Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of John Whinnery by Joan Bromberg on 1986 August 18, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/4959
For multiple citations, "AIP" is the preferred abbreviation for the location.
Whinnery trained many prominent maser and laser scientists in his University of California, Berkeley Electrical Engineering laboratories. This interview touches briefly on the functioning of the Electrical Engineering department from about 1950 to the early 1970s, Whinnery’s own research into laser subjects like thermal lensing, his relation to National Science Foundation funders, and his students and their projects.
We want to start with the early years and the effect of the maser as it became discovered, and your reactions to it. Of course, you were very much involved from the beginning, weren't you, with the submillimeter waves and the drive to generate submillimeter waves.
Well, we were working with microwaves. Of course, my career at GE was in microwaves, and then here, and at Hughes, and the first I heard of the maser principle was in a classified conference on millimeter waves at the University of Illinois. I thought it was Sevel who made the report for Townes' group, but according to the history that I read, it seems to be Nethercot. In any event, the description of the idea was made there, and most all of the other proposals for millimeter wave generation were electronic, higher voltage electron beams, things of that nature, and I guess it would be nice to say that it caused a sensation.
I was wondering, what did it do?
Well, I think most people thought it interesting, but were a little skeptical about its working and being practical, and I think it was — was it the same year or later? — at what was then called the Electron Tube Conference, up in Canada, Ottawa, I think it was, that —
Weber, Joseph Weber —
Yes, Weber's proposal for, again, the inverted system. I think Rudy Kompfner was one that said something about, "You know, that's an exciting idea," but he went on to say later that by the time he got home, he forgot about it and really didn't think much more about it. Again, the reaction in general was, yes, it's interesting, but nobody quite saw how to do it, and they didn't immediately rise up and say, "Oh, this is the solution to all of our problems." So I wasn't aware of much more, until the formal announcement of the Gordon, Zeiger and Townes maser actually operating.
What kind of impression did that make on people?
Well, that certainly made a big impression. Nearly everyone recognized that this was something new, and in our department, Sam Silver, who had studied solid state under Slater and then had been in microwaves at the MIT Radiation Lab, was the one who picked it up immediately and said, "Oh, we must do some work in that field." He was interested in what you could do in solids at that time. I don't know that he had any specific invention, nothing like Bloembergen's three—level maser, but he believed, and others around the country likewise were saying, "There must be some processes in solids that you could invert."
Was he a professor in the engineering department?
That is correct. Yes.
And also was he in the Electronics Research?
Yes. The Electronics Research Lab, I notice you have some questions about that later. Would you like to go into that now?
Well, you know, the reason I ask that is, this immediately reminds me, this case of Silver immediately sounds a little bit like Strandberg at the other end of the country, who was in both physics and the Research Laboratory of Electronics and interested in the possibility of developing this in solids, so I thought that was an interesting case.
Sure. Let's talk about that a little bit. Let's see, Silver after the war went to the Naval Research Laboratory, and then came here a year or so afterwards, so it would be 1947 or so that he came. He gave courses both in physics and EE, but was always in this department. His title was Professor of Engineering Science. He was one of the tremendous influences in this department. His research, for the first several years, was continued in microwave antennas, well really very fundamental diffraction problems, and existence proofs, things of that nature, following up work he had done at MIT. But he gave courses in solid state, which at that time was new. Of course we had very excellent solid state people in physics, but in this department there was nothing until then. At that time, the research projects were uncorrelated. We had several in the department, primarily three big ones, the one on antennas, funded by the Navy; one in microwave tubes, funded by the Air Force; and one in computers; it may have been a little bit later that Paul Morton was starting that project. But they were not correlated and somewhat competitive, and it was Silver who argued that we should have a common laboratory and common policies on all of these. At that time I think the MIT Laboratory for Electronics was in existence, and he saw that as somewhat of a model. Maybe Stanford's, I think Stanford's perhaps was also in existence.
Yes, it was in existence by 1950.
Well, in any event, he headed a committee and made this proposal, and it was accepted by our dean, M. P. O'Brien. Silver could have been the first director, but he chose not to be. That was a period when I was on leave at the Hughes Aircraft Company. I wasn't quite sure I was going to come back, I enjoyed the work there so much. But he came down and spent quite a lot of time trying to convince me that it would make a difference if I would come back to head the laboratory. And eventually I did. It wasn't an easy decision, because things were going extremely well there and I was enjoying it. So that would have been 1953, I guess.
I guess you were at Hughes '51, '52.
Yes, I guess so. I have to look it up again.
I have '51, '52, the microwave tube section, Hughes. 
Yes, I guess it would be '51 then, 1952. And so it was during that period. Of course Silver was still active primarily in electromagnetic theory problems with emphasis on radiation, but still very much interested in broadening not only his work but that of the laboratory itself.
When did you physically set up the Electronics Laboratory?
Well, let's see, at that stage we had moved into this building. We moved in here in 1950. And as it is now, there were several different laboratories, with different emphasis. The antenna laboratory I think was on the third floor. I'm not sure; it had been in several locations. The computer laboratory was definitely on the third floor, on the north side, and the microwave tube laboratory was on the second floor on the east side. By that time there were beginning to be a number of other research specialties, particularly circuit research, and I don't remember exactly where but that was in a relatively small room. So it was throughout the building. Perhaps I should make clear that this laboratory, perhaps more than some others, was really an instrument for just the research that we saw going on in the department, and more a mechanism for having common policies, helping with proposals, and with the administration of these.
And now the funding began to be Joint Services?
Joint Services came somewhat later. Let's say that came in 1961.
Oh, that's very late.
Yes but we were one of the first to be added to the group started during the war. So we had funding from a number of agencies, but even though it came from different agencies, the policies, for example, on research assistant appointments, would be common. In the early stages, each group would try to outbid others on salaries, which was part of what Silver objected to.
I see, because I usually think of laboratories like Columbia Radiation Laboratory or RLE or the Stanford as being laboratories that are essentially on one contract, paid for by the three services but funneled through maybe an ONR contract or what have you. It did not work that way in this?
No, but I think that's not true of those any more. It was in the early stages, that the Joint Services was the primary support, but I think in every one of these now, they have to have many sources of support.
And so you started out with many sources.
Many, and then we had the Joint Services, which was the dominant one for a while, but now again, it is a very important one, but we still have to have many sources of support.
I see. What was the relationship between the Electrical Engineering Department, as a department, and the Electronics Research Laboratory?
Well, first of all, it's a very close relationship. The research in the Electronics Research Laboratory was really that which the faculty wanted to do, and of course could get support for. We had a few permanent research people, but I think, in contrast to some of the other laboratories, much fewer.
And all the research that was done through the department would be done through the Electronics Research?
Most of it. It wasn't required. We had one or two faculty members that, for reasons of their own, either they didn't get along with the director, or they disapproved of the fact that it was supported by Defense funds, went on their own, and that was perfectly all right. But then of course they didn't have the support structure that it supplied. But they weren't made to go through it. I would say there were maybe two exceptions out of a faculty then of 40 or 50.
I see. So getting back to Silver a little bit, did he inaugurate much of a program in solid state lasers, in response to the Gordon, Zeiger, Townes?
He made some experiments, but as far as I know, nothing came out of them that was publishable. I believe the results were negative. I was following it with some interest, but not in any tremendous detail. And the way I got into it was through a graduate student, if it's time to talk about that. At that time my research was essentially all on microwave tubes — traveling wave tubes and related tubes — and I had a number of very good students. We had a marvelous person, an engineer, who built the tubes, George Becker, and one of the very bright students who had just finished a Master's project on a velocity analyzer for electron beams—at that time we required theses for the Master's degree as well as for the Ph.D. — was Amnon Yariv. He had come to our group from control theory, because he wanted something much closer to physics. Until talking with him recently, I didn't quite appreciate how much was affected by the conference at Boulder in selecting his next research direction.
Would it be the one where Feher and Gordon [reported on the] solid state [maser]?
Yes, and heard them talk about it, and got tremendously excited in that field. I didn't remember exactly where it started, but he wanted something really more physical for his Ph.D. than even the tube work that we were doing, and we talked about some possibilities, including the solid state maser. Of course, I knew I didn't have the background to supervise that alone, but first of all, George Feher talked to him. George Feher was at Bell Laboratories at that time. He talked with him and gave him some suggestions, including the adiabatic fast passage technique, which Amnon did use, and Alan Portis, in our physics department, also gave him suggestions and promised to help if he got started on it.
And Silver, was he involved?
Not directly. He was interested in it and certainly supportive, but thinking back, I'm not entirely sure why he didn't get more directly involved. Maybe he saw this as a little different from his ideas.
Because I somehow got the impression from Yariv that no one in the engineering department, in the EE department here, really knew much about masers at that time, and you're giving me a much more nuanced feeling, that there was interest and there was thought about it.
Well, there certainly was interest. But it is puzzling. Certainly there was no problem between Silver and us. We remained good friends all the time. But it may be that he was involved with something else at that stage. But the person who did come into it was Jay Singer, of course, who was appointed some time during the period after Amnon had started on this, and before he finished. I'd have to look up the exact dates. And also Shyh Wang, I believe, came during that period. Certainly by that time I had left heading the Research Lab and was department chairman. Some of this recruiting was done while I was department chairman.
So Singer was recruited to electrical engineering?
I see, and Wang was also.
And Wang was also.
Now, were you still taking a friendly interest in what went on at Hughes? Because they were doing considerable.
Let's see, I was consulting there for a while after I left. I probably was still consulting at that stage, it sort of attenuated over a period of time, but at that stage, I was.
Do you remember being in contact with Lyons or Birnbaum?
No. The group I was consulting with was still the microwave tube group, so I didn't really have association with the quantum electronics people at Hughes, other than through meetings and so on.
So it would have been more sort of peripheral.
Yes, that's right.
And then of course there were some people closer to home. There was some work at Varian, I guess. There was some work starting at Stanford. I know that Stanford started solid state work, maser work in late '56. Did people here get pretty close to the Stanford electronics laboratory?
Yes. Of course, as in anything, it depends on individuals.
How about you?
I'd have to look at the dates, but — whom did you mention?
Well, I didn't, but Heffner, I think.
Yes, Heffner and Pantell and of course Tony Siegman became very much interested in it, and certainly we had interactions with all of these people. I guess the thing that Amnon was doing was different enough from what Stanford was doing that it was somewhat of interest to them, and what they were doing was somewhat of interest to us, but it wasn't a matter of close collaboration.
Was it a matter of keeping each other informed in seminars etc.?
Yes. And a lot of it occurred through the technical meetings and research reviews. We each had research reviews.
That's like a kind of quarterly report or what is it?
Yes, at that time everybody wrote reports — some contracts had monthly reports and some quarterly, but nearly every contract had regular reporting. Also, Stanford had what they called a TAC meeting, Technical Advisory Committee meeting, every year, which was very well attended. We had one which, at that stage, wasn't nearly so large.
I see. That's an interesting thing which I didn't know about. In other words, a lot of members of the technical community would come yearly and sort of listen to your reports on your work?
Yes. The Stanford one for a number of years grew into a huge meeting, literally hundreds of people. It started out as a report on their Joint Services Project, but they broadened it to include the whole program, and at first it was just I think for the military people supporting it, but eventually, they had people from companies and other universities, hundreds of people.
That's interesting to me, just as a part of the social apparatus and the way the work is done. It's something I didn't know about. I think of people learning about each other's work through journals and so on, but for me this is a new arena, in which —
I think it was a very important one for a number of years. I think now, there are so many other technical meetings that even though most of us have some review meetings, they're pretty much now back again to where the sponsors are the main ones to come. But at one stage, it was a tremendously effective and well attended part of the reporting procedure.
And if a researcher, for example, wanted to do a blow by blow study of maser work here, they would use these yellow reports that you've got there, the progress reports?
Yes. That would give a summary certainly of what was going on. The reason I have some of these out, the Joint Services Program is having its 40th anniversary celebration in September, September 25, I believe, in Washington, and as one of the speakers on this, I've been going back, trying to remember what was done when under the Joint Services Program.
I think I vaguely remember that. That's probably going to be an interesting meeting, probably some interesting history.
Townes and Bloembergen are speaking.
As we talk, I also like to get a feeling for where the paper documentation is that people might want to see. Now, something that I wonder about is, the maser comes in and you get a whole bunch of noise issues arising with the maser, like the lower limit of noise, and what to do about noise when you have negative resistance, and stuff like that, and you, in the microtube business you were very much interested in noise, and I was wondering if you have any memories about the way in which maser noise problems intersected with the noise problems you were already interested in, or whether that was a big effect, or whether that was just a little thing?
Well, certainly at that time, the primary interest of the maser was probably its low noise characteristic. The other important characteristic was the very high stability, and as it's turned out, those two properties of masers are the things that are still used, although much more specialized than we saw at that time. The problems of analysis of noise were really quite different. I did do quite a lot of work on the physics of what was happening in fluctuation phenomenon near the potential minimum of electron tubes, and in the sense that everything is a statistical fluctuation, of course, there's a lot in common, but the process of what happens to it is quite different. So I'd say the interest from the low noise point of view was certainly there.
That was a practical interest, or theoretical interest, or some of each?
Practical, but I didn't do anything in the analysis of noise for masers, other than listening to what others had said. Perhaps I should go back on the work with Yariv. I guess I didn't see myself as moving into this field at that time, so much as helping Amnon get started in it; having these people who were expert and helpful, Feher and Portis and Singer and Wang, coming along, that part of it, I thought, was taken care of. I did try to follow it in detail, but moving into administrative things more and more I guess he finished before I became dean of the college, let's see, when was his degree? Yes, 1959, so he finished just before I became dean. But anyway, as department chairman, I was getting more and more involved in the administrative end of things. So as department chairman, we were trying to hire some people in this field, and we did see it as an important field, and I saw it as a way of getting him started and associated with the people who could help. But probably at that stage, I didn't see that as a sudden shift to this as a career. But getting back to your question about the impact it had on engineering education, I think that certainly it had an impact here, not only in our trying to find faculty who could be expert in this field, but — we didn't start courses called quantum electronics at that stage — but certainly in bringing into the courses some of the concepts of the maser principle.
I was interested in your articles on education, that as far as I understand what you said, that the attempt was to bring more science into electrical engineering education, and at the same time to bring more diversification into it, and I was wondering if I should read that into the fifties? You say at one point that also some new things are being added, biological science, people going into that phase of engineering and so on. I was wondering if quantum mechanics was brought in as one of these options, or whether it was brought into the very core, if that makes any sense to you?
Well, I think the main shift to a more scientific base for engineering education really occurred during World War II, and it was a combination of things, with all of the new things that happened during World War II, and the fact that engineers and physicists worked together on such things as radar, and I guess some of them on the atomic bomb. Until then, the doctorate for engineers was fairly uncommon. MIT of course had a good program, Stanford, Caltech, a couple of others, but if you look at the numbers of Ph.D.s in engineering, it was extremely small. But the engineers who had to work with Ph.D.s from other fields began to see the advantage of this kind of an education, so right after World War II — and it's amazing to me that it happened so suddenly — the enrollment for graduate degrees in engineering schools jumped way up, and continued increasing for several years. And with it, the understanding that a scientific base, rather than just a certain handbook approach was necessary for an understanding of all the new developments.
I was interested in the fact that Yariv has told me that when he got interested in this maser thesis, he then went over to the physics department and took a year's worth of courses to give himself the physics background, and I wondered whether two years after that or four years after that, an electrical engineer in this department wouldn't have had to do that, whether you were bringing in the quantum mechanics or whatever, or whether they would always have had to do that?
I think both things have happened. First of all, we have much more in the department than before. We have quantum electronics courses and we have some very good solid state electronics courses that are given by persons who understand the fundamentals well. Shyh Wang certainly knows the physics of it. But at the same time, we encourage our students to take the courses in physics too. Of my graduate students, I guess all have taken a number of physics courses, including the final graduate quantum mechanics course in physics.
Here's a phrase in one of your articles that really intrigued me, something like, how these changes occurred in the course of many protracted battles of the faculty curriculum committees, and it occurred to me, it would be really interesting to know something about the details of these battles at Berkeley, and maybe even some of the documents in which they're recorded, because that would give you a kind of a case history of how engineering education was changing, and some of the arguments pro and con in the fifties, and I think that would be an interesting body of material for those people who are doing the history of electrical engineering.
Yes. Well, I don't know, coming to the documentation, whether anybody could find what was written about it in the minutes of these meetings, and of course not all of the most interesting part got put down. It got edited first. Every change of course is difficult, because you have to leave out something, and whatever you leave out, somebody sees as terribly important. We had a wonderful group of the persons, you can call it the earlier generation, that had built up the department, but they didn't necessarily agree that all the changes that were being made were a good idea. And you know, coming back to such things as cutting down on some of the machinery courses, when power again became important, one could argue that they were right. But we never did give those up completely. It was a matter of trying to see as best we could the priorities of the moment. But the great thing about our senior colleague is that, although we had a lot of arguments in the meetings, whenever it was decided, everybody worked together to make the new thing work. We'd get a pretty heated meeting, and yet we'd go off and have coffee together afterwards and decide what to do next. So yes, there were some very very tough curriculum decisions.
And then, it sounds as you explain it, the attempt to understand what the shape of engineering to come was going to be —
That's right. And if you go back to when I was getting my bachelor's degree, there were circuits and machines and electronics. Basic electronics was just coming in, but it was relatively simple. But as we were starting to build in the solid state area, there was also the plasma area, systems, control theory, computers — so diversification was part of the problem. And in addition to the need for physics, a much deeper understanding of mathematics was needed than people had had in the earlier days. When I was in school, if you knew about Bessel functions, you were considered to be pretty sophisticated, but that's not at all the level of mathematics that our systems theory people have to work with now.
Then how big an impact should I be thinking in terms of quantum electronics, in terms of, all this new stuff is coming in. Is quantum electronics just one small part of it, or is quantum electronics an enormous phenomenon, or how should, you know, somebody who doesn't know, how should I weight what's going on in electrical engineering with respect to this one field of masers and lasers?
Well, of course, that's difficult. I think you'd have to say that if the maser had not gone on to the laser, and nothing else had happened, then it was way overweighted, just by virtue of the fact that it was exciting to a lot of people.
So it's a little hard to separate out whether people were interested because it was new and exciting, or because they saw the potential. Well, it was both.
It's interesting that they're both in there.
But it turns out that, and I think this is correct, that the number of masers built per year and sold now aren't very man, that they're for specialized purposes, radio astronomy, low noise purposes, atomic clocks, and so on. So if that had been all that had happened, it would have been overrated. It was like the interest in solid state before the transistor made it obvious that it was something that everybody had to be concerned with in electronics. How much it's vision and faith, and how much it's jumping onto something new because it's interesting, is a little hard to unravel. But I guess that's the way we do make progress, by looking at things with new potential, and the fact that they're interesting helps to get good people involved.
That again is another new thing for me. You just said that there was kind of more excitement about the maser than it potentially panned out, just as a maser.
As a maser. Yes.
I didn't realize that. And on the other hand, probably the reverse was true on the laser, because it probably panned out beyond anyone's real —
Yes, I think that's right. During all this period, of course, there were the discussions of moving up to higher frequencies using terms like IRASER for infra-red and UVASER for UV. It would be a little hard to say how many people had enough vision to see what they would really be useful for, and how much was just that it would be interesting to do. But I think there had to be a large element of the letter.
Well, so, now comes the time when you yourself personally start to get interested in laser research.
Yes. It really happened after I finished my appointment as dean of the college, and that incidentally was a relatively short appointment. I'd been asked to take the appointment for "two to three years" and I did it for four, and fortunately had the original letter that said two to three years. One had to decide at that stage whether to be an administrator the rest of one's life or get back to being a faculty member, and it was fairly clear in my mind that I wanted the latter. So fortunately, I took a leave and took it at the Bell Laboratories. I think if I hadn't taken the leave and had just gone back to teaching, the busy work of every day would have made it very hard to have caught up with anything, and even if I'd taken a sabbatical, and tried to study on my own, it wouldn't have been nearly as effective. The group at Bell Laboratories was an excellent group, and were also people who, in large measure, had changed from microwaves to lasers, as I was trying to do. So they knew all of the pitfalls and problems and how to explain things in my terms. These were people like Kompfner, Cutler, Ashkin, Tien, Louisell, and then of course there were other excellent people. I worked with, Jim Gordon and Porto and Leite who had come through the physics route. So it was a wonderful mix. At that time. I had three former graduate students at Bell — no, Yariv had left, but Marty Pollack and Billy Kluver were there. They also had started in microwaves, and Marty was really just starting at Bell but each made contributions to lasers in a relatively short time.
But what you just said was that you were already thinking of going to lasers, even as you were coming there. Is that what you meant?
I didn't really say, I'm going to go into lasers, so much as, I wanted a place where I knew good work was going on. So I think that the place came first and the subject second. But it was really together, I guess.
Anyway you certainly chose a group that was working in that.
I'm assuming that the thermal lens work comes out of an interaction with that group.
That's right. The story on that's rather interesting. When I went there, and talked with John Pierce, and of course Cutler and Kompfner, the problems that were the most immediate from their point of view were transmission of the light waves and modulation. At that time fibers were just so [high] in loss that even though it was known you could guide light with fibers, they were ruled out; periodic lens systems seemed more likely. These were real lenses which were placed every ten meters or so. And the problem of modulating semiconductor lasers, which we now know can be modulated very easily, had been pretty much ruled out because they had a very poor spectrum. They also had to be cooled, at that point, cryogenically, and they didn't last very long. They knew there'd be some improvement in all of these, but the fact that the fibers could be improved as much as they turned out to be, and also the semiconductor lasers, are two of the great accomplishments in optical communications. But anyway, in modulation I worked with Art Ashkin. We looked at a number of things which were quite interesting. Although nothing developed for that period. Franz-Keldysh effects were one, and these are still being investigated. On the propagation issue, we were looking at gas lenses, which later Bell did quite a lot of work with.
As it turned out, that was not the way to go, but there was some nice work with P.L. Tien and Jim Gordon. I did try to find out what others were doing, and got to know Leite and Porto quite well. They pointed out these strange fluctuations they were getting when they put cells in a helium neon laser looking at Raman shifts. They said, "Well, it looks like a thermal effect, but we've looked at that and there's just not enough loss in these materials to explain it." So I worked with them and with Jim Gordon in trying to figure it out. We looked at a lot of phenomena, at first ruling out the thermal effect. Carbon tetrachloride had loss in that range of about 10 to the minus 4 per centimeter. We had a centimeter. It seemed impossible that this small loss could be causing the major effects. So the paper we wrote on that is a rather funny paper, because we talk about all the things that might explain the effects because we were still not 100 percent sure it was thermal, even though by that point we had analyzed the thermal effects and found them possibilities. Finally Jim Gordon went down and made some experiments of his own, and came back convinced that it had to be thermal because of the time constants. I sat down and made some analyses, with his and Louisell's help, and it became evident that when you realized the size of the gradients, it was possible and all of the time constants seemed to be right. So that's the way that happened. It was one of those observations where the obvious was dismissed because the magnitude seemed wrong until a careful analysis was made.
At this point, you came back and set up the laser laboratory, is that true?
Yes. Wang and Singer had really done the quantum electronics work in the period between. Singer had a couple of very excellent students, Alan Pine and Larry Lin, but he was moving back into the bioelectronics area, in which he still is doing excellent work, and Shyh Wang of course was always interested in solid state, of which quantum devices were just one aspect. At that stage he was doing nice work in magnetic effects. So the laboratory did exist, but it really was down to about two students, and so before I came back, in discussions with — let me see, after I left the Electronics Research Lab, Sam Silver became director, then Don Pederson, followed by Angelakos, maybe it was Angelakos — I don't know, one of those, in conversations with them, we agreed that we wanted to build that up, and the department also was recruiting another person. That's where Steve Schwarz came into the department, from Caltech. And George Becker, who had been our marvelous person in the tube field, moved over. We brought him back to Bell Labs while I was there to talk to the people there about making lasers, and all of our first lasers were ones that he made.
So now you had this kind of basic lab. I remember that Mike Bass told me for example, that at one point he came in — no, he must have come into the physics department.
So the physics people must have been building up a laser lab almost parallel to your building up one in the middle sixties.
Let's see. Of course, Townes came back, but emphasized astrophysics.
That must have been Hahn and Bass and —
I see. Well, Hahn certainly, made fine contributions, I think he didn't think of it so much as doing research on lasers. Lasers were a tool, and he did some excellent work in nonlinear optics and photon echo effects, things of that nature. Sumner Davis was also interested in lasers for spectroscopy and worked with us to get Bloembergen and Ed O'Neill as visitors. He also worked to get Ron Shen as a permanent member of Physics.
I see, so this actually should be scratched, because I think what went on there was a slight confusion in my head which wasn't very helpful for telling the story. I do want to know how the lab was built up here, and I suddenly got mixed up with what was happening in physics.
I think our relations with physics have always been very good. They are probably not as close as we would all like them, we all get so busy, but there's never been any problem. We have joint seminars. We trade equipment, and certainly have tremendous respect for all of their the work. Now it's Ron Shen, Peter Yu and Roger Falcone that we interact with primarily.
So how did you decide what was the major research directions you wanted to follow? Of course the thermal lens is clearly a research direction you continued along. And I did want to ask whether the final achievement, or the beginning of an achievement, of low loss fibers in the late sixties began to make you deviate from that.
No, not at that point. Let's see, how did it happen? After I got back, the first thing I did was to start a course which, looking back on it, was a pretty rudimentary course. I didn't really know enough about it. But I had learned from the excellent people at Bell Labs what some of the problems were, and where some of the literature was, and that was probably the main thing I got across. I had a marvelous group of students, you know, people like Erich Ippen, Bill Clark, Obert Wood and Marvin Klein. I made them do quite a lot of the work in each of the areas that were then just developing. They wrote papers and in many cases those are the subjects on which they turned out to do their theses. I have the list of those papers from that first course, that might be interesting to look at some time.
Yes, it would be interesting.
But I was fairly clear that I wanted to continue work on the thermal lens affect, and on modulation. I guess Erich Ippen's work grew out of the modulation goal. We started off with electro-optic deflection, and then interaction with surface acoustic waves.
So when you choose these topics, you just choose things that look as if they're —
We tried at least to follow the literature, and see what we thought was an interesting problem that we could do something with. But I must say that there was chance, and trial and error. We discussed a number of problems, before settling on ones that we thought that we could do something with, and which had students' interest. I guess I always let the students have more say in what finally comes out than some people do. I know some of my colleagues talk about "assigning" a Ph.D. project; that always startles me, because I never assign them. We discuss. And in many many cases, it's the student's ideas that come to the fore.
Now, you know that I have this feeling that engineers in lasers do slightly different things than physicists in lasers. Is that just a lot of nonsense? Or do you think the problems you were selecting, for example, in this department, might have been —
No, it's not nonsense. In some areas, it isn't as different as some people might think. You might be doing, for a while, pretty much the same thing, but the motivation is very different. Let's take the area of short pulse work, that we worked on for quite a while. Presumably we were interested in short pulses because we hoped to have a generator that you can do something with in communications or information processing. The physicist would be interested in them in two respects — one is what they tell about the inner workings of the transition, time constants and other fundamentals, but also because they want to use them to do some dynamic spectroscopy. Now, we are interested in the fundamentals too. It isn't just, does this work? But why does it work? And what implications does it have? So in this intersecting area, where both are trying to get shorter pulses as happened in the mode — locked dye lasers for quite a while, the physicist and the engineers might be doing very much the same thing. This is where we'd exchange a lot of information and equipment. But the goal is certainly different.
I wonder if some historian someday wanted to really do a detailed examination of some cases in which decisions were made on what piece of research to do next, where would he or she have to look, just to the memories of the participants, or do these things get written up in any way?
Well, the proposals, of course, if you could find them, are —
All these things get recorded in proposals?
Not everything, but in general, if a person has ideas of what they want to work on, why, you tend to write a proposal about it, and —
So those —
I think of proposals as asking the government for money, but these proposals were just within the department?
Well, I was thinking actually of the proposals for financial support. You ask for money, but you have to tell a good deal of what you're proposing to do with it. Now, many proposals are based on straightforward continuation of existing work. Some of my friends have said that the way to write a proposal is to describe the work you did during the last year, because then you know you can do it. But most are somewhat ahead of that and many contain ideas for new directions. Some of these, of course, never pan out, but they do indicate the thinking of what the person hopes to move into next.
And those proposals would be in the Electronics Research Laboratory files?
Yes, but I don't know how long they keep them, space being what it is.
One of the really hard things to get at, and I think it's a very important part of the history, is the relation between what was going on in research laboratories and military interests. When you're working on the thermal lens, I think of that as possibly being close to some of the military interests in things like thermal blooming. I'm just wondering if you could give us any insight, to the extent that it's not classified, into the relations between what was going on here, and the — now, at some point you change your sponsorship from DOD to NSF, mostly, which is also interesting. Anyway, you see the drift.
Yes. First of all, let's talk about the sponsorship. From the time I came back and started work in the quantum electronics area seriously, I have had both DOD and NSF sponsorship. I guess the emphasis has shifted a bit. At the beginning, it was probably more DOD, and later it was more NSF, although there wasn't any particular conscious plan to do it that way. That's just the way it turned out, partly, of course, coming with Al (Elias) Schutzman's program in optical communications.
Was the optical communications more interesting to NSF?
Well, that's quite a story in itself, and is in fact being written up by a group at Syracuse, that I can give you the data on. It was really Al Schutzman's vision. He had the communications program. He was getting a number of proposals in the optics communications area, both on theory and devices, and so he got a group together to see if there couldn't be more of a relationship among the university workers in the field, and the people on the industrial side. He started annual meetings. Well, first he started with meetings twice a year, but finally they came down to annual meetings. He called the grantee — users' meetings, and although he's no longer in charge of the program, they're still going on and they're very successful.
Where is Schutzman?
He is still at NSF, but he's now working on the Engineering Centers programs.
I see, so it was as an NSF person that he started this.
That's right. So in a sense it was an experiment, to see how it would work to pick one area and have some interrelationships, not only between the workers in that program but the expert people on the outside, and that one worked beautifully. The total funding is relatively small. If you compare it to the budget just of Bell Laboratories in that field or most of the military organizations it would be extremely small, but the influence it's had has been considerable.
That's quite interesting. You always think of the Department of Defense as doing that, I mean, as ONR will organize meetings among people who are doing certain things, but I never heard of NSF doing this.
Well, they certainly don't do it universally, but they have done it in other areas. In fact, it had been done in the materials area, and I believe Schutzmann modeled his program after that.
Did that have a particular impact on your work?
Yes, in recent years. That program was started around 1970, but I can look up the exact date since I have all the reports on those meetings.  So the first period that we've been talking about was before that. But the DOD's support was for quite basic work, as the NSF work was, so there wasn't a major difference in approach to what we did for one or the other, nor was there a conscious shifting from one to the other. It's more just the way ideas evolved. But to come back to the thermal lens effect, yes, we did see the relationship of this to the problem of thermal blooming in the atmosphere, but most of our work was done with liquids. Some parts of that could be extrapolated, where the theory could be used, but it was really the work by the group at United Aircraft, I think, that provided the beautiful, very definitive experiments on the effect of wind and convection in the atmosphere.
No, I didn't know about that.
You can look up some references on that. That probably was the central basis for the military designs.
How would the military have permeated or not permeated into work here? I mean, if one was trying to understand the interaction between the work here and military interests, how would you picture that, characterize that?
With the agencies we've dealt with, it isn't terribly different from NSF. I'm talking about ONR, AFOSR and ARO. They do have priorities, you know, certain things that they're pushing at any given time, but I think in the beginning that wasn't so much true. Now we've found more and more, there are certain areas that they're interested in more than others, but they do expect pretty basic work, and —
So it might push some people into those particular areas, but not in any particular direction within those areas? Or you don't even think it does that?
Well, it depends. I think there's more of a tendency in recent years to become more programmatic, and say that "we aren't interested in a certain material but only in another material", because they have certain things in mind. And it depends on the individuals that you're dealing with, of course. But by and large, we've had very broad support, and very good understanding of the basic work that we can do, in contrast to development work where a product meeting certain specifications is needed at a certain time.
So in the thermal lens business, did they do anything about that particular thing, like putting you on review boards for?
I was on one panel concerned with the limitations on high — power CO2 beams in the atmosphere and presented something of the work we had done. We had some work on self convection effects, but not on forced convection, and all in liquids as I have mentioned. I think the work of United Aircraft that I mentioned came later, and may have been inspired by that meeting.
I see, so really in a sense one of the uses they would make of you is to review other people's programs, more development-oriented programs?
Yes. I was one of several people of course to have inputs on that problem.
Avco, were they part of your, or they didn't come in?
Avco certainly was concerned with very high powered lasers, and in that sense, was concerned with the thermal blooming, but I don't remember any work they did, other than using the work of others. Maybe they did.
Were you also consulting at this point? Would your knowledge flow into more practical work by consultantships?
Let me see, was I?
Say late sixties, early seventies.
At that time, I was on some committees. For instance, the Science and Technology Advisory Committee to NASA on the Apollo Program. So my industrial consulting during that period was sort of minimal. I was still doing some at Hughes, some at TRW, but it often didn't have much to do with the things I was working on. It would be more on broader range issues, such as the direction they should go in the future, things of that nature. In some cases, it did impact on laser problems, but it wasn't really the kind of technical consulting where I was using the detailed information from a research program.
Kind of broad, general —
Yes. It was more policy, in some cases.
So then if I summed it up by saying, a little bit of consulting and rather more serving on government committees?
At that stage, yes, that is correct. And, well, not only government committees, but there was the Commission on Engineering Education, I was chairman of, and so on. A few things like that.
NASA, and then this commission, and the reviews —
And I guess I was on the IEEE board of directors, so — there were plenty of things to do. Now I think back on it, I don't know why I got into so many things.
Well, it was an active time, and you had a very active research program at this point in all sorts of areas. Well, the rest of my questions really have to do with just specific researches. I just have four here really, and they might not exhaust the most important ones. Mode locking, FM modulated lasers, integrated optics and liquid crystals, and then there was a year you spent at Stanford, or some length of time you spent when you got involved with some problems there. So these are really just — you might want to work from your bibliography, if you prefer. I only went up to the middle seventies, by the way.
Well, if I can start back a little bit, because you were asking about the initial program, and we talked about the thermal lens work, and the deflection modulation work of Erich Ippen. These sort of grew out of my period at Bell Labs but other examples of that period — Marvin Klein's work on ion lasers, and Bill Clark's on microwave discharges in helium neon lasers, grew more from the papers that they did in the course. We obviously discussed the problems, whether they were important ones and whether we had something to contribute. But in considerable measure, they were student — generated projects that fit into our overall plan and funding authority. The mode-locking and the FM of lasers really were started to be part of the same project, related to the question of modulating the lasers — how to get information on the lasers. In the FM, there had been quite a bit done on that by Yariv and by the Stanford group, but at that stage, we thought we had a somewhat different approach. The idea started with something we called mode code modulation, to use the different modes a laser might operate in, to carry information.
Well, you could call it frequency shift keying, I don't know that that name was even current at the time, but you would use the different longitudinal modes of the laser, and selectively excite them to produce a given code. For a binary digit system you would only need two modes. The companion to that was use of the transverse modes for simple pattern formation, if one could find how to excite them. Dave Auston took the latter idea, and realized that a more useful, perhaps more basic, idea was that of transverse mode-locking, which had not been done at that stage. He did some beautiful work, both in analysis and experiment, in showing the effective scanning of beams you could get by locking the modes together with certain phases.
He was also a student of yours.
Yes. I don't know that transverse mode-locking has ever turned out to be a terribly practical thing in use, but it was a beautiful piece of work. We followed it up a bit when Auston went to Bell Labs. There was another thesis on the subject concerned with circular cyndrical forms. So the mode-locking really started with that. Then, when Andrew Dienes came — we tried to keep a series of visitors from industry or other countries in our group, and several had been from Bell Laboratories: Andrew Dienes was one of these. Cutler was the first one to come, and Peter Smith, P. K. Tien and Ivon Kominow were others.
It sounds like a strong connection with Bell Laboratories. Is it a special connection with Bell?
I guess I'd have to say yes. It wasn't any attempt to rule out others, but it turned out, through a combination of things, that we had many connections with Bill. Not only I but others among my colleagues had spent some time at Bell, and we had so many people from Bell here, and they hired so many of our students, that it turned out that we probably had more of a relationship with Bill than most others. But we had quite a bit going on with other places, Hughes, and right now, locally, Hewlett-Packard. So there wasn't an attempt to single out one. Partly it's that they had so many very good people and so many interesting things going on.
Did that shape what went in here, would you say, the strong connection with Bell?
Oh yes. Certainly. And again, it wasn't so much a conscious decision, to do something because Bell recommended it, but it was the interaction with very good people. They made suggestions, and in some cases really supervised the work of students. So, coming back to Dienes, when he was here, he got deeply involved with several of our students, and in particular suggested the mode-locking for short pulse work. I have to go back and try to think what the students were doing, before he came. It was on some related work, particularly (Omar) Teschke. But in any event, he influenced the direction of the program and the individual students a great deal during the year he was here. So he deserves credit for that aspect of mode-locking for short pulse work, which we followed up for quite a few years, but it really started with the suggestions to Dave Auston, which Dave took in a somewhat different direction than I had thought of, and really a more important direction.
Lots of interesting things in this conversation, in what you're saying, that throw a lot of light on the way in which research gets done.
Yes. Not everybody works the same way, but certainly in my case, in so many of these things, the students either initiated the ideas or really formed the approach, after the general direction that was pointed at the beginning. Or made the key decisions. That was true in our integrated optics work. That's an interesting field, because after all these years, it still isn't very much integrated.
No, I didn't know that.
The work that's been done has certainly been important, Tien, at Bell Labs, and several others there really started the field. But the important thing first was to study the kind of guides you could have on the surface of a crystal, usually — well, it didn't have to be a crystal. But most important have been electro-optic crystals such as gallium arsenide or lithium niobate, and the interactions you could have with them. Of course the idea at the beginning was to put a lot of components together, as we do with integrated circuits.
Was that something like the early maser, in that it generated a lot of excitement at the time?
Yes. I'd say it's not as fundamental an idea, but it certainly generated a lot of excitement. There were special meetings devoted to it, which they still have, and a special issue of the JOURNAL OF QUANTUM ELECTRONICS, about a month ago, was devoted to integrated optics, so it's by no means a dead field. But at the same time, the important work that has been done so far has been the guiding and the individual components that have resulted, and not so much the integration. A few things have been integrated. And there will be more. But most people think it won't ever be like integrated electronics, with thousands of things on a single chip. Maybe that's not right, but that's the way it looks now.
Because, I don't know why, but I have a feeling that it just might have looked very exciting in the early seventies.
Oh yes. It did.
I guess what I'm trying to say is that I feel a sort of abrupt movement into it, as a —
Yes, there certainly was. Tien, you know, was a national lecturer for the Optical Society, and went around and described his work, which is really very fundamental work on optical wave guiding, and nearly every laboratory that was in a position to do so did something in that direction.
Certainly Amnon Yariv did at Caltech, and the group at MIT. And there has been important progress but it hasn't resulted in large scale integration. The work that we did with liquid crystals for integrated-optics applications came about because of Ron Shen's work in physics on the use of liquid crystals in nonlinear optics. I believe it was his suggestion that somebody ought to look at these from the point of view of switching and modulation. In any event, the student who really got us into that, Chenming Hu, was taking a course from Shen and certainly used many of his ideas. Liquid crystals have large electro-optic effects, in comparison with typical solid crystals. The problem is that the effects are very slow. There was some hope that some of the things we could do would improve response time, and there has been some improvement in the years people have worked on the problem, but it's still relatively slow. For some switching purposes, that's not too serious, but for high speed modulation, it wouldn't be suitable in any of the forms that now exist. But it was a very interesting area to look at, and because of the work in physics, it looked like we might do something complementary.
That's interesting, especially, and also it's interesting that the student is kind of the —
— he was the connecting link. That's right.
And again, it's sort of different from what I had thought. Well, this acousto-optic filter tuner, is that something we should talk about?
That's the work at Stanford. Well, let me tell you about it. Steve Harris at Stanford had developed the acousto-optic filter, one in which the matching of the acoustic wave to the optic wave determined the wavelength at which it would pass the optical wave, and by changing the voltage, you could change the phase matching, and therefore the frequency that would get through. It was a very nice device and I think he was involved in a company that was making these. In any event, one of the things that he proposed was the use of this in tuning dye lasers, but the simple attempts to put it in the dye laser to see if it would tune didn't work. At that time I had proposed going to Stanford for a six months sabbatical. Steve suggested that I look at this to see if there was a fundamental problem, or if it was just a matter of finding the right parameters. At the time I was there, Bill Streifer, who was then on leave from the University of Rochester, was also there. I think he went back to Rochester before he went to Xerox, but later was at Xerox.
So it's pretty much your way of doing things, to go to a group and sort of ask them what problem might be a good one to work on, is that?
Well, it didn't really start that way. I asked if I could spend a sabbatical leave at Stanford, and Harris said, yes, if he could think of a good problem. I did have work of my own to do at that time, but of course the point of a visit to another active place is to see fresh ideas.
I'd say in other places it was more that I knew the work going on, and I would go there and get involved in it, not particularly saying "Do you have any good problems for me to work on?" It was certainly true of Hughes, that the problems evolved out of the interactions with the group. Certainly, I found that going to these other groups has been an extremely important part of my career. For five or ten years, after each visit, I could very definitely trace the influences. But in general, it happens by just going and working with them, knowing that the group is good, but not particularly worrying about their problems of the moment. So in Stanford's case also, I wanted to spend a sabbatical there because it's a tremendously good group. Tony Siegman, whom I know best, was on leave as was Pontell, but Harris and Byer, were there and they really are excellent people. But anyway it was Harris who said, "Well, I'll see if I can think of a problem," and then he came back with the suggestion for analysis of the tunable filter. It turned out that Streifer and I worked on it together. Streifer is just a marvelous analyst, has good physical pictures and a command of both formal analytical tools and use of a computer and I learned a lot by working with him. We made the analysis, published the paper, but I guess it's not anything that ended up as a terribly useful device for tuning dye lasers.
Was this also an experience that influenced your next four or five years?
Probably not as much as the others. We actually tried a surface version of that — this is in the period of integrated optics, and had some results. Mark Chang was a student on that. But the tolerances were so severe that it wasn't a terribly useful device. So I'd say it influenced things a little, but not as much as the period at Hughes or at Bell had done.
What else should we talk about?
You asked about the low loss fibers changing the direction of our work. This ties in with the NSF program of Schutzman's, and in one sense, I'd have to say, not a lot but in another sense yes, since all of our devices were directed toward possible communications applications. The main emphasis we placed on the work for quite a while was on short pulse phenomena. The fibers have very low dispersion and take quite short pulses. As you go to shorter pulses, of course you have higher data rates, so we hoped the short pulse work would have application to fiber—optics communications. But our short pulse work had really started in an earlier NSF program. Schutzman used his grantees to start the program in optical communications which we have discussed. As this program grew, it did have a lot of influence.
I guess I don't completely understand that. The shorter pulse work is independent of working in thermal lenses, and could just as easily be applied to fibers in —?
— oh yes, it has nothing to do with the thermal lenses. No, the thermal lens work, we haven't worked at for 15 or 20 years.
Because in a way, this was a trivial question. It was just, whether the fibers coming in is what decided you to de-emphasize thermal lenses —
Or whether you wound up before that.
I think it's just that we could see other things to do that were interesting to do seemed needed. Now, the thermal lens effect is still used. Quite apart from its serious effects in de-focusing, it's still used by chemists as a technique for very sensitive absorption, spectroscopy. I'm always amazed—I still get some papers for review — that they're now down in the 10-8 and 10-9 loss per centimeter, measured by these thermal effects. So, obviously, the field isn't dead, but we really left it some time ago.
It just became an analytic technique.
Yes, or to know that it exists in lossy materials, and how to compute the thermal blooming kind of usually undesirable phenomenon. Now, Bill Bridges at Caltech feels there are still some fundamental problems. He used some of this work when he was at Hughes, and there were some things that worked fine and some that didn't check. So, when I was on leave at Caltech last spring, he tried to convince me to get back into it. But we didn't do enough to know whether it's just a problem of definition or measurement, or something really interesting and fundamental. But coming back to fibers, I think the main thing that that has affected us is the type of lasers we work with. It's quite clear that if you are interested in communications and information processing, the semiconductor laser is the device you want to work with. Dye lasers are fine for experiments and to study materials and to study devices, but you certainly aren't going to put those in communications systems, at least not in those of commercial type. So I'd say that our stress has changed from dye lasers to semiconductor lasers, partly because of the communications emphasis with fibers, and the program that I mentioned with NSF.
So if I'd gone forward, I would have gotten a lot of semiconductor work, is that right?
Yes. Right now, of the five students I have, all working with semiconductor lasers in one form or another. A couple of them are still using the short pulses from the dye lasers to study them, but the emphasis is really on the processes in the semiconductors. The other thing I think we might talk about a bit is the relationship to my colleagues here. I mentioned Shyh Wang. Steve Schwarz and Ken Gustafson (T. Kenneth Gustafson), Have been the other members of our group. We don't actually have a formal group called the Quantum Electronics Group, but everybody refers to it as that. We do work very well together and very cooperatively, more so I think than in some places. In particular, we're all pretty much in the same laboratory, and share a lot of the equipment. Some of it has to be dedicated to a particular experiment, but we share a great deal of the equipment, and laboratory space. There's an area of the laboratory that's understood to be one of our primary spaces, and shifts from time to time among us, as we have different needs.
How large a group is that in terms of students and in terms of funding?
It's interesting that the student population has remained remarkably constant for quite a while, and part of it I guess was the limitation with four people. Let's say that we can handle maybe six students each, so around two dozen students have been typical of what we've had for quite a few years. Funding, I'm afraid I don't have the numbers for you.
Well, maybe order of magnitude.
I see. I suppose that we assume that each person has to have, for that many students, at least 50 K per student so we're talking about 250 K per faculty member per year. Some students would be on fellowships or teaching assistantships. So we'd be talking on the order of a million dollars per year, but I might be off quite a bit, without checking actual figures.
Well, at least that gives—and how far back does this group go?
Gustafson was the last of the four to be appointed, and I think he came about '65 or '66.  So the group has really been together more than twenty years. Schwarz has recently concentrated more on millimeter waves, so isn't now so active in optics. We have just appointed a new young person, John Stephen Smith, a student of Amnon Yariv's at Caltech. I mentioned the cooperation in space and equipment, but of course much more important than that has been the intellectual interchanges, and joint projects. Schwarz and Gustafson had some in tunneling detectors. Shyh Wang and I just finished with one student, jointly, on semiconductor laser arrays, and another is now working with us. Ken Gustafson and I worked together with Moty (Mordechai) Heiblum on tunneling devices. So we have more joint supervision of theses than I think you'd see in most places, I don't know that this is so much a planned strategy as that it's just worked this way. A student would find it useful to get information from both of us, and we would decide then that it was more accurate to list us as joint supervisors. I do this with Andrew Dienes a great deal, ever since the period that he got us started in the short pulse work, although he is now a faculty member at UC Davis. He continues to contribute a great deal to our program.
That is, I think, pretty unusual. I mean, you don't see that very often. You see one supervisor's name on a paper.
And a student. So that's a 20 year old situation, this group.
Well, the joint supervision aspect sort of grew. I think we always worked together. We have a joint seminar, and tried to exchange ideas. Also, if one person is on leave, then another person helps in supervising the students. Of course, some supervision is done by long distance, but there is a part that has to be done locally. But I think the part that's really unusual is the formal joint supervision of theses. We have had, well, I guess a total of a dozen by this stage.
And so should one understand, I guess, as far as intellectual interactions, you've already said that, that one should understand, especially recently, your work in terms of these intellectual interactions.
I'm sorry, I didn't quite get the point of that.
I think you've really answered my question. I just rephrased what you said. Well, thank you very much.
Professor Whinnery's biography and bibliography are on file in the Laser History Archive.