Sidney Millman - Session I

Hoddeson:

I am interviewing Dr. Sidney Millman, formerly of the Bell Laboratories, now at the American Institute of Physics, and he very kindly agreed to have a conversation with me about some of his experiences at Bell Laboratories. My interest in Bell Laboratories has to do with some research I’m beginning on the history of solid state physics, and I’m very interested in looking at the various threads, research threads, that went into the development of the solid state physics research group at Bell Laboratories. People came in at different times, for different reasons, from different places and different areas of research. I’m interested in: what brought you to Bell Laboratories, where you came from, what you found, whom you worked with, and how you eventually ended up being the director of a research division in solid state physics.

Millman:

That’s a very good question. Actually, I did not come to the Bell Laboratories primarily to be involved in solid state physics. My route was quite different. During World War II, I was involved in research on magnetrons, the microwave transmitters, for radar at Pupin Physics Laboratory, Columbia University. The Columbia Radiation Laboratory was set up early in 1942 by Professor I. I. Rabi as a branch of the main body of the radar effort which was carried on at MIT. We also became acquainted with the activities on magnetrons at Bell Laboratories in a group headed by Dr. J. B. Fisk. The Bell work was at longer wavelengths than we at Columbia were assigned to work on. And in fact, one of the tubes or magnetrons that I had most to do with inventing, the Rising Sun Magnetron, was adopted later by Bell Laboratories for further development and design for ultimate manufacture by the Western Electric Company, and that was the tube that was going to be used in radar for fighting the Japanese War, if the U.S. were to invade Japan. Fortunately we didn’t have to use it. The tube was successfully used on an experimental basis in bombers, and they could see the increased resolution coming from shorter wavelengths, as expected from 1 centimeter compared with 3 centimeters, the shortest wavelength used in radar before 1945.

Hoddeson:

What has happened to the Rising Sun Magnetron?

Millman:

Oh, it’s still good but I don’t know whether it is actually in use at present. If you go to the Columbia Radiation Lab, you’ll find a cross section of the anode of the Rising Sun Magnetron as an emblem on its reports. Anyway, we got to have a very close association with the Bell Laboratories group, and as the war was approaching an end, Fisk asked me whether I would like to join the Bell Laboratories. He was the one who recruited me.

Hoddeson:

I see.

Millman:

When he first recruited me, I was going to join one of the more basic physics research groups, but as it turned out, because of the way the recruiting worked out at Bell Labs, in 1945 I agreed to join the Bell Laboratories in the research group on electron tubes, traveling wave tubes and so on, and not the solid state physics group.

Hoddeson:

When was that?

Millman:

That was the end of 1945. I came in December 1st, ‘45. And I was a part, to be sure, of the Physics Research Department, as it was called at that time; now that organizational unit is called the Physical Research Laboratory. I was a part of physical research, but essentially only in name, because Mr. Fisk, in addition to being an Assistant Director of Physical Research, had direct responsibility for one of the groups, which was called the Electron Dynamics Subdepartment, 1160, of the Physical Research Department. This Subdepartment included people like A. M. Clogston, L. R. Walker and J. R. Pierce. We were not doing physics research as you would infer from the term. We were more involved in, I would say — it was sort of an applied research group, most of us were involved in traveling wave tubes. And I think at that time we were still working on magnetrons too. Anyway, magnetrons, traveling wave tubes and so on.

Hoddeson:

Were these things you were working on then used for some of the solid state physics research?

Millman:

No. The electron tube research was entirely unrelated to solid state physics.

Hoddeson:

Unrelated completely?

Millman:

The only connection with — I think it’s purely organizational, in the sense, Mr. Fisk, who had previously during the war headed a group such as this, magnetrons, traveling wave tubes and so on, was advanced to the position of Assistant Director of Physical Research. In other words, he was essentially designated to become the Director of the Physics Department, after Harvey Fletcher retired. And since he had himself direct knowledge and expertise in activities of this group, that was brought, for organizational reasons, directly into physics, you see, rather than because of any direct connection between them. So we were only nominally in this physics department or the physics laboratory organization.

Hoddeson:

Were you working in the same building with everybody else?

Millman:

No, as a matter of fact, even that wasn’t true, because the physics activities, the basic physics activities, the solid state research and physical electronics was at Murray Hill whereas we in Department 1160, the electron dynamics research people, were still in West Street. Now, we knew we were going to go to Murray Hill ultimately, and I stayed in West Street for over three years, didn’t move to Murray Hill till early ‘49. Prior to that time, we were organizationally moved out of the Physical Research Department, after Mr. Fisk left, which was in early ‘47, we were put in a different organization, under J. W. McRae, Director of Electronics and Television Research. In other words, we were more nearly properly organized, because we were a little closer to communications research. Subsequently, we had acquired a different director, when McRae was promoted and transferred to the development area of Bell Laboratories. A man by the name of W. H. Dougherty took over in charge of the Electronics and Television Research Department. And so I remained in this business of traveling wave tubes, tubes in general, applied research, until January 1, 1952. This research had no connection whatever with solid state physics. In fact, after about July 1947 we were not even organizationally in physical research. But it didn’t make much difference to me anyhow.

Hoddeson:

Were you still interacting strongly with the group at Columbia at that time?

Millman:

No.

Hoddeson:

You were independently working at West Street on these traveling waves.

Millman:

Well, Columbia was never involved in traveling waves. Columbia was involved in magnetrons. When I came to Bell Laboratories, I did some work on magnetrons, and then gradually moved over completely to traveling wave tubes research which was the main focus of activity of our group after the war. The traveling wave tube was invented abroad by R. Kompfner and taken over at Bell Laboratories, with a great push provided by John Pierce because of his recognition of its great importance to broad band communications, That was an important part of the research at Bell Laboratories, and the traveling wave tube was an important research activity of that group, that I, along with Clogston and others, was involved in. When I got back into what you might call pure physics, it was at the level of an administrator, having been promoted to the position of Director of Physical Research on January 1st, 1952. At that time I dropped all connections with tube research.

Hoddeson:

Were you at Murray Hill at that time?

Millman:

Yes. The tube research activity, along with the tube development work, was moved from West Street to Murray Hill early in 1949 when building 2 was completed. Building 1, the oldest one of that complex that includes the present library,^[1] was there since 1941. Then at about the end of ‘48, or early ‘49, when building 2 was being completed, we moved in about January or February 1949, but at that time we were located right close to the development part of the tube activity. Research and development were placed very close together, because of the common interest in electron tubes.

Hoddeson:

That must have stimulated research. Being together.

Millman:

Well, in general one tries whenever one has an opportunity to arrange groupings — you try to locate people in research who have a common interest with the appropriate development departments. That’s part of the strength of the organization. Organizationally we try to separate research from development. Research people are not required, although they could be encouraged to take an interest, in what goes on in development, if they’re interested, if they see some relevant research problem to work on, but they’re not required to pull their chestnuts out of the fire in a development problem. Nevertheless, by having research and development close together for people having common interests, it helps to stimulate coupling between the two groups. So we were located initially right close to the tube people, the development people, and we were working in research, until January 1st, 1952, at which time even though the research work of the group wasn’t all of a sudden abandoned, Mr. Sidney Millman was transferred out of this group and promoted to become the Director of the Physical Research Department. This coincided with a more general reorganization of the research area.

Hoddeson:

For some special reason?

Millman:

Well, there comes the time, perhaps because of the rapid growth of Bell Laboratories, that the higher management felt that it was best to make a thorough reorganization. Just to show you, prior to January 1st, 1932, both Pierce and I were both department heads, or rather subdepartment heads, as we were called at that time. In fact, we both were co-subdepartmental heads of one subdepartment. After that, each of us became a director. I was moved completely out of this applied research area, and put in physics research, to become Director of Physical Research, and Pierce became a director of the tube research activity — that included traveling wavetube research which had been expanding too, and he became a director of that laboratory, and the man who was our director at the time, W. H. Dougherty, was promoted to a higher level called executive director or general director. So there was a rather big reorganization at that time, and I, with my previous physics background in basic research at Columbia, was probably considered a good choice to head the Physics Research Laboratory, though even I did not come with any direct experience of any of the physics groups, such as solid state physics or physical electronics.

Hoddeson:

I see. So then you had these groups directly under you — Grisdale, McKay, Morgan, Shockley, Germer. Was K. K. Darrow the only person in general physics?

Millman:

Well, general physics was just a euphemism for giving the subdepartment that he was in some kind of a name. Darrow actually was only working for Bell Laboratories nominally. He was essentially running the American Physical Society as its secretary. It was Bell Laboratories’ contribution to the American Physical Society, by donating Darrow’s time, but he had to report some place, and it was most natural that this be in physical research; and thus nominally he reported to me.

Hoddeson:

I see. In what way?

Millman:

Somebody — any given scientist or employee has to have a boss, to worry about his salary or anything else, somebody’s got to be interested and know something about what the scientist is doing.

Hoddeson:

Where was Darrow working at the time?

Millman:

Actually, Darrow never moved out to Murray Kill. He continued to have an office in West Street. The West Street location of Bell Laboratories was still in existence for many years. So Darrow remained — I mean, organizationally — as he always has been in physics, physical research. He started at Bell Laboratories as a research physicist, but very soon changed his role to become a writer of review papers in the BELL SYSTEM TECHNICAL JOURNAL. Later on, when he became secretary of the American Physical Society, he phased out of writing too. The American Physical Society secretaryship got to be a full time job for him.

Hoddeson:

As director, you must have been very interested in finding out in much more detail what each of the groups was doing.

Hoddeson:

I notice here in the 1952 organization chart that the transistor physics group is separate from the physics of solids before, in the earlier charts, they were combined.

Millman:

Yes. Right. Well, as a matter of fact, I was told that the reason for the breakup of the two groups had to do with some friction that developed in the original large solid state physics subdepartment. The split into two subdepartments happened two or three years before I became Director of Physical Research. It had no connection with my assuming directorship of that particular department.

Millman:

Well… not for very long after that. If you look at one for ‘50 or so, you will find them separate. Well, first of all, it was a very large group, so the idea of breaking it up into two is not unnatural anyhow; but what’s more, I had heard that Bardeen was a little unhappy, in the sense that Shockley had tried to talk him out of working on superconductivity. Bardeen began being interested in superconductivity. Now, Shockley, who was of course the main driving force for the transistor, and was the head of the group, realized the potential fruits of the invention of the transistor, could see a lot of problems, experimental and theoretical research, which would advance the transistor further, and it looked to him a little odd to have, here is one of the strongest theoretical physicists like Bardeen, instead of finding awfully good related physics problems in semiconductor physics, and he could of course pick any of the basic research work he wanted, why not stay with the transistor, when there’s so much pay dirt there? Well, Bardeen developed an interest in superconductivity and saw even greater challenges there at that time. I heard this more or less indirectly from friends, because I was not directly involved.

For these reasons and because the group was large enough anyhow, they divided up the activities, and by having essentially one group devoted almost entirely to semiconductors with Shockley as head, it included people like Haynes and Hornbeck and Goucher and so on. The other group, which was still fairly large, reported, to S. 0. Morgan and was devoted more broadly to solid state physics. It included magnetism, with Bozorth, Williams and Nesbitt, and the theorist P. W. Anderson. Holden in ferroelectrics, as well as — he was the one who cams through with the unusually narrow electron spin resonances for organic materials, which was used as a probe for calibrating magnets. It had a complicated organic chemical name, abbreviated as POV, but it gives a very, very narrow line width for calibrating magnetic fields. Anyway, Holden was a physical chemist, and he worked on ferroelectrics and on crystals, too. You see Schawlow in that group. Schawlow was, at that time, interested in superconductivity. Later of course he got involved in lasers.

Hoddeson:

And here we also have under physical electronics, we have people like Herring and Wannier and Peter Wolf, I see.

Millman:

Well, at that time, prior to 1956, there was no theoretical physics group as a separate group. It was considered good to have the theorists spread among the experimentalists, and it wasn’t till 1956 that the values of having separate theoretical physics groups was recognized, and that was established in ‘56. But in ‘52, there was no theoretical physics group.

Hoddeson:

It was entirely experimental.

Millman:

No, I wouldn’t say that. We had theorists such as Bardeen, Conyers Herring, Phil Anderson, Peter Wolf and others. They worked on theory, but they were spread or scattered into various groups, where the heads of the groups were almost always experimentalists. Except I don’t know how you would label Shockley; experimentalist or theorist or both.

Hoddeson:

Now the theorists have a Journal Club. Had that already started by that time?

Millman:

Oh yes. The Journal Club existed before I became Director of Physical Research.

Hoddeson:

Was Herring the steam behind it?

Millman:

I wouldn’t be a bit surprised, yes, I think so. I think certainly ever since I knew him in ‘52 he was running the Journal Club. That’s right.

Hoddeson:

What about colloquia, seminars? Now they have quite a number of them fairly frequently.

Millman:

Well, I think seminars existed at that time too. The general colloquia I remember in the early fifties. There were three kinds. In addition to seminars specific to some particular group, there were three main physics seminars. One was a solid state seminar on Friday morning, every third Friday. Another one was the Journal Club, run by Herring generally, and thirdly, there was this more general colloquium. I have to go soon…

Hoddeson:

Well, maybe the best thing to do in the few minutes we have left — to make the most of the minutes is for you to give me some suggestions as to where I can get certain information.

Millman:

What information?

Hoddeson:

Well, one part of the information I’m interested in relates to Bell’s policy towards having visitors from the outside: consultants, people who would have short term appointments, summer visitors, foreign visitors at the Labs, to exchange information with. That’s one kind of information. Another kind which might be the subject of a whole separate discussion, that maybe we should put off, I was speaking recently with Betty Wood; she mentioned that you recruited her one year to teach in the CDT program, I’d like to know as much as possible about that program.

Millman:

I’d be glad to tell you, I’ll tell you briefly now. We can talk more later. Essentially, just about the time I became a Director of the physical research laboratory, the Bell Labs decided to overhaul its training program, what they call Communication Development, the Training Program they had for their engineers, and inject more physics into that program.

Hoddeson:

Why?

Millman:

Well, by that time, it was apparent that transistors, and solid state physics in general, were going to play an awfully important part in the technology of the Bell Laboratories. The original program was set up by Kelly back in 1948, the engineers, coming from schools with a bachelor’s degree don’t have quite enough training — we ought to give them some specifically Bell systems kind of training. And maybe a little more general sophistication, so they had about a year of graduate work, in the Laboratories. But in 1952, the feeling was that they didn’t know enough physics, even the people coming from the very good engineering schools like MIT or other places, we’d find out that they had no solid state physics in their courses. What did they have in atomic physics? Well, maybe a week or two of atomic physics in a general physics course. At any rate, the thought was that, in this extra year of training they were going to get, they ought to have some modern physics.

Hoddeson:

Who made this decision?

Millman:

Well, Mr. Fisk, who was back in the Bell Laboratories and now was on a level called executive director or general director. Fisk got me to set up a committee to decide what physics courses it would be good to give these entering engineers. I got people like Herring, Galt and others on the committee, and we decided on a curriculum. We decided we ought to have three courses, one on atomic physics, one on solid state physics, and one on the physics of waves. And for several years I was the one who was involved in getting lecturers and instructors to teach these physics courses to them. So I was the “angel” for the physics courses.

Hoddeson:

Do you think some memos describing this effort might still be in existence?

Millman:

I believe there were some in 1952. Well, I’ll be glad to tell you more about this. I remember asking Betty whether she would be interested in teaching atomic physics. She came to the Physical Research Department not with the usual background of having gotten a Ph.D. in physics, but rather in crystallography. She picked up bits and pieces of solid state physics in connection with crystallographic structure studies and I knew she was a very good expositor, very good at teaching. I thought it would also be good for her to learn more atomic physics in a systematic way. Who learns more than the instructor who teaches? So I asked her if she’d like to teach. I didn’t want her to give the lectures in atomic physics because she might feel inadequate, she could be one of the instructors, taking sections. Her first response was highly enthusiastic, but she asked whether we could wait until the following term to give her more time to prepare. She was very conscientious. I said, “Fine.” This was all on a voluntary basis. So I said we could keep her in mind for the following term. She took it on and thanked me several times afterwards. She was very appreciative of the opportunity to teach and to learn more physics.