Robert Maurer - Session I (original) (raw)

Hoddeson:

Before we start the formal interview, I’d like to record some of your informal comments about early solid state physics.

Maurer:

A very good example is, and I have forgotten the name, a man at Cornell who, in the early part of the century, studied the properties of germanium and measured conductivity and Hall effect with erratic results, which weren’t understood. It turned out during World War II, when Lark Horovitz at Purdue began to investigate germanium as a possible semiconductor diode, that that earlier work at Cornell was exceedingly interesting. What made it interesting was the theoretical background that had accumulated since that work had originally been done. Without that theoretical background, it was just a set of data that was relatively incomprehensible.

Hoddeson:

That’s exactly the point. So much good work on solids was done from year 0 up to about 1926, but it was in more or less random pockets, and separated. It was only within the framework provided by quantum mechanics that some of the work began to join under the “umbrella”, as I like to call it of, solid-state physics. The second section of the book our international project will be writing — a section on the quantum theory of solids, describes this framework.

Maurer:

I would like great emphasis upon the work of A. H. Wilson, who demonstrated what we now call the band theory. While it was true that, until after World War II, its limitations were not really appreciated, nevertheless, it provided the conceptual framework within which all of the work, particularly on semiconductors, was done in the late thirties and forties.

Hoddeson:

Quite true. In fact, as I understand it now — and I haven’t done all my homework on the Peierls papers yet — the basic band idea is already there, although in obscure form, so that it couldn’t have had the impact without Wilson’s work.

Maurer:

I never investigated the exact distribution of credits here. I only know that in my own case, it was Wilson’s paper that was important. I’m sure that none of these things were done, you know, completely individually.

Hoddeson:

No.

Maurer:

There was too close contact...

Hoddeson:

As I heard it, Wilson was trying to understand the Peierls papers in order to present them at a colloquium. And then he wrote his beautiful papers in 1931 and 1932, which weren’t in fact the last word. For example the third paper got the rectification of metal to semiconductor junctions in the wrong direction, because the barrier model wasn’t right.

Maurer:

That’s a different matter. Surface phenomena were still in a sort of super-primitive state. For example, one didn’t understand at all the role of surface states. There was, in the late thirties, I’m a little vague as to time, but there was an understanding that when you terminated a crystal with a surface, you got possible solutions which were different from the volume-band solutions of the wave function. But the property of these surface states weren’t really understood, nor was the role that they could play. I remember very vividly during World War II, we at the University of Pennsylvania were investigating silicon and germanium diodes, and a graduate student and I decided to measure the work function of germanium.

Hoddeson:

Right.

Maurer:

And the reason for doing it was that according to our picture the work function of p-and n-type germanium should differ by the band gap, about three quarters of an electron volt. Now, we measured the work function and found that p- and n-type germanium had, within the error of our measurement, the order of perhaps a tenth of a volt, the same work function, and we didn’t understand it. We didn’t understand it until Bardeen published his paper on surf ace states, where he showed that by essentially keeping the equivalent of a film of metal conductor on the surface, surf ace states could make p- and n-type germanium have exactly the same work function. So, those problems just weren’t appreciated during the thirties, and I don’t think it’s quite fair to hold Wilson’s very primitive theory of rectification against him in the sense that it somehow detracts from his understanding of the volume properties of the wave function.

Hoddeson:

What is curious to me is why people didn’t get the surface state idea before, because it was almost there in the late thirties in the work of Tamm and Shockley.

Maurer:

That’s right.

Hoddeson:

Although it wasn’t a real surface that they were talking about, but rather an ideal surface. Still the idea was there, but people didn’t put it together.

Maurer:

Tamm’s contribution here — this was a Russian contribution that was really important to surface states. And Shockley’s contribution was good. Actually, it was just a failure, I think on their part, to have the kind of insight that Bardeen had into the problem. Bardeen didn’t really solve the problem as to whether these surface states were surface states which were there because the surface was imperfect, or because they were there because there were some defects on the surface. It’s simply possible with surface states to have the properties of the surface that Bardeen found. It’s just very common in physics, and all science, that somebody makes an advance and carries it so far and then stops, and somebody else sees the implications, It’s more a psychological problem.

Hoddeson:

In the transistor story, it’s very curious, because Shockley was wondering since 1945 why the field effect calculations didn’t describe what he was observing. But he had himself done some work earlier on surface states; it took Bardeen to find the explanation. You would think that Shockley would have thought of it, but he didn’t.

Maurer:

In one sense, I think that when a person makes a really good stroke, when he really achieves something, there’s a letdown, and the very f act that you have made that achievement makes it difficult immediately to push ahead and exploit the consequences of it.

Hoddeson:

That’s very interesting. I’ll have to check that theory out against other examples.

Maurer:

I think there are many, many examples of it. One aspect of the history of solid-state during the thirties that I think deserves very great attention is the fact that it was still not really appreciated to what an extent defects and impurities determined the real properties of crystals. And not merely in connection with electronic phenomena, but in connection with metallurgical and mechanical properties of solids. Of course, the concept of a dislocation came very late. I remember in '40-'41, let’s see, '42, as I remember, Seitz was presenting the ideas to the metallurgists at the University of Pennsylvania. And they were new to the metallurgists. They developed during the thirties.

Hoddeson:

There are some papers of Prandtl in the twenties, by Orowan in the thirties, by Taylor in about '35 …

Maurer:

Those were probably the critical papers, although they were building upon Volterra’s work

Hoddeson:

Yes, Volterra, very early in the century.

Maurer:

Volterra’s work, though, was really fairly abstract; considerations concerning continuous media and not crystals. But I don’t know the early history of this. I do know that in the middle thirties, when I was a graduate student, we were not learning about dislocations, although we were then learning about the quantum-mechanical electronic properties of crystals. The role of defects and impurities in determining physical properties was very poorly understood, right up to World War II. Some of the work of the Germans — Pohl at Gottingen and Carl Wagner, and Schottky — was very, very critical, of tremendous importance in focusing attention upon the fact that there were point defects in crystals, that in ionic crystals these point defects could trap electrons and therefore influence very strongly the electrical and the optical properties of materials. After World War II, in 1946, '47, we started at Carnegie-Melon to use radioactive tracers to study the self-diffusion of ions like sodium in sodium chloride. We were building upon some earlier work that had been started by Wagner and interrupted in Germany by World War II. Some people had used radioactive tracers, naturally radioactive materials, to study diffusion in primarily metals, very early in the century.

Hoddeson:

Where?

Maurer:

Largely in Germany. But, with the availability of artificially radioactive materials, you had much more scope, and the work of people like Wagner had shown already that the defects and impurities were important. But, when we started the work, the first of it was carried out by Dillon Mapother as a graduate student. It was still a matter of controversy whether the low-temperature mobility of an ion in a crystal like sodium chloride was due to some impurity-induced defect, or to grosser macroscopic defects like microscopic cracks in the crystal. And it was really Mapother’s work which demonstrated that the accidental concentration of impurities determined the low-temperature diffusion coefficient of sodium

Hoddeson:

Do you know how the work got started?

Maurer:

It got started because Seitz was aware of the work of Pohl and Wagner and Schottky in Germany. And we were also aware of the opportunities — that the availability of artificial radioactive materials conveyed a new scope for experiments. For instance, immediately after the war, I had an opportunity to talk with Carl Wagner who was able to come to this country under the auspices of the United States Army. And he told me of experiments on the electrical conductivity of alkali halide crystals that he and Huckel had done in Germany, which had been interrupted by the war. And I told him of our intention to study the ionic conductivity of doped crystals and the diffusion coefficient. And Wagner encouraged me to go ahead, but he said at the same time he would try to publish what they had already done as quickly as possible He was really a tremendous source of ideas. But then, there was another graduate student at Carnegie named Howard Etzel. Etzel did a conceptually very simple Ph.D. thesis. He Simply measured the electrical conductivity of sodium chloride as a function of a deliberately added small amount of cadmium, and demonstrated that the theoretical concepts of Wagner and Schottky, which were simply based upon well-established chemical thermodynamics, worked beautifully The only really difficult part of the work was the purely chemical part, determining the concentration of deliberately added impurity, which was in the range of from 10 to — let’s see, we worked about one part in a thousand — the order of one part in a thousand atoms. In the case of cadmium, it turned out polarographic techniques could be used rather well. We had help there from people at the Naval Research Laboratory. But that was a classic paper, which was important for about ten years, and yet it was very, very simple in concept. But, it’s hard to understand today how bewildering the erratic behavior of crystals appeared to be during the thirties because one didn’t really understand the role played by impurities.

Hoddeson:

What I would like to do now is go through your life, considering highlights of your work in chronological order. As you think of more general comments on solid-state physics, please make them.

Maurer:

That’s fine, that’s fine, but you must always remember two things. First, my memory isn’t that good. Secondly, I’ve never worried much about the past, and so I’m not one of those people who kept meticulous records, either mental or physical, of what went on.

Hoddeson:

Few people have. I see that you were born in Rochester, New York, in 1913, on March 26th. I’d like to know just a little bit about your parents and your background in general, so that I can get an idea of how your interests in physics began.

Maurer:

Well, it’s not particularly relevant, but my father was the son of a man who migrated from Alsace to America in 1871, because he was in trouble with the Prussians. The intention was that he would stay in the United States for a few years until the trouble blew over. The trouble was, he married a German girl in Utica, New York, and his family disowned him for at least 20 years, so he stayed in the United States. My father was a bookkeeper at the Eastman Kodak Company and eventually an office manager. His education stopped at the grammar-school level. Like many men at that time, he took night school courses to prepare himself. My mother had nothing but a grammar-school education. They had five children. They were grimly determined that we were all going to be educated.

Hoddeson:

And were you all educated?

Maurer:

More or less. The three boys all went through universities. I have one brother who is a Professor of Medieval Philosophy at the University of Toronto. The other is a physicist at the Eastman Kodak Company. The two girls didn’t fare so well. During the thirties, there wasn’t the money to send them to college. One sister is dead; one sister has been a secretary at the Eastman Kodak Company for a lifetime.

Hoddeson:

Did you go to public schools?

Maurer:

Well, no, no. We were Roman Catholics, and when I was a boy, we originally went to St. Bridget’s school, in Rochester, New York. By the time we entered high school, we had moved to the suburban part of Rochester, New York. My younger brothers and sisters went to a parochial school, St. Thomas in the town of Irondequoit. Eventually, the boys, including myself, went to Aquinas Institute, which was the Catholic high school in Rochester. My sisters to Nazareth Academy, which was the girl’s school.

Hoddeson:

How was the science education in Aquinas High School?

Maurer:

Terrible.

Hoddeson:

Did you have any interest in science?

Maurer:

Yes, oh yes, I took the chemistry and physics courses, and enjoyed them tremendously. And, at that time, in high school, I fixed my intention of becoming a scientist.

Hoddeson:

Do you remember any specific incident that was very important?

Maurer:

No, as I remember, neither the physics nor the chemistry course was competently taught. But it was very exciting. It stimulated one, and I guess that was the important thing. I and my brothers, with the encouragement of my parents, were tremendous readers, and we would just rake the public library for books. In high school, I was trying to read a terrible book called “Slosson’s Creative Chemistry”, which just thrilled me. I found all these books on chemistry in the library, and things like bowdlerized versions of Einstein’s theory of relativity terribly exciting and tried to understand them. Books on astronomy. But, I had a very great mish-mash of poorly digested ideas. I went to the University of Rochester because it was the only financially feasible place to go; I could live at home.

Hoddeson:

Were there any good professors there at the time?

Maurer:

Superb ones.

Hoddeson:

Who were important to you?

Maurer:

One of the most important was a man named Willard Line. He was a professor of chemistry. I became a chemistry major, and Line taught quantitative and qualitative analysis. It excited my imagination; it sounds strange now, but I loved it, and it was really a course in inorganic chemistry. In my sophomore year, I flunked German because I was spending all my time in the laboratories, and I reached a point where I was tutoring the other students in the class; knew the periodic table backwards and forwards. That was a tremendous influence

Hoddeson:

Who were some of the other professors or courses that played a role in your background?

Maurer:

Well, there was a very impressive man, Dr. Helmkamp, who taught organic chemistry and he was a very attractive and stimulating fellow; unfortunately, I didn’t enjoy organic chemistry. I didn’t have the intelligence or experience to realize that a man named Wigg, who was there, a physical chemist, was a really first-class research man. And I passed up the opportunity to work with him. But by the time I had reached my junior-senior year, I knew that I was more interested in physics than in chemistry and such, but the physics department at the University of Rochester was very poor at that time, and I wanted to do graduate work. Well, I took more math than would he normal, even to the extent of plunging into courses which were beyond my ability as an undergraduate. But there, I took very little physics, and started to do graduate work in chemistry, because there was an opportunity there to have a very poor assistantship, which could help me to support myself. However, at the end of one year of graduate work in chemistry, I heard that Lee DuBridge was coming to the University of Rochester to head up the physics department, and I hustled over to speak with him and ask him whether I could do graduate work in physics with him and also get some form of support from the physics department. He permitted me to become a graduate student in physics. Originally, I had an assistantship working for a man named T. R. Wilkins, who did nuclear physics, but I wanted to work with DuBridge. I’ve forgotten; I think it would be within that first year, DuBridge found a job for me at the Rochester Medical School servicing some ultraviolet equipment that they were using to irradiate bacteria, and I began a Ph.D. thesis on the photoelectric and optical properties of metals with DuBridge.

Hoddeson:

What was his major interest at that time?

Maurer:

At that time, DuBridge had established a reputation for work on the photoelectric and thermionic properties of metals, but he had seen the opportunities that were available in nuclear physics with the new types of accelerators, in particular, the cyclotron. And, when he came to Rochester, he raised on the order of 100-150,000 dollars, and began the construction of a small cyclotron, so that, in the late thirties, Rochester was one of the physics departments that was able to do some really quite frontier nuclear physics.

Hoddeson:

I haven’t looked at DuBridge’s background yet. Where was he trained?

Maurer:

Let’s see now. He came out of one of the small Midwestern colleges like, what’s this one up north in Minnesota? He might have come from Carlton College, one of these. He did a Ph.D. thesis, I believe, at Washington University, no; I’m not quite sure about that. Might have been at Berkeley. He was a post-doctoral fellow.

Hoddeson:

Well, I can look this up later.

Maurer:

Some of it is in his book, Photoelectric Phenomena. He was a tremendously able man, and the thing that he did at Rochester was he simply demonstrated that the right person in the position of headship can turn a department around in two years. Immediately after he came, he brought Sid Barmes from Cornell to build a cyclotron. He brought Fred Seitz; he brought Milton Plesset, who had just, was just out of Copenhagen, where he had been working with Bohr. When Seitz left, at the end of a couple of years to go to GE, he brought Weisskopf there. During the period from 1935 to 1939, when I left, the department at Rochester was just enormously stimulating. There was a man there, Brian O’Brian, head of the Institute of Optics, and he was a controversial, but very stimulating figure, and Stu Campbell, who worked there in optics, was simply an absolutely’ first-class scientist; so during that period, it was an opportunity that was enormous to be able to work there.

Hoddeson:

Was there any interaction at Rochester between the university and Eastman Kodak?

Maurer:

It was a very loose connection. People like Rudolf Kingslake, a British optical man, associated with work at the Institute of Optics; I really don’t know what the formal arrangements were. Eastman Kodak and Bausch & Lomb loaned equipment, I think, in some cases, gave equipment, to the Institute of Optics. I believe their financial support was responsible for establishing the Institute. But, as a graduate student, we had very little contact with the people at Eastman or at Bausch & Lomb.

Hoddeson:

Were there visitors to the university frequently in those years? Do you remember any particularly impressive talks that were given? Sometimes people have a tremendous memory of someone coming who changed their perspectives.

Maurer:

Well, if you want a ridiculous anecdote. one of the things that does stick in my memory, was when I was an undergraduate and a sophomore, the University sponsored a series of lectures on the topic of general relativity since 1915, I think, or 1917, by a man, Ludwig Silberstein; he was an elderly German, who had been associated with Einstein. He gave a series of about 8 lectures and was so enthusiastic that he occasionally lapsed into German. He had an audience of about 10 people for his first lecture. By the time he reached the third lecture, his audience was down to 2: the head of the physics department, T. R. Wilkins, and myself. And I carefully took notes on everything that he said, and I understood practically nothing, but it was exciting.

Hoddeson:

Did you have any summer jobs in those years that were important in your career?

Maurer:

Well, it was very difficult to obtain summer work during the Depression, and I continued to use the summers to do my thesis work at the university. The only outside job I had all the time I was an undergraduate, and part of the time when I was a graduate student, I babysat. We lived in a suburban area, and our neighbors liked a good, steady source of babysitters; it was 50 cents a night. But, in ‘39, I left Rochester with my Ph.D.

Hoddeson:

Let’s see. The paper on the solubility of silver in mercury was before your thesis.

Maurer:

I started that work as an undergraduate senior project. In order to get a degree with, what did we call it, with honor, you had to do a senior thesis.

Hoddeson:

I see.

Maurer:

And, I did this under a very nice, very decent physical chemist named [Arthur] Sunier, who was the most meticulous person. I don’t think he was very creative, but he was exceedingly precise and careful. And I had a lot of fun doing that. What I managed to accomplish was adequate for the senior thesis. But then, during the one year of graduate work as a chemist, I continued the measurements and actually refined them a good deal, so that at the end of that year, there was this very, very modest paper [“The Solubility of Silver in Mercury III,” Robert J. Maurer, J. Phys. Chem., 42, 515-519 (1938)]....

Hoddeson:

It’s largely data.

Maurer:

It’s simply a measurement of the solubility; there is no theory in it whatsoever. It was a very modest piece of work.

Hoddeson:

Now, the paper on photoelectric an optical properties of sodium-barium, is this your thesis? [“The Photoelectric and Optical Properties of Sodium and Barium,” Robert J. Maurer, Phys. Rev., 57, 653-658 (1940).]

Maurer:

That was the Ph.D. thesis.

Hoddeson:

And I gather, this was really the beginning of your long interest in photoelectric and optical properties.

Maurer:

Well, I knew by the time I finished that that I wanted to work in solid-state physics. I wasn’t at all sure where I wanted to work. I had a talk with people at GE about the possibility of a job there, but I found that they weren’t really very much interested in somebody who was thinking of work on what we would now call semiconductors, because I had become intrigued by the possibilities of semiconductors. But, I found that the electronics industry wanted people to work on vacuum tubes and they would tell me “Well, if you’re interested in looking at secondary emission in tetrodes, we might have a job, but if you’re interested in solid-state electrical conductivity, I don’t think we’ve got anything.”

Hoddeson:

But this was already at the beginning of the war, when people were looking at radar crystals. One would imagine the industries to be right up there.

Maurer:

I don’t really understand it exactly. You see, I had a contact with GE. The last two years as a graduate student, I had what was called a Coffin Fellowship, a reasonably prestigious award actually, because of their scarcity. And that was supported; my stipend was supported, by the General Electric Company. So as a matter of course, they invited me to come and interview for a possibility of a position.

Hoddeson:

They were then at Schenectady?

Maurer:

They were at Schenectady. And one of the people I talked to was Saul Dushman, whom I knew by reputation, a very impressive man. But as a result of talking with him, I came away with little feeling that there was any interest at GE in the sort of thing that I was interested in. In the meantime, really, it was DuBridge who, with his contacts at MIT, suggested applying for a post-doctoral appointment there in the electrical engineering department actually, with a German named Arthur von Ripple, who had relatively recently come to the United States from Germany, and who headed up the group called the Laboratory for Insulation Research. But Von Ripple’s interests were very broad, and extended all the way from fairly conventional studies of properties of insulating materials to dielectric breakdown of crystals, and the mechanisms of electrical conductivity in crystals.

Hoddeson:

Do you remember which institutions he came out of?

Maurer:

Well, he came out of the mainstream of ideas in places like Gottingen. He was married to James Franck’s daughter, Franck was his father-in-law. I’m sure he very much influenced by Pohl and Wagner and Schottky and so on. He wanted me to work on dielectric breakdown, which was actually not of very great interest to me. But I made an arrangement with him that I spend part of my time looking at some semiconductor phenomena. I had to pick something easy and relatively straightforward to do because I didn’t have a lot of time to put on it. So I worked on cuprous iodide.

Maurer:

I don’t think that von Hippel took very much interest in what I wanted to do. Copper oxide was one of the few semiconductors like selenium that had found a practical use in industry as a rectifier.

Hoddeson:

That was due to Grondahl’s discoveries, mainly.

Maurer:

Actually, a lot of the interpretation was due to people like Wagner and Schottky. They were the people who had some understanding of what was going on I didn’t feel that I could study copper oxide itself, but cuprous iodide was an easier material to prepare and to manipulate. Unfortunately, it turned out that the amounts of excess iodine which it easily picks up are so great that the phenomena in cuprous iodide are really quite complicated.

Hoddeson:

Let’s see, this paper, on cuprous oxide work [“Deviations from. Stoichiometric Proportions in Cuprous Iodide, Robert J. Maurer, Journal of Chemical Physics, 13, 8, pp. 321-326 (August, 1945)], this was published in ‘45, but I gather you did it before.

Maurer:

Yes. That was all done around 1941-42, before I left MIT.

Hoddeson:

Publication was held up during the war?

Maurer:

I forgot, I suppose so. I really have forgotten, I imagine that I pretty much forgot it after I left. You see, in ‘42, I got involved in war work at the University of Pennsylvania, and everything else got pushed to one side.

Hoddeson:

Right. Well, let’s discuss that separately. Let’s discuss the work done in ’40-41, before you began the war work. Let’s talk about that cuprous oxide work. Now, you chose it because of this K. Badeker in 1905.

Maurer:

Yes. Badeker had done some very early work, and was in fact one of the people who observed the anomalous Hall effect in cuprous iodide, which was completely un-understood at that time. But what was clear was that you could put excess iodine into cuprous iodide and you could influence the electrical conductivity tremendously. Well, I felt that a rather fundamental problem was not just to expose the material to different external pressures of the vapor of iodine and measure the conductivity, as had been done with oxygen and cuprous oxide, but actually to measure how much iodine was taken up by the cuprous iodide and relate the effect on the conductivity to the actual amount of excess iodine present, rather than to some more remote variable, like the pressure of the vapor in equilibrium with the material.

Hoddeson:

So this is already the heart of the concept of the imperfection, in a way.

Maurer:

Yes. Well, one would like to know if one atom of iodine enters cuprous iodide, how much effect it has on the conductivity.

Hoddeson:

Were people thinking in those terms at that time?

Maurer:

Oh yes, oh yes. See, Wagner and Schottky had developed theories. Since people weren’t able to measure the actual amount of, say, oxygen picked up by cuprous oxide, what they did was develop their theories in terms of the relationship between the pressure of oxygen in equilibrium with cuprous oxide and the conductivity. You tried to develop a theory of how the conductivity was a function of the vapor pressure, the equilibrium vapor pressure of the gas, say oxygen in contact with cuprous oxide. Now the reason for doing it that way was it was too difficult to measure the actual amount of oxygen which had been absorbed by the cuprous oxide. But in the case of iodine, first of all, it was obvious that a lot of iodine was picked up. And secondly, the iodine atom is heavy, so you can actual1y weigh the amount of iodine that is picked up by the cuprous oxide, and I built a microbalance that could operate in an atmosphere of iodine. I hung a strip of cuprous iodide on the end of the balance, and then I varied the pressure of iodine of vapor, and actually found out how much iodine was in the cuprous iodine as a function of the vapor pressure of iodine in equilibrium.

Hoddeson:

This balance was developed by J. Blewett. Is that the same Blewett who then went on to do great work in linear accelerators?

Maurer:

I think so. I don’t remember.

Hoddeson:

He must have been at GE at that time.

Maurer:

Yes, that’s right.

Hoddeson:

— if it’s the same man.

Maurer:

I’m quite sure it was, quite sure it was, because one of the graduate students at Rochester at that time was a girl who subsequently married Blewett —

Hoddeson:

Hildred?

Maurer:

That’s right, that’s right. She worked in the Institute of Optics, if I remember. But the …

Hoddeson:

Had anybody worked with cuprous iodide before? I mean, did you choose this, or did von Hippel point out that this particular substance …

Maurer:

Well, I picked cuprous iodide because, in looking through the literature, I didn’t think I could work with cuprous oxide, because of the high temperatures and the small amount of oxygen that was picked up, but I found the cuprous iodide business in the literature, and I made the usual calculations and concluded that I could make the measurements on cuprous iodide. Now the very thing that makes it possible to make the measurements makes it very difficult to interpret the measurements. Cuprous iodide absorbs a lot of iodine. That means there are a lot of defects in it. When you have so many defects, they interact with each other, and the whole interpretation becomes too difficult, so you’re on the horns of the dilemma. You must remember that at this time, the Germans had published quite a bit of data on various semiconducting materials and how their conductivity changed as a function of the pressure of a gas in contact with them. Not just cuprous oxide, but tin oxide, things like that. It was a relatively simple experiment. You would put the sample with leads in a tube. If it’s an oxide, you filled the tube with oxygen, you heated it up to some temperature, you now could vary the temperature and vary the pressure of oxygen independently, and you measured the conductivity. And they did a lot of that, .and they developed a thermodynamic, statistical-mechanical theory which is really just chemical theory applied to electrons. But I wanted to get the direct relationship between, to be able to write that the electrical conductivity is a function of a number of excess iodine atoms. Well, it was fun, and at a later date, at Carnegie at one point, I picked up the subject again, and did much more, had a graduate student who did some much more extensive work on it. But it still is not a sufficiently simple system to be really interesting, and in the meantime, the elemental semiconductors had become, had almost erased interest in everything else. They were so important.

Hoddeson:

Let’s see, there are two more subjects which you did work on in this period. One you mentioned, the electric breakdown of glasses and crystals [“Electric Breakdown of Glasses and Crystals as a Function of Temperature,” A. von Hippel and R. J. Maurer, Physical Review, 59, pp. 820-823 (May 15, 1941)], but you also did work on the deviations from Ohm’s law in soda lime glass at MIT [“Deviations from Ohm’s Law in Soda Lime Glass,” R. J. Maurer, Journal of Chemical Physics, 9, pp. 579-584 (August, 1941)]; these are worth commenting on in detail.

Maurer:

Well, the dielectric breakdown was done because of von Hippel’s interest in the subject. He had a theory of dielectric breakdown, and wanted data to test it, so several of us worked over a long period of time on that problem. In the course of making dielectric breakdown measurements, we always monitored the current through the sample as a function of the electric field. That was done partly, or even largely, to insure that the experimental conditions were what we thought they were, that the sample was properly positioned between the electrodes, that the insulating wax around it was in place, there weren’t air bubbles, things like that. But I realized that one could make some use of these current-voltage curves to test a relatively simple extension of Ohm’s law. People had pointed out that Ohm’s law was really an approximation and that, at very high fields, it should break down, it should begin, and the relationship should become what is known as a hyperbolic. Sine relationship between the current and voltage, and then at very high fields, the relationship should become exponential. I observed such behavior in crystals, and then realized that it was far easier to make measurements upon glass, because the glass could be blown into a shape that was ideal for making measurements, and avoiding some of the nasty experimental difficulties of crystals. So I made measurements on the soda lime glass, and was just delighted to find that I could fit the current-voltage curves to the theoretical form beautifully. Unfortunately, one of the parameters didn’t seem to be physically reasonable and I invented an absurd theory to account for that. It led to a dispute with Onsager at one point, where I think neither of us really understood what the other was talking about.

Hoddeson:

What struck me in looking at this work was that here again you run into the role of the deviation from the periodic structure.

Maurer:

Well, the current is carried in glass, typically in soda lime glass, by the sodium ions, and here one has an irregular structure, a random network, and you have the sodium ion jumping around inside of it and getting a directed motion due to the electric field. I was already at this point becoming interested in the details of ionic mobility: how do ions move around inside of a crystal or a glass? It’s obvious that they do; the process is very important for things like the buildup of oxide layers. It was clear that defects were involved, and the defects could not only affect the ionic motion, but also, when electrons were present, the electronic motion. So, I don’t think I had any very grand program at that time; I seized upon the glass mostly because it was readily available, and I could just test a really small thing, the deviations from Ohm’s law. Did they really occur the way the simple theory predicted it should?

Hoddeson:

You note in the paper that similar work was going on in Britain in the British Electrical and Allied Industrial Research Lab. I was wondering whether that was an industrial laboratory.

Maurer:

I think it was one of those laboratories that were supported by industry, but I really don’t remember anything about it. There was a good group that was doing dielectric break down, that was a very serious group.

Hoddeson:

There was an interesting little argument in the literature, in Phys. Rev., between the group of Austin and Whitehead, who were at this British Electrical and Allied Industrial Research station, and they were objecting to, I guess, the edge effect argument.

Maurer:

The truth is that it was very difficult to make dielectric breakdown measurements that were meaningful, and there was always a question as to whether you were doing it right. On top of that, there was no good theory of what was happening. One of the more interesting theories was created by these British in the area of very, very thin samples. I think, in retrospect, it’s clear the phenomenon was just too complicated to be handled at the time. And I’ve got to confess that I never could get terribly excited; I like to work on very simple things, and this was too complex for me.

Hoddeson:

Before we move to Pennsylvania, I just have one more question about MIT, an institutional question. You were working in the circle of Arthur von Hippel, and there was also the very important group of Slater working at that time. I was wondering whether there was interaction between the two groups, and how the groups differed from each other — I’m trying to gather information about the environment for doing solid-state research at MIT at that time.

Maurer:

Well, it was a very favorable environment. Everything was really very informal. We went to all of the physics colloquia, and their EE colloquia that were just as interesting as the physics ones. That was a very exciting period because people were developing analog computers at MIT, for instance, performing mechanical integration. In addition to the people in Slater’s group, there was a very superb group in X-ray crystallography.

Hoddeson:

Who was the principal person?

Maurer:

M.J. Buerger. And we had close contact with them. Von Hippel’s laboratory was one of the first of the so-called “interdisciplinary” laboratories.

Hoddeson:

How large was it?

Maurer:

Oh, not very big. There was von Hippel; he usually had a couple of post-docs, and about, I would say, four graduate students. Usually, there was one of two people around who were guests of the group and using the equipment, who were nominally attached to maybe Buerger’s group or some other. Four graduate students would be a little small, more like, I would say, six or seven. And there were chemists. Jim Schulman, who recently retired from the Naval Research Laboratory, was a chemist; he was working on selenium and electrically depositing selenium layers to make photo-cells. There was an X-ray crystallographer, there were electrical engineers, like Gordon Lee, who developed one of the first fast, really fast cathode-ray systems, cathode ray tubes, capable of actually observing, detecting a microwave. Demetrius Gelatis, one of the physicists …

Hoddeson:

Was Fan there?

Maurer:

Who?

Hoddeson:

H. Y. Fan. He worked with von Hippel, but maybe not in that period.

Maurer:

Not in that period, not in that period, no. That must have been later. I was there for three years, and I was sort of the second wave. There had been one man whose name I’ve forgotten, who was a post-doc before me Shepherd Roberts, who did some very, very excellent work on the titanates, the barium and stratium titanates, and eventually went on to GE to be very successful there. He and von Hippel did some superb work on the titanates at that time, recognizing their remarkable dielectric qualities and what they can do, the fact that they have very practical applications But Shepherd Roberts and I were there both in a sort of post-doctoral position, and we were the second wave. I left in ‘42, and it was during the war that the lab really took off and became very large, and had a much larger program, a much broader program. The lab was still in a rather primitive state when I was there. It was a very exciting atmosphere, because we interacted the crystallographers, the chemists, the physicists, the metallurgists, and we fought with them. They were always trying to take our space away from us, and we were always fighting to keep it. You knew that if you let somebody’s nose in the door, why, pretty soon, he’d walk off with the lab. But it was all done in a sort of friendly, competitive spirit. And the atmosphere, especially in the physics department, was just very exciting. I took the opportunity at that time, for example, to sit in on Slater’s lectures. Slater was probably the finest lecturer that I’ve ever listened to. His lectures are a hundred times better than his books, and his books are very good. He was one of the few lecturers who talked in complete sentences which were grammatically correct, and who used the blackboard to its full advantage. He was a superb teacher.

Hoddeson:

Did you know Shockley?

Maurer:

Yes, Shockley was around at that time, I don’t really remember anything in detail. I remember things like one of the crystallographers they were doing early work on electron emission from points, and imaging the electrons on a florescent screen inside of a sphere that surrounded the point. This, of course, eventually became a technique for studying surfaces. I remember Slater was going to show one of these, and he had a sphere, and it was covered with cloth, and when he took off the cloth, Shockley had painted a face on the sphere. Shockley was notorious for a whole series of pranks, but I don’t remember that at that time I had any real contact with him as a person.

Hoddeson:

What about the other members of the Slater group, people like Krutter?

Maurer:

Who?

Hoddeson:

Harry Krutter was in Slater’s group at that time.

Maurer:

No, I was primarily interested in the calculations that the group was making. I don’t remember that I really knew any of the people as such. I was very naive, very shy, and didn’t really make friends very easily. I didn’t reach out very easily.

Hoddeson:

There was another important group around that I am aware of, the group surrounding Van Vleck at Harvard. Was there interaction between your researchers and...

Maurer:

No, we would occasionally go over to Harvard, of course, to hear a colloquium or seminar, but between our group and that, I think there was very, very little interaction.

Hoddeson:

In the mid-thirties, there had been strong contacts between the group around Slater and the group around Wigner, at Princeton. I was wondering, whether in this later period, when you were at MIT, there were still active contacts. Or had that passed by then?

Maurer:

I don’t know.

Hoddeson:

We move then on to the University of Pennsylvania. This is where you begin to do war work, isn’t it?

Maurer:

Yes. But what happened was that I had this post-doctoral appointment; at the end of three years, you expect to move on. However, MIT was expanding its effort; the Radiation Lab was beginning, and Slater offered me an assistant professorship in physics. The understanding was I would work with Wayne Nottingham, and I knew Wayne pretty well at that time, because Wayne was interested in working on the surface electron emission. I had done my thesis in this area, so I knew Wayne even before I came to MIT.

Hoddeson:

I would like to know more about Wayne Nottingham; he was apparently very, important in that period.

Maurer:

Oh, he was a tremendously important person.

Hoddeson:

There doesn’t seem to be much information about him that’s easily accessible. Or maybe you know where I can find some?

Maurer:

Well, I can tell you a few anecdotes. He was a complete experimentalist, He was a most careful, meticulous and thoughtful person. Every one of his graduate students had to begin in life by making what was called a Compton electrometer. In those days, electronic circuits for measuring small currents were just appearing, and a lot of work was still done by using what was called a Compton electrometer, which was one of the most bastardly, instruments ever invented. Before Wayne would let anybody do a thesis, he had to build his Compton electrometer, which he would then use in measuring small currents. That was a real introduction. People used to say of Wayne that what you ought to do is put him in a room with a very small window and give him slips of paper, telling him what measurements you want, and that he ought to pass out little slips of paper with the data, because Wayne was not very strong theoretically, but he was tremendous as far as intuition was concerned. He was one of the people who really introduced high- vacuum techniques into surface physics. He had a real appreciation of the fact that even at a pressure of 10 8 mm of mercury, you would find it hard to keep a surface clean. He was a rather arrogant person. He used to offend the people at GE by coming and telling them that their vacuum technique was inadequate. But his own techniques were good, and he was very meticulous and careful. He was very single-minded. I remember one incident during the war; he had made some surface measurements, and he had attempted an interpretation of them, and when he presented this before a group, a kind of information meeting of people, I think it was Weisskopf who was present. And when Nottingham was finished, Weisskopf got up and said, “Wayne, I can’t accept your interpretation; it violates the fundamental principles of quantum mechanics.” And Wayne leaped to his feet, and said, “So much the worse for quantum mechanics!” The room just broke down with laughter. But that was typical of Wayne. He was really a very, very nice person. I didn’t want to work with him, however, because I knew that I would be a pair of hands; he belonged to the old school, and his graduate students and post-docs did what Wayne Nottingham told them to do. So, I was very pleased when Fred Seitz offered me a chance to go to Pennsylvania.

Hoddeson:

Just one more question about Nottingham. I understand that the Nottingham conferences were very important in the history of that period.

Maurer:

This was where the leading edge of surface physics was discussed at that time.

Hoddeson:

Were they going on when you were there?

Maurer:

I guess so, I don’t really remember. There was all of this work, for instance, the problem of thermionic emission from a surface whose work function varies from point to point, patchy surfaces, enormous controversies. All of that Nottingham was in the thick of.

Hoddeson:

Maybe there are records left behind of those conferences someplace.

Maurer:

There probably are. Actually, an awful lot, of these problems were eventually solved by Leroy Apker at GE. Leroy Apker did the really definitive work on tungsten, thermionic emission from tungsten.

Hoddeson:

Is he alive?

Maurer:

No, he killed himself. Leroy Apker was a graduate student with me in Rochester, and then I went to MIT, and about a year later, he went to GE. At GE, the first thing he did was to do the thermionic emission from tungsten properly. He did the definitive work. He got the, what is it called, the solid-state physics award?

Hoddeson:

Buckley prize?

Maurer:

The Buckley prize, mostly for that work.

Hoddeson:

Seitz invited you to come to Penn.

Maurer:

You see, I had known Seitz since I was a graduate student at Rochester and he was an instructor there. He gave us a course, for instance, in solid-state physics.

Hoddeson:

He was writing his book then.

Maurer:

That’s right. It was a very bad course, because he was writing his book and spending all his time on his book, and the course was way over our heads. But he was a most stimulating person, and I kept in contact with him after I went to MIT. In fact, one of the contacts was this amusing one; I submitted a paper, I don’t really remember which paper it was, and I got it back from the referee with a lot of, I thought, very nitpicking comments, punctuation was criticized, things like that. And it happened that just after I had got the paper back, I was in New York at a Physical Society meeting, I guess at Columbia University. As I walked up the steps of one of the buildings, I met Fred Seitz, said hello., and he asked me how things were going, and I said, “Oh, I just got a paper back from a nitpicking referee, who I don’t think really understands what the paper’s about, but who’s death on punctuation.” And Fred smiled and said, “I’m the referee.” Well, it was at that meeting that he suggested I come to Penn. He was trying to put together a solid-state group there. At Penn, the facilities were hopeless. The physics department was in a building that had been an orphanage for girls. The Randall Morgan Laboratory of Physics had a fine ring to it, but it was an old structure, a minimum of remodeling. One of the first things I had a graduate student there do was dig a trench and lay a cable across one of the yards into one of the basements so we’d have electrical power. The cockroaches were two to three inches long. I had a graduate student who wouldn’t work at night because he was scared of the cockroaches. It was an incredible place, but it had marvelous people. Gaylord Harnwell was there; unfortunately, he left. He was head of the department, and I shook hands with him; he greeted me when I came. Six months later, I shook hands with him again, the second time I saw him; I was leaving. I was only there for about six months.

Hoddeson:

Who else was important there?

Maurer:

Park Miller and Andy Lawson.

Hoddeson:

Andy Lawson is also not alive anymore?

Maurer:

No, I think Andy died down at Riverside; he ended up as head of the department at Riverside. Park Miller is still in California, he’s teaching at one of the California colleges. Those were just very exciting young people to be with, full of ideas, very, very smart physicists, clever as the devil. And life was just exceedingly erratic and informal, because we had, we couldn’t do anything right. We didn’t have the resources. Leonard Shiff took over as head of the department, and Gaylord Harnwell went to the underwater sound laboratory in San Diego to head that up. Leonard Shiff was so young, and looked so young, that one day, one of the officious people tried to shove him into a line of freshmen who were signing up for the draft. We had a couple of older people on the faculty, but these young men, most of them under thirty, ran the department. It was a remarkable department. I was only there for six months; I went there in June, I left the following winter.

Hoddeson:

While you were there, you wrote a paper on the electrical properties of semiconductors [“The. Electrical Properties of Semi-Conductors,” Robert J. Maurer, Journal of Applied Physics, 16, pp. 563-570, (1945)], a review.

Maurer:

That was just one of those reviews. Seitz would get me involved in doing those things, because he was always doing them.

Hoddeson:

What were the major subjects of interest at that time?

Maurer:

Well, at Penn, the major interest was that the radar had appeared and the only decent detector that they had for radar waves was the silicon diode. People were working on diodes, vacuum diodes that might work, but didn’t have the properties; you couldn’t make them small enough. The old crystal detector from the early days of radio, with the cat’s whisker, used galena, zinc sulphide. Now people had hit upon silicon. We didn’t know much about it at the time, but Bell Labs was really already well into the silicon business.

Hoddeson:

They came on silicon about 1939, when Russell Ohl did it.

Maurer:

That was a remarkably far-sighted set of ideas, you know, to go into studying silicon.

Hoddeson:

But that is where it began. I came across that when I was looking at the background of the transistor.

Maurer:

As far as I know, if you go way back, you’ll find there were people like Badekerand this man at Cornell, who studied germanium. But nobody really did a serious job until the Bell Labs people picked up silicon, and that was a remarkable bit of far-sighted thought.

Hoddeson:

Were you working on silicon?

Maurer:

We were actually taking silicon and making diodes with them, and trying to understand what it was that made a good diode with a good rectification characteristic. We didn’t know whether the key element was the silicon or the cat’s whisker or the way the cat’s whisker was put on the silicon. We used to sit in front of oscilloscopes, and you would put a tungsten wire down a silicon wafer, and get very poor rectification. And you’d tap it, and the rectification would improve, tap it some more, still better rectification, one more tap, no good, no more rectification. And we used to talk about our calibrated hammers with which we tapped the devices. We had a very imperfect understanding of the role of impurities, we really didn’t understand that much of the silicon that we had was so heavily impure that you really couldn’t make any strong, sensible conclusions from measurements made on it. We were trying to do some fundamental measurements on the properties of silicon and germanium.

Hoddeson:

You worked on germanium too?

Maurer:

Yes. I measured the work function difference of p- and n-type germanium because I could handle germanium. Silicon, with its higher melting point, was more difficult to deal with. Park Miller, for example, tried to measure the work function of silicon by a thermionic technique which really didn’t work, but he gave it a good try. I was there about six months, and had this one graduate student who doing a thesis. Actually, our results went into a report, but I think they disappeared. Meyerhof, later on, right after the war, I believe, made the same measurements and showed that p- and n-type germanium have the same work function.

Hoddeson:

This report, you said disappeared. Was it a war-time report?

Maurer:

Every month, we’d write reports about what we were doing.

Hoddeson:

It would be interesting to try to find out if they exist anywhere.

Maurer:

Oh, they’re probably in warehouses, somewhere in Washington.

Hoddeson:

Were they OSRD reports?

Maurer:

We were sub-contracting to the radiation lab; we were working under a sub-contract with the MIT radiation lab, so they, probably ended up at MIT.

Hoddeson:

They would be worth looking for.

Maurer:

Yes, probably. There are things there which, you know, just got buried. On the other hand, there’s an awful lot of verbiage to read, and I’m not sure that the cost-benefit ratio is very favorable.

Hoddeson:

But it’s a good idea to have these papers, because then one can verify people’s recollections. [pause]

Maurer:

I used to know that the reason that Germany had so many great musicians and composers was simply that everybody learned to play a musical instrument. Most of them played musical instruments; I’m sure, rather badly. And there was a lot of composition which was done badly, but this provides the total environment within which a few people of exceptional talent can do exceptional things. If you try to support just the exceptional people, it doesn’t work. For one thing, you can’t identify them. But it makes collecting history like this just — you’re really going back and lifting the cobwebs off of a lot of stuff, most of which might as well remain covered with cobwebs.

Hoddeson:

I think that, unless you do that, it’s very hard to identify the major developments, the ones that really shaped the frameworks that were important in any given period.

Maurer:

I don’t really think so. My own feeling is that if you went to about, probably as many as twenty or twenty-five main people, it would have to be an international group. That’s what’s important.

Hoddeson:

Well, we’re doing that too. Sometime I can show you our list of people we definitely don’t want to miss interviewing. But then there are other people who played a role in the growth of the field, who made important contributions, who may not have been in the small first rank category, but who nevertheless contributed.

Maurer:

It may have been important at the time, but the importance wasn’t really that major even then, and in retrospect, you can really afford to pass it by. I think you need about, as I say, twenty to twenty-five people who were the prime people during the period.

Hoddeson:

Not all of them are good interview subjects.

Maurer:

Are you talking with Dave Lazarus?

Hoddeson:

We hope to do so next fall. Dave Lazarus came out of the Chicago group, didn’t he? And Zener and Cyril Smith were also in that group.

Maurer:

Dave had worked on diffusion and defects in crystals and elastic properties in crystals, and when he came here about 1949, he started to work on diffusion in metals, and certainly in the post-war period, he was one of the two or three prime people in the field of diffusion. So, his knowledge not only of that period but of the immediately preceding period, the war-time period, is probably rather valuable.

Hoddeson:

Yes.

Maurer:

Is Smakula still alive? Smakula was an important German physicist, who came to the United States after the war, and really carried on the von Hippel laboratory; he took over the von Hippel laboratory. And it was the beginning; it was really the groundwork for the present materials research laboratory. Smakula was very much interested in crystal growing at the time he was at MIT, but Smakula’s memory of the period of the thirties, when he was in Germany and closely associated with all of the developments of Wagner and Pohl and Schottky — it would be very valuable. But I’m afraid he’s dead. His name was associated with a very, very famous formula for relating the number of defects to the optical absorption, the intensity of optical absorption. The Smakula formula was the formula relating the optical absorption by F- centers to the number of F-centers in the crystal.