the three, and you observe the fourth. And he claims that you could do anything like that, except possibly make point contacts by applying fields. He found no way to make whiskers that way. That was all. It was almost true. It was a burgeoning field in which every step you took, even wrong, was more or less a historic start. Too easy to be true.
Hoddeson:Would this have happened if the transistor had not been discovered in December, 1940, the point contact transistor. Or --
Aigrain:Itís not obvious, that would happen. Of course, first, itís not obvious the Navy would have put me on a ship, which would have gone aground quickly, because see, they would not have been as interested in spite of the power of persuasion of Yves Rocard. So, Dugas might have started working on it. But Iím not sure. The transistor was really, it was very important modulation, even though it diversified very quickly, but still.
Hoddeson:Right. And also another question is the first transistor was of course the point contact.
Hoddeson:And you did study those problems. Was there any thought at that time of producing another type of transistor?
Aigrain:Well, yes. In the first place because, when I studied the point contact transistor, my own feeling was that it was really a junction transistor. The point contact was just a complex way to make p-m junctions. And my thesis, I donít even know whether I have a copy, itís all, my thesis wouldnít be interesting nowadays, --
Hoddeson:-- do you have your early papers somewhere?
Aigrain:I may have them somewhere at home, I donít know. I will try to find them. My thesis tried to develop a theory of p-m junction, minority carrier injection. The Davidov paper mentions minority carrier injection but he doesnít treat it. The Shockley paper on the p-m junction appeared shortly after my first notes on the p-m junction it appeared in the [???] is a talk, is so much more powerful than the miserable mathematical treatment that, I was applying, that of course, his paper is the one which is cited. And which this is perfectly Ö But I actually published a theory of p-m junction shortly before Shockley. So we were of course interested in junction transistors very early. The junction transistor was reported for the first time at the Redding Conference, in Ď49, by Shockley. It was the first experiment on transistors.
Dobrzynski:Where did you publish your paper? The one --
Aigrain:At that time, I also presented my own theory, but the idea of the p-m junction transistor is Shockley. I just collaborated. Thatís clear. But at the same time, we had been working on... transmit my policy to my successor, you know? As of today, Iím unemployed. Since this morning. The glorious uncertainties of democracy. Because of, we had started working on field effect, and of course, [???] about field effect transistors. The technology of, which we had at our disposal, did not make it too easy Remember; silicon was unknown as a material. Germanium is not too easy to make an oxide layer, which, we tried to work at that time with molybdenum sulfide, because molybdenum sulfide cleaves, and itís very easy to make a thin layer, a single crystal layer of melidium(?)sulfide. Because you can peel it off, better than mica. But of course, molybdenum sulfide is very hard to make synthetically, and we tried to use natural crystals, which have very poor homogeneity, varying doping and so on, and so it led to some interesting work on the properties of compensated semiconductors, which had not been studied very much. We were not successful in making it practical, field effect transistor at that time. The other thing we did, however, which was important -- oh by the way the transistor had become so important that around 1952, just at the time I was finally leaving the French Navy, and I was offered a job of associate professor at University of Lille for two years, then in Paris, but then I stayed in my laboratory in Paris -- at the same time, an electronics company called C.S.F. (just since merged in the Thomsen group) and the boss of that company, Maurice Ponte who is himself a physicist, Maurice Ponte got interested in starting a laboratory on semiconductors physics, and he approached Dugas and me saying ďI want you to come to my company.Ē The offer seemed good to usÖ Dugas said yes and went with the company. I myself became a consultant to them and kept in touch. But I wanted a university career. And so, the university laboratory was from then on under my leadership alone, since Dugas was in industry. That in a way was a great help, because of course, in industry, they were able to set up crystal growing equipment and so on, and this was a great support to the laboratory. It was also a source of ideas about fundamental phenomenal worth to be investigated.
Dobrzynski:When was it created?
Aigrain:The laboratory was created in 1952.
Dobrzynski:You were also at the College de France at that time?
Aigrain:Well no. At the College of France, I was Assistant to Professor Laval -- but as you know, there was no laboratory at the College of France. I helped Laval with his lectures and that was it. At the Atomic Energy Commission; I was at the Atomic Energy Commission by Rocard, but because RocardÖ thought that the members of the Committee had responsibilities, operational responsibilities, which wasnít true, so he hired me and found out in a few days there was nothing to do, and so, I would say I more or less consulted with them for one year, and -- and left as soon as --. So this appears on my curriculum vitae but I would say, there were two jobs which took very little out of my time. 90 percent of my time was concerned with my laboratory. And itís about 1952 --
Hoddeson:Wait, one more question. Where did you get the germanium from? The materials in general?
Aigrain:Well, we got the germanium at the beginning in part from the telephone administration laboratory where Walker and McHaray started making germanium. We started making germanium crystals ourselves. Getting germanium metal was easy. You can buy that. Purifying it, just about the time zone melting came along, and thatĎs easy with germanium. Making crystals with germanium is relatively easy. At that time, people were not looking for non-dislocated crystals and things of that kind. I mean, you formed very poor crystals, compared to present day standards, -- but they were perfectly sufficient for all the work we did. No problem. And we had set it up our own crystal making too progressively. Well, the little problem we had, and which was important, for the future, which was a handicap for our lab, is that we were very slow in studying silicon, and we were very slow in studying III- V compounds, excepting lead [???] because we just didnít have the technical and financial capacity to set up the production of GaAs or silicon crystals, because these are much more difficult to handle than germanium or Gudium(?) which have low melting point. Thatís true. And that was a handicap of sorts, as we see later. Because in 1952, I got another young student started, a man named Benoit a la Guillaume. He was there spending his one year research, and Iíd come to the conclusion, as with Philip, that if you want to start some research protects which have a low probability of success, a problem where youíre not likely to succeed, this system where you have one year, Master of Science level thesis may be excellent because after all, if it fails, the man is still going to get his degree, as long as he has shown reasonable ability in making the experiment, nothing of any consequence to the student, while if you start the student on his doctorís thesis, on a very uncertain subject you may be creating a little prob1en for him, if after some time you find it doesnít work. So I started Benoit a la Guillaume on the project of trying to see whether, when minority carriers recombined, after injection, in a semiconductor, some light was not emitted, which might lead to a spectroscopy of semiconductors. That was reported for the first time, the combination light of semiconductors, at the 1954 Amsterdam Conference. And by the way, Benoit a la Guillaume has been working on the combination light in semiconductors ever since. And the field has of course opened tremendously. Itís still a very active field. So that was a very important, I think, contribution. What follows may be less brilliant. In 1956, there came to my laboratory a young American who was an RCA fellow, Jack Pinkove who is still, with RCA now. By the way, at that time, when we look at the literature I think he was signing as Pancheshnikov which was his original name. He was an American of Russian descent, and changed his name to Pinkove to simplify it, as Pancheshnikov was too complicated. We started working on the idea that the emission of light from recombining carriers in semiconductors could be stimulated. And the problem is that we were working with germanium, not with a III-V because we did not have III-V at that time, and we pursued experimentally the hope to make a semiconductor laser with germanium, taking advantage of the fact that, If you have a simultaneous emission of a photon and a phonon, you donít even need a comp1ete inversion of population to get stimulated emission. But unfortunately, the numbers came out just wrong. You have always competition between stimulated emission and free carrier absorption, and in germanium, the free carrier absorption is a factor of two or three too large. So it was a border line case, so we were unable to make a semiconductor laser with germanium, and nobody has ever been able to make it, for this same reason. But we had really a theory of stimulated emission quite well worked out. We reported on it in í58, in leading places on the theory, and of course as you know itís about two years later, in 1960, semiconductor lasers were first reported on, with junction transistors with GaAs an electron beam pulse in semiconductors with cadmium sulfide by Basil (Soviet Union)
Hoddeson:There was also some Japanese and Russian work.
Aigrain:Yes, Barsov. I think the first report --
Dobrzynski:Thereís a book by Sze that mentions that you, some Russian and some Japanese reported independently.
Aigrain:Independently on the idea. But anyway, the fact that we were working on germanium -- we had the wrong material. In fact, we had set up an electron beam point system, and after the report on the GaAs laser, we had the GaAs laser working. It was just necessary to put our hands on the GaAs and it worked. After that, we studied -- another one of our students, Debever (?) who is now a professor at Marseille Luminy, passed his thesis, in which he reported on the lasing action in semiconductors on, I think fifteen different compounds I think. Itís a very [???] study... importantÖ butÖ
Hoddeson:It sounds as if you had a very big group of students at this time.
Aigrain:By that time, 1960, I would say we had close to 100 --
Hoddeson:-- A hundred --
Aigrain:-- (researchers) who were working; the work had expanded tremendously. We were both at the Ecole Normale Supérieure and shortly after 1960 we moved a good part of the lab to the Halles [???] where a lot of space was available.
Hoddeson:A hundred is a very big number. Perhaps you could tell us -- more about the way your hundred was, it must have been a sub-organization.
Aigrain:Yes. Well, you know, some of the first people who come in we knew were good and they quite naturally become the leaders of the sub-groups. These people were at the time theoreticians like Philippe Nozières. There was Julian Bok.
Hoddeson:Letís see, Nozières was not working on semiconductors or was he?
Aigrain:Nozières was working at the beginning on semiconductors. He later diversified to mostly magnetic problems, which isÖ
Hoddeson:-- this is before he then went to Princeton to work with Pines.
Aigrain:Pines? Yes. He went about toward the middle of that period. He went to Princeton, and Pines then came to our lab twice, because we always had visitors from aboard, he came three times, rather. So some of the theoretical group got less interested in semiconductors, and started moving toward many body problems, plasma effects and things of that kind. Julian Bok was still there, had taken on a group working mostly on plasma effects in semiconductors and hot electron problems. Benoit a la Guillaume was working on the combination light. Collette Rigaux had been started working. She had taken up a problem at that time; it was te11urium, which turned out to be interesting from a fundamental point of view, because itís one of the very few non-Centro symmetric elements. You see, itís very easy to handle to make crystals and there a lot of extremely interesting magneto optical phoneme to be observed in tellurium, which developed as a line of its own. Tuillier who since died had started a group which had been working on absorption on semi-conductors, as a byproduct of that 1951 paper. A man named Garetta who later went into industry and has probably since died, had started working on Peltier heat, refrigeration. That was started in 1951. Realizing that with semiconductors, it might be possible to take practical Peltier heat or thermoelectric engines and be able to make ice. It was the first test, you know, at room temperature, and made ice, with semiconductors at the end of 1951. Kept up the actively which since disapproved. So the groups had more or less diversified. I donít remember the exact date at which a physicist or e1ectrochemist-actually which had converted to semiconductor physics, and was not trained in our laboratory, but which had converted to solid state physics during a stay in Berlin -- named Balkanski, came to the lab and started a group which had been working mostly on excitons and which later separated into an independent laboratory. Some of the people at the lab -- by that time, semiconductors had become quite in fashion, so some of the people at the lab left and established groups of their own in provincial universities. Godefray (?) in Dijon, the Debever and others, [???] in Marseille. Also some people came to the lab and learned about semiconductors and set up their own group later, like Rodot in Bellevue. So I mean, we had become a little place which was planting little plants everywhere, and creating a healthy competition, in France. But we always kept very good relations with these groups on all sides, sometimes competing with them, but we used to meet all the time, exchange... Itís always been a very happy atmosphere. Groups have kept... Another line which we started at that time was that of radiation effects in semiconductors with Pierre Baruch.
Hoddeson:When did that start? (Crosstalk)
Aigrain:That started in a rather strange way. It started because we happened to get a van de Graaf. The van de Graaf had been acquired for the nuclear physicists, who were made to wait quietly while the linear accelerator in [???] was being built by the Ecole Normale Supérieure, and as soon as that accelerator was running, they lost interest in the van de Graaf, so we used it for radiation effect in semiconductors, and there too another man came in, Ansel, who had the idea to apply the equipment to a method of nuclear reaction in thin film analysis which more or less became a skill connected with the lab, a more or less independent activity, so we had a kind of diversification at that time. The work on plasma effects in semiconductors actually started because of a theoretical work Iíd done myself, about the prediction of wave propagation in magnetized solid state plasma. That was the heliconÖ
Hoddeson:How did you start the early roots of that? Do you remember what work --?
Aigrain:I think, strangely, what started me on this problem of what happened was a combination of two things, one purely pedagogical, -- I was teaching of course, and when you start teaching, sometimes you try to find a new way to express something to your students, and once on occasion you find that you may have a new approach to phenomenonís. And the teaching of Hall Effect, even though itís such a simple phenomenon, is not as easy, itís not as easy to explain what the Hall Effect really is, if you want to get into detail. One day, I had the idea of --
Hoddeson:This is now what year, about?
Aigrain:It was about 1958-59. I reported it I think at the 1960? I reported it at the [???] Conference on Semiconductors, I donít remember what year was the [???] Conference. Thatís easy to look up. So I had the idea to look at the Hall effect in a plate of semiconductor in which, you see, the electric field would be rotating, -- and in which at that time, because of the Hall angle, the current would also be rotating, but would either lag or lead the electric field, so that you had the active component of power and the reactive component of power, and then the Hall effect appeared as a marginal part of the impedance. And actually it makes a number of analyses much easier to look at the Hall Effect that way, in a rotating frame of reference. And just about that time, during a visit to the RCA lab, they told me, they have been looking at the phenomenon which we had predicted, with the right valve but strangely we got the wrong sign. ďItís the wrong sign and I have to tell you why -- because we never computed the sign. We were so convinced of the sign, we computed the magnitude and had the sign there, and got it wrong, and thatís what should be called the photo power magnetic effect, even though our report speaks about the photodynamic magnetic effect -- and essentially the idea is, that if you illuminate a semiconductor, in a magnetic field, the same geometry as the photoelectric magnetic effect, what happens is that, the hole-electrons pairs which diffuse perpendicular to the surface, they are separated by the magnetic field, which deflects them, and so you have a current on the front face of the semiconductor, but this plate is isolated. There can be net council of the current in terms, in three terms, through the current loops which of the semiconductor. So as a whole is that you have a count-too, and this current loops produces a magnetic field. So you have a reaction of the illuminted semiconductor and the magnetic field, acts as a non-zero permeability material, and the strange thing is that the effect can be quite large. Itís not a small effect. In extreme cases, very small illumination, it can be larger than [???] ThatísÖ in extreme cases. But anyway, even in the case of standard semiconductors, which -- of --course -- was our domain the effect is 10-4 or 10-3 which is large for -- and you know, after computing that, we applied the general idea of Lanslaw that [???] spontaneous always tends to react against magnetic fieldÖ If you look at, as a whole, you convince yourself very quickly itís paramagnetic, and the people at RCA reported that. They tried to measure it in our problem. They were trying to measure something else. They were trying to measure the influence of illumination on the diamagnetic properties of impurities, you see, in the semiconductor, and of course they couldnít see it, because this effect is so large that it dominated everything. They asked me to look at this, because they were getting the right values with the wrong sign. Of course they were right. They simply looked at the problem, through this theory, that should have it the opposite. But after that, as well as they could compensate for it, they were finding a lot of effects which they could not explain, which have never been explained, and probably do not exist anymore. It was probably an artifact. And when I came back I started thinking about, what could cause these artifacts? And well, I just put on paper -- itís a very simple theory of helium really simple, goes in four lines -- just put it on paper, if you have a semiconductor in a strong enough magnetic field, even at room temperature, you could have essentially absorption as propagation of the electromagnetic waves, and circularly polarized, along magnetic field lines. That these waves are slowed down tremendously, and so we decided to work on that. There are a lot of funny things about solid state physics. A meta1 physicist, Raymond Bowls, of Cornell University, had been working trying to measure the magneto resistance of simple meta1s at liquid helium temperature, essentially sodium, and in order to do that, since at low temperatures, these pure metals are of enormous conductivities. Very difficult to make contacts which donít perturb the experiment. So he tried to make a conduct less metal -- that is, take a sample, put it in a strong direct magnetic field, and then he would induce a perpendicular current by a loop, get it down and measure the decay of the induced magnetic field, due to any currents by Lingy. Instead of a simple exponential decay, he was getting at astonishing low frequencies, 50 cycles, and whose lifetimes in the seconds, and just in [???] So he started working very hard on it, all 1961, Philippe Nozières was at the [???] Raymond Bowls told him, and presented the theory, which was essentially the helicon propagation, Philippe Nozières told him, ďThat theory was published years ago.Ē Well, many physicists in these conditions would have been offended, very downcast. Raymond Bowls was honestly wonderful. He was the first one to observe the waves, and gave us much more than proper credit. So it turns out that this helicon wave propagation was very useful. Possibly not for the reason we had expected at the beginning, but in two ways. In the first place, it makes it possible to propagate a high frequency field in a compact medium, and was helpful for nuclear magnetic resonance experiments in metals and semiconductors. In the second place, it turned out that it was a good way, combined with others to studied Fermi surfaces in metals. And so we studied plasma effects and thatís also why we started an activity on plasma, on the Fermiology of metals. By the time, when a Laboratory gets to that size, what happens is, the young people who did come for me to train, they are getting a lot better than me, and they are getting ideas, so most of the ideas of the lab, starting from Ď56 and so on, just are not mine. They were the ideas of the others, all, the guys in the lab. They started in many ways, and I was the old man who was still able to give some advice, but they had the ideas.
Hoddeson:The way you described it, your laboratory was the place where this area of solid state grew up, and put out lots of graduate, seeds --
Aigrain:-- in semiconductors, yes.
Hoddeson:-- and sent them out. Were there any important independent developments that we should --
Aigrain:The only one that was really independent, completely independent, that we should get very good relations with, was the one at C.N.E.T. at the telephone research laboratory.
Hoddeson:That was actually a basic research lab?
Aigrain:-- That also included basic research, and there was first [???] and then some younger people, one of which was, [???] Bernard, who by the way is now director general of that telephone administration research lab.
Hoddeson:So he would perhaps be a good person to see.
Aigrain:Heís a very good person to see, too. Now, that line developed independently of our lab, though with, close scientific relations and sometimes cooperation. See, after my first proposals on semiconductor lasers, it is Bernard and a younger man named Durafour who first expressed in a simple way, the famous Bernard and Durafour condition for related emissions in semiconductors, still the basis. So, we kept working very closely.
Hoddeson:What about exchanges between -- you mentioned that Nozières going to Princeton and David Pines.
Aigrain:David Pines, Smoluchowski (crosstalk) came several times to my lab; Weiner, of RCA, also spent one year in our lab. We often had shorter visitors from Bell Labs and so on, and my students spent time in the United States on many occasions, summer visitors, myself -- you know, Iíve crossed the Atlantic more than a hundred times in my life, and --
Hoddeson:From your point of view, where are the main centers of research in the US in semiconductors? Bell Labs, of course.
Aigrain:Bell Labs was clearly number I -- probably still is. But of course I would say every important university had a center; every important company, GE, RCA had centers.
Hoddeson:Were there other companies besides the telephone one in France, where basic research in semiconductors was done in an industrial context?
Aigrain:Well, thereís a Bellevue laboratory of Rodot which started, in contact with us, but Rodot decided to convert to semiconductors and came to our lab for one year to finish learning, to train and then started their own lab, and they had very close contact with -- SAT which SAG was anther industrial company, mostly on infra-red cells. They weíre working; you know, on, among other things, Cadmium telluride. I suggested to them they should work on cadmium mercury telluride, tertiary alloys and every time they see us they say. It was a good idea because it did work beautifully, because after all the first variable gap, adjustable gap, with that mixture, and has been used operationally for detectors. And then they converted mostly to photo-voltaic cells which we never worked on. So this was another group. There were also, there was one group which had a very special line. It was Nikitine in Strasbourg who studied, carried out a lot of studies of purely optical, of exciters in compound semiconductors.
Hoddeson:In the US, there was a big transfer. People were being trained in academic centers and then going to work in industry such as Bell Labs and GE.
Aigrain:This is unusual in France. But if there is one area in which it occurs to a significant extent, it is in semiconductors.
Aigrain:I would say a good deal of the important people, some of the first people moved and worked in the various industrial companies: Thomson...? And so on, were trained in my lab or those, I would say more or less satellite labs, even though I never considered them that. And moved to industries. One of the areas, I think, of physics thatís of benefit to industry. More than significant -- relatively high. And I would say those who did not move to industry, almost all had consulting jobs with industry.
Hoddeson:Oh, they were doing that.
Aigrain:And those who joined the lab after 1960, which have been very important, Andre Libchaber who had been trained at MIT. He was a Frenchman, telecommunications engineer, got trained at MIT, joined the lab, but started his own line, on connective phenomena in liquid helium, which has now become one of the world leaders. So, we also tried, you know, to leave room for young groups of people who looked brilliant and seemed to get an interesting subject.
Hoddeson:In 1955 or Ď56, you started, together with André Guinier and Jacques Friedel the "diplôme díétudes approfondies de physique des solides". We would be interested in some background on that. Because --weíre interested in the gradual growth of solid state as a subfield.
Aigrain:All right. Now, in 1954, Jacques Friedel who graduated from the Ecole Polytechnique and the Ecole des Mines and had taken advantage of the possibility of [???]to be sent for two years to do research somewhere. He had spent his two years, possibly more, in Bristol with Nevill Mott and Jacques Friedel is a very brilliant physicist, and he came back as a very brilliant solid state physicist, mostly interested in the physics of metals. And at that time, Rocard and myself supported him to get a job at the University of Paris, where he got started his laboratory at Orsay which at that time was under the University of Paris. It was not yet an independent center itself. We very quickly became very good friends, and we have exchanged students. People like de Gennes started in my lab and went to his lab and Parodi and several others, so it was very quickly a friendly cooperation. The second thing, that in 1954, was established what we call the "troisième cycle". It is a graduate cycle of university studies which had not existed in France. The only thing that existed then was a doctorate system which included no courses, students just did research, for any length of time. Sometimes it was fast and you were lucky, as in semiconductors, I was very fast, but sometimes it took ten years. And that was it and you got your degree. It is Mendès France (and Gostan Berger was director of higher education at that time), who decided to set up something more like the graduate system of American universities, and that included courses of a more specialized nature than what was given at the BS level, and on more modern subjects. And one of the first ones of their courses to be established was established in cooperation with Jacques Friedel and myself, who were co-directors of a course of solid state "diplôme díétudes approfondies". Since Paris and Orsay separated and Orsay became a university of its own standing, and then later Paris split in Paris 6 and Paris 7, and the Ecole Normale was always involved in this. This system went on. Itís still common undertaking of all these universities who are still teaching together these courses. Now, of course Friedel taught, I taught and many other people taught in that system, and it was for a long time by far the most important source of training for solid state physicists in France. Another thing that was very important, I must add, is the Ecole des Houches of theoretical physics, which had been established -- I donít know when, when was the first? Ď53? by Cécile DeWitt. She is now a professor at the University of Texas. Who had the idea to set up a -- it was the first summer school of the world. There had been the Gordon conferences but there were no schools. Here, it was professors from the world over teaching a select group of students from the world over. It was supported by NATO and by the minister of education in France.
Hoddeson:Did she gather all this funding?
Aigrain:She gathered all this funding. She got all the support. She managed to find some old buildings in one of the most beautiful places in France, in Les Houches facing the Mont Blanc.
Aigrain:Physicists are often mountaineers, so everybody was tempted to come there. And she really started the high level teaching of theoretical physics, and that was theoretical physics in general, but depending on the year, and there are several sessions in a year, many of them were on theoretical solid state physics. And the influence of the Ecole des Houches not only on solid state, but on French modern physics in general, has been tremendous.
Hoddeson:Not just French.
Aigrain:Not just French. Well, still I would say half the students were French. But students and professors were from the world over.
Hoddeson:Right (crosstalk) Proceedings -- all over the world --
Aigrain:-- tremendous -- and especially Iím happy that one of the last things I could do as a minister of research was to grant a decoration to Cécile DeWitt, who now lives in the United States, but was not been forgotten, by everybody, in France.
Hoddeson:When did the term solid state start to be used in institutions in France? Probably after the war.
Aigrain:I think itís probably right on, 1948-49.
Aigrain:-- you know, really, at that time, the word already existed in the United States.
Hoddeson:The American Physical Society set up the division of solid state I think in í47, orĎ48. So it was just --
Aigrain:You know the start was based on transfer of science and technology from the United States, and also from England in the case of Friedel.
Dobrzynski:People like Curien, Abragam, Castaing, were doing related things?
Aigrain:They were doing related things. But you know, we always had a strong school of crystallographers and radio crystallogaphers. That was very strong with people like Guinier people like Laval; there was a strong tradition there. Mineralogy also. So these people would say I belong to a different branch, very, very important, very good, that had started pre-war. Of course Curien is of about my age, they were both born in 1924, so their work was after the war, but it was in relation with Laval and Guinier.
Hoddeson:Then of course you have a very strong tradition in magnetism.
Aigrain:Magnetism dates back to Pierre Weiss.
Hoddeson:(crosstalk) -- early part of the century -- Langevin.
Aigrain:Yes I think the big boom has been the molecular field theory by Pierre Weiss around 1920 in Strasbourg. And people like Néel and so on were students of Pierre Weiss. So magnetism is a tradition and was very, very well established pre-war.
Hoddeson:I guess the work on dislocation and defeats came later.
Aigrain:Well, the work on dislocation and defects had two sources -- first, the radio crystallography aspect, with Guinier and then Friedel.
Dobrzynski:The lattice dynamics was a subject of interest to Laval?
Aigrain:Laval essentia1ly. Laval of course, very interesting character, by the way, you know, he couldnít read at the age of 14.
Aigrain:Yes. He was a goatherd, and [???] and itís only because a new pioneer school teacher appeared in the village, and noticed that this boy was not going to school, and forced him to school. So ten years later he was doctor of science. And then Professor at the college of France. So he was really brilliant. His thesis had been the Laval effect of thermally diffuse scattering of X rays connected with the lattice dynamics, and so.
Dobrzynski:But you donít consider him as being on the solid state physics community?
Aigrain:Oh yes. Heís definitely. You canít interview Laval because he is dead, thatís all. You can interview Curien.
Dobrzynski:But the point is that in the projects now there are ten subjects about solid state physics, and one of the questions that arise is, should there be another point on lattice dynamics?
Aigrain:Well, -- see, lattice dynamics has been a subject itself, and at the same time, you find it everywhere in all the rest of solid state physics, because obviously, you can do -- you canít do semiconductors without speaking about phonons. You canít do ferroelectrics without speaking about soft Medes. And so on. So you find it permeates the whole of solid state physics to a large extent. You find people working on lattice dynamics practically all the time. You also find lattice dynamics which is used in tellurium because of its very strong electron-phonon coupling, because of its strong piezoelectricity. So of course, Piezoelectricity is in the French tradition, since Pierre Curie.
Hoddeson:Then there are also these isolated developments in the history of French solid state physics. The work of Brillouin. Now, there --
Hoddeson:There I know nothing. Did he work in a vacuum or did he?
Aigrain:He worked in a very isolated way. His book on wave propagation, in periodic structures, is a wonder. ItĎs really an astonishing book, because it was written in 1924 or 1923. But even so he had a lot of prestige, but he didnít have any students. But this is a little problem with the College of France. The College of France is a very special institution, itís not degree granting. The professors give 20 lectures a year, but they donít grant any degrees. So, some of them didnít have any students. But he had a tremendous indirect impact.
Hoddeson:Well, Iím interested in putting together the development of the quantum theory Of solids now, which mostly took place in Germany around, Munich, Leipzig and Zurich, though it had important contributions from people in other centers, such as Landau and Frankl and Wilson and Brillouin.
Aigrain:And Brillouin and so on.
Hoddeson:And one wonders how the contacts were established. Itís hard because one canít interview most of those people.
Aigrain:Most of those people are dead.
Hoddeson:I mean Peierls and Bethe who did some work at the end of their lives -- I mean, there are probably ways of finding out but Brillouin is somebody I don Ďt quite know how to get information about yet. How he got started, and did this wonderful work that laid the basis for Brillouin zone theory.
Aigrain:Well, you know, his father was already professor at the College de France, Marcel Brillouin and his father worked on wave phenomena all his life. So itís quite possible he got interested there. I understand his first interest was not so much in three dimensional structures. He started working on what is called in electric filter theory, that is, periodic L-C structures of various kinds, which are used as pass bands, stop bands and so on, because they show of course a band structure as a function of frequency. And they have been used for -- filter theory started developing before the First World War.
Hoddeson:Thatís very interesting.
Aigrain:And of course they are periodic structures, in which what propagates is an electrical signal and you immediately find -- of course, in one dimension the Brillouin zones they are just segments -- so he got interested in that, and then moved to two dimensions and three dimensional structures.
Dobrzynski:Gérald Leman told me that Blanc Lapierre had some connection with Brillouin
Aigrain:Thatís quite possible. And Blanc Lapierre would be interesting for an interview. Heís not a solid state physicist. His own line has been mostly noise theory, stochastic processes and so on. But he may have known Brillouin well enough. I never Brillouin because you know, he moved to the United States during the war, just before the war, and never came back. Now, he was working at IBM. So itís possible to find some people at IBM who have known him quite well.
Hoddeson:Sometimes in an industrial context, they make people keep notebooks. Bell Laboratories is very easy. It was easy to trace the history of the transistor because of the necessity of laboratory notebooks.
Aigrain:Well, I doubt if youíll find any notebooks of Leon Brillouin because you know Leon Brillouin left France during the war, and finally resigned from, the College of France, and resigned I think about in 1949, because it was at that time that Laval succeeded him.
Hoddeson:Oh, Laval succeeded him?
Aigrain:Yes. That chair was established in 1801, for a man named Billault.
Hoddeson:Yes, of course.
Aigrain:Who was at that time 26? And at that time --
Aigrain:1801. At that time, professors at the College of France never retired. They could stay in office until they died. He died at the age of 87. So he held his chair for 61 years. His successor was a man named Bertrand who had been his assistant, who was already rather old when he succeeded him, and who stayed until 1902, died at the age of 90 years, still in office. In 1902. So during the first 100 years of this chair, there were two holders. The next one was Marcel Brillouin who held the chair from 1902 to 1932, because by that time they had introduced retirement. Otherwise, Marcel Brillouin -- he only died in 1948 -- could have stayed 16 years more. Leon Brillouin succeeded him, and resigned after only 17 or 18 years, but still Laval in 1950 was only the fifth holder of the chair.
Dobrzynski:Is there some scientific connection?
Aigrain:No, when Laval was nominated, he had been assistant, as I said, mostly helping him with preparations of the courses, -- Laval liked the whole equations to be written down. He was against the Einstein convention for compressing tensor, so he explicated all the subsequent simple-states of the tensor. That took an awfully long time, to prepare. But at the same time, there was a lab attached to it, except this lab was -- there was a lab being build, and there was an old lab in the old buildings of the College of France which apparently, which nobody had gotten in, and I was asked to take care of cleaning up things and so on, and when I saw the mess which had been left -- it was kind of, in fact, ghost of the early Marcel Brillouinís experiment was still rotting on the spot. So I doubt whether Marcel or Leon ever kept a notebook, because he was certainly the most orderly man there.
Hoddeson:Tell me, at this point; I think that perhaps one should save a discussion of your work in science administration for another discussion, because this is now --
Aigrain:Well, yes, thatís an unfortunate thing, I mean, as a doctor, you have done science, and somebody asks you to do science administration.
Hoddeson:Why unfortunate? You made a big impact through administration. But itís a question of whether you would like to go on now or --
Aigrain:Well, listen, now, because of the special circumstances we are in, I would like to keep in touch and find out, after the new government has been appointed.
Aigrain:Because I may have some motions to go through as a result.
Hoddeson:Yes. This is an excellent interview.
Aigrain:Now, one point I would like to make -- I was trained as an electronic circuit engineer. Most of my work was in semiconductor physics. But I always kept up with electronic circuits, and I was always active in applied electronics during that period, and Iíve always found that it was very stimulating. It was sometimes fun.
Hoddeson:Did you notice the time?
Dobrzynski:I have just one questionÖ
Hoddeson:Weíll turn the tape recorder off.