Oral-History:J. T. Wallmark

About J. T. Wallmark

J. Torkel Wallmark

J. T. Wallmark (1919-2007) received an electrical engineering degree at the Royal Institute of Technology (1944), where he researched electron use in vacuum tubes and gas tubes under Hannes Alfvén. This work contributed to understanding of the solar atmosphere and led to the development of the Trochotron tube, a decimal counter for computers, which was manufactured by Raytheon. After a year of industrial employment with Standard Radio, Wallmark continued his graduate studies with Alfven, developing and testing tubes as Alfven's research assistant. In 1947-1948, a stipend from the Swedish-American Foundation put Wallmark in residence at the RCA laboratories in Princeton, NJ, where he submitted multiple patents and participated in development of beam tubes for high frequency applications. Wallmark's award-winning work on the electron beam tube was published in his 1953 dissertation.

After completion of his graduate studies, Wallmark returned to RCA, a decision shaped by his desire to work with transistors. This interview details Wallmark's projects and patents at RCA, including work on transistors, semiconductors, and memory devices. Wallmark describes his early confidence in the importance of transistors, and he details the work environments of RCA Labs.

From 1964 through 1966, Wallmark took a leave from RCA and worked at Chalmers University in Sweden. He describes the research secrecy of industries and universities that impeded semiconductor work during this period. In response, Wallmark created spin-off companies for the developments he made in his laboratory research. This interview also details Wallmark's creation of the Innovation Center to support commercialization of research, and he analyzes the interactions of industry and university.

Other topics covered in the interview include the Electron Devices Research Conference at Cornell in 1948, working with Jan Rajchman, Wallmark's advisory board activities, Wallmark's students, and the IEEE. The interview concludes with Wallmark's thoughts on strategies for encouraging innovation.

About the Interview

J. T. WALLMARK: An Interview Conducted by Rik Nebeker, Center for the History of Electrical Engineering, 17 July 1996

Interview #303 for the Center for the History of Electrical Engineering, The Institute of Electrical and Electronics Engineers, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, IEEE History Center at Stevens Institute of Technology, Castle Point on Hudson, Hoboken, NJ 07030 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

J. T. WALLMARK, an oral history conducted in 1996 by Frederik Nebeker, IEEE History Center, Hoboken, NJ, USA.

Interview

INTERVIEW: J. THORKILD WALLMARK

INTERVIEWER: RIK NEBEKER

DATE: JULY 17, 1996

PLACE: GOTHENBURG, SWEDEN

Childhood, family, and education

Nebeker:

You were born the fourth of June, 1919, in Stockholm. Could you tell me something about your family background? What did your father do?

Wallmark:

My father was in the Postal Service. I have a brother and we were both trained as engineers. As part of the Engineering program, I wanted to have a stay in a foreign country at the very end, getting into technical work abroad.

Nebeker:

Working outside of Sweden was an attraction for you?

Wallmark:

Yes. It is a necessary part of education I think, in a small country like this.

Nebeker:

Were you always interested in science and technical things as a youngster?

Wallmark:

Yes, I think so. Although I actually started out in medicine, but I quickly changed over, since medicine involved learning a lot of Latin which I didn't quite see the point of.

Nebeker:

You went to gymnasium in Stockholm?

Wallmark:

Yes.

Royal Institute of Technology

Nebeker:

And then to the Royal Institute of Technology?

Wallmark:

Right.

Nebeker:

What was your line of study there?

Wallmark:

This was Electrical Engineering. I actually didn't like the studies very much. It was a little bit too much of the "What you should know" rather than what I wanted to know. It was not really until my fourth year that I started to like doing it, because I became part of a small research team in one of the departments there. It was dealing with new kinds of uses for electrons such as vacuum tubes and gas tubes, and things like that. This was for Hannes Alfvén, who was a very remarkable person and was later a Nobel Prize winner.

Nebeker:

Was this in weak current electrical engineering?

Wallmark:

Yes. Electrical Engineering encompassed weak and strong currents, communication, and information engineering, but also power engineering

Nebeker:

Alfven was a physicist.

Wallmark:

He was a physicist actually, yes. He was interested in things out in space, very far away from Earth. He was also interested in practical applications of what he knew. He became an inventor and with his cousin invented a very remarkable new vacuum tube. It was intended to replace the so-called Ericsson 500 Carousel Selector, which was the mainstay of the Ericsson industry at that time. It was a selector for telephone stations. He wanted to do this with an electron beam in crossed electric and magnetic fields. The idea originated from his studies on the solar atmosphere. I was early coming in on this business.

Nebeker:

Could you explain how this was related to studying the solar atmosphere?

Wallmark:

At that time very little was known about conditions in space, how electrons move around the sun in other parts of the cosmos and then come into the Earth's atmosphere, causing for example the Northern Lights and things like that. Alfven's contribution was a better understanding of the solar surrounding and the way the particles moved in the electric and magnetic fields present. He showed that they formed beams of a new kind; not straight beams, but beams that were coiled like trochoids. These beams could then be steered in a very intricate fashion. I can still remember seeing this being demonstrated in a large vacuum vessel with enough gas present so you could see the beams. The gas particles hit by the electrons gave off a faint blue light, and you could see these beams moving around.

Nebeker:

What was the practical application of this type of collision?

Wallmark:

Alfven considered making an electronic counterpart to the mechanical telephone selector. The mechanical selector had the disadvantage of being large, clumsy and subject to wear. The electron beam could be moved very much faster and more easily. We worked for many years on this invention.

Nebeker:

Was there any success with that new tube?

Wallmark:

Yes, to a certain point, we never succeeded in making a complete selector that was practical, because first the transistor came around and offered other solutions that were probably better. The other part was that in order to use this selector, the entire system had to change, and telephone technologists abhorred changing the telephone system. Later on, they had to do that anyhow under the impact of the transistor but at the time, there was very much resistance to doing that. So the invention succeeded only partially. It was used and a license was actually sold to Raytheon to manufacture decimal counter tubes for computers. Counting by ten was a very interesting proposition, fitting the existing ways of making computers. But ultimately the transistor, counting by two, reigned.

Nebeker:

This tube was called the Trochotron?

Wallmark:

Yes, from the trochoidal path of the electrons.

Nebeker:

I see. And it was actually manufactured in some numbers?

Wallmark:

It was manufactured by Raytheon and sold for many years.

Nebeker:

What was your involvement in that?

Wallmark:

I was a graduate student, fresh out of the university. I had spent a year with Alfven on doing other things. I was also in the industry for a year learning how to build tubes. And then I came back to Alfven, and was his experimentalist so to speak, building all the experimental tubes. From 1945 to 1947, it was a very interesting time.

Nebeker:

Now was Alfven interested in tubes more generally, or was this one particular idea he'd gotten?

Wallmark:

Well he had practical ideas like that, and many other ideas also. This was one of the more spectacular ones, but he was usually interested in applications. At the same time, his main interest was up in space.

Nebeker:

You found that very stimulating?

Wallmark:

Oh yes.

Nebeker:

So you started that in your fourth year at the Institute you said?

Wallmark:

Yes.

Nebeker:

You got your Engineering degree when?

Wallmark:

1944

Nebeker:

What did you do then?

Wallmark:

Well, I spent about a year with Alfven after that.

Standard Radio employment

Wallmark:

Then I spent one year in industry at Standard Radio, learning how to make tubes. This was just after the War. We made the first Swedish electron tubes at Standard in 1945. Standard was an ITT branch in Sweden.

Nebeker:

I talked with the son of Arne Schleimann-Jensen who was involved in the beginnings of the tube industry.

Wallmark:

That was at Ericsson in a new daughter company called Svenska Electronica AB.

Nebeker:

Yes, it was affiliated with Ericsson. But that's not Standard Radio?

Wallmark:

No. That was a parallel one and came a little later.

Nebeker:

Okay, and what exactly was your work at Standard Radio?

Wallmark:

To make tubes.

Nebeker:

So you were sort of a production engineer for that tube?

Wallmark:

I was constructing and manufacturing a new tube that they could not buy from the outside.

Nebeker:

Sounds like a challenging project. Did you succeed?

Wallmark:

Yes.

Nebeker:

And then you say you went back to Alfven?

Wallmark:

Then one of the largest strikes in Swedish history hit the company. The company offered me half pay, and no work. So I looked around for something to do, and I contacted Alfven to see if I could do something with him.

Graduate studies, experimentation for Alfven

Wallmark:

So I went back to Alfven, and he happened to have this project where he was very much in need of someone to do the experiments. I could do those experiments.

Nebeker:

Can you tell me more about those experiments?

Wallmark:

Alfven wanted to sell his invention to Ericsson. Ericsson said, "Oh yes, this is very interesting. It is a fantastic idea to make an electronic switch. But, how about this, and how about that and how about the other?" There were many questions that Alfven could not answer about reliability, and current, and speed, and noise, and all kinds of things. Could it be manufactured? Did it actually work? So Alfven had to set up experiments to try out all these things. This meant building lots of tubes and dimensioning every part of the tube and the necessary magnets that supplied the required magnetic fields. So I employed a glass blower I had at the time to blow small, thin tubes, and one technician to evacuate and prepare the tubes. Then I made measurements and constructed the next tube. We made about a tube a week for a couple of years.

Nebeker:

Were you making continual changes in the tubes?

Wallmark:

Oh yes.

Nebeker:

I see here that you were Research Assistant there until '53. Is that how long you continued working with Alfven?

RCA laboratories, Princeton, NJ

Wallmark:

During that time I spent one 1947 to 1948 on a stipend at RCA laboratories in Princeton, New Jersey, working on tubes. At that time, RCA was working on beam tubes for high frequency applications. I took part in making one such high frequency tube, using a secondary emission multiplier.

Nebeker:

How did your connection with RCA come about?

Wallmark:

That was on a stipend from the Swedish-American Foundation.

Nebeker:

Did you select RCA because of the tube work they were doing?

Wallmark:

I selected RCA because they had new laboratories in Princeton, New Jersey which were set up at that time. The head of the laboratory, Dr. Elmer Engstrom, was of Swedish descent and he liked to have young people from the Scandinavian countries as trainees. He had had about a dozen when I came in, and a dozen more afterwards.

Nebeker:

You were involved you say in the development of a beam tube?

Wallmark:

Yes. In 1948, while I was there, the transistor was invented. I participated in the Electron Devices Research Conference during February of 1948 at Cornell University. Then rumor came out that Bell had something that was a new amplifier without a filament. It was very, very secret and nobody could find out what it really was. It raised a tremendous amount of excitement at RCA. The head of the Tube Department at RCA, when he came home, called together the patent people at RCA, and he said, "Let's look through all existing patents and find a clue to what this new invention might be." They selected about 150 possible patents, gradually whittling them down to 10. Then ten people were selected, each to try one of these to see whether it was the way to go. While this work was going on, the announcement of the transistor came. During that time, there was a lot of other strange activity taking place in the wake of all this excitement. I remember an outside inventor came to the laboratories and wanted to sell a "black box," saying it was a replacement for the tube but did not need a filament. For this, he wanted a million dollars. He was willing to demonstrate it, but nobody could take it apart or take X-rays or anything like that. He got a radio from RCA, and he replaced the tubes with his Black Boxes. We tested the regular one and this with the Black Box side by side. They worked about the same. Maybe the Black Box one was a little bit better. But RCA didn't buy it. The inventor was a radio technician and he didn't have much status. It was not bought and it was not bought by any other company that he applied to. But it caused a lot of excitement. Then came the transistor announcement in The New York Times on July 1, 1948.

Nebeker:

Do you know what the black box was?

Wallmark:

It contained a tube and a small battery supplying the filament. I don't know if he actually sold anything but he was traveling around to many industries in the United States with this invention.

Nebeker:

I see. Who were you working for at RCA?

Wallmark:

I was working for Ed Herold, who was head of a part of the Device Group developing new tubes.

Nebeker:

How was the atmosphere at the RCA labs?

Wallmark:

It was fantastic. Laboratories have a life span like people have. They are born and there's lots of excitement. Then there is a lot of growth and they mature. Finally they whither when they get old. When I was there, it happened to be at the high point of the labs.

Electron beam tube and Ph.D. thesis

Nebeker:

So you went back to Alfven's group after this year at RCA. And continued the same sort of development?

Wallmark:

Well, I then continued a little bit with a new electron beam tube which I invented at RCA. I got a large award from RCA for doing this. There was a lot of excitement about it.

Nebeker:

Can you tell me a little bit about that particular tube?

Wallmark:

This was a beam tube with a secondary emission multiplier at the end to increase the transconductance of the tube. It was not known when I was at RCA that you could steer the beam by the "space charge" of the beam itself. You could get a much more sensitive amplifier with this new way of steering the output of the tube.

Nebeker:

Steering in what sense? What was the idea, to use this as an amplifier?

Wallmark:

Yes, an amplifier tube. I made this my thesis, so that I had to continue working on it. It was called "Space-Charge Deflection: A New Principle for the Control of Current in Electron Tubes," and was published in 1953.

Transistor and semiconductor publications

Wallmark:

But at the same time, I was well aware of the power of the transistor. So I took up transistor work at the same time. I got part-time work with a small Swedish group, Elektrovarme-institut, trying to understand what the transistor was about.

Nebeker:

Did it ever reach the production stage?

Wallmark:

No. The transistor took over. It was too late.

Nebeker:

I see. So you came back thinking the future lay in semiconductors?

Wallmark:

Yes. I wrote an article in the largest Swedish newspapers at that time about the transistor. The article was called "Farewell to the Electron Tube".

Nebeker:

What year was that?

Wallmark:

Around 1949. This article caused a lot of stir in Sweden. We had the Swedish Radio Company call up my boss, Hans Alfven. He asked him what was going on and how he could allow anybody to write something so foolish and erroneous. Alfven came down to me and told me this and he was very amused.

Nebeker:

He agreed with you?

Wallmark:

Well, he didn't quite know. He was quite willing to allow new things to come out. He was interested in this. But he never got into this new field of research because he was more interested in space. He supported me. But the other thing that was very much an element of this was that it made me think I had to do more on transistors. Gradually I got into that field.

Nebeker:

So this period from '48 to '53, when you were still with Alfven's group.

Wallmark:

Yes. But I was part-time with this new transistor activity.

Nebeker:

I see. You were also completing your Licentiate? Or was that your Doctorate?

Wallmark:

No, I had passed the Licentiate in 1947. This was a Doctorate, which I completed in 1953. At the same time, I had some other part-time activities to liven things up.

Working with Hans Alfven

Nebeker:

I wanted to ask you, before moving further in your career, a little bit more about Hans Alfven as a person. How was he to work with?

Wallmark:

Great. He was a very remarkable man. He was out on new ground all the time. That's what Nobel Prizes are given for. And he was very good.

Nebeker:

Was he authoritarian in his attitude?

Wallmark:

No not at all. He was very liberal in his attitude toward other people. He had many people around him working in various ways which were both new and different.

Nebeker:

Did you interact with him very much?

Wallmark:

Yes, because he interacted in personal ways, but not necessarily formal ways. I was working on things that were very different from what his interests were. But he supported that and he was interested in how it was going. We talked often about things like that.

Nebeker:

So he would ask you how your work was going?

Wallmark:

Yes.

Nebeker:

Would he ever have suggestions?

Wallmark:

He would have suggestions, Yes. But more along his own lines. But the interaction was much more intense regarding the Trochotron work. That was daily.

Nebeker:

You were still working to develop that as a practical tube in these years?

Wallmark:

Yes.

Nebeker:

What were these other activities you had in these years?

Swedish Technical Research Fund

Wallmark:

I was also the Secretary part-time for the Swedish Technical Research Fund, which is government agency that provides money for research at universities. This was a group of about eight people who were the tops in Swedish research and industry. This was only part-time activity, but it was an interaction with very remarkable people. That was interesting.

Nebeker:

Were you involved in the decision making?

Wallmark:

No.

Nebeker:

But you were exposed to all these proposals?

Wallmark:

Right. I learned a great deal from meeting people, learning about projects and the evaluation process.

Nebeker:

Are there other activities there you think of?

Wallmark:

I think that's it. Those are the main ones.

RCA employment, transistors

Nebeker:

Okay. In 1953 you went back to RCA.

Wallmark:

Yes.

Nebeker:

How did that come about?

Wallmark:

I realized that I had to, if I wanted to work with transistors, RCA was a much better place. I had a very nice time at RCA. During the year I spent there, I had made six suggestions for new patents that were actually accepted. One of them was rewarded with an RCA Laboratories Achievement Award. I also obtained a three year consulting arrangement with RCA to follow up this work. So I fit in very well at RCA. I found it a very stimulating place to work where I could contribute.

Nebeker:

Was that a good place to learn about transistors?

Wallmark:

When I came back, I was put on tube work to start with, because I didn't have a background in transistors. But I very quickly switched over to transistors.

Nebeker:

What was your success in arousing interest in Sweden in transistors in those years up to 53?

Wallmark:

<flashmp3>303 - wallmark - clip 1.mp3</flashmp3>

The work in Sweden was of course to follow up developments in the United States. The first thing I had to do was to make a rectifier diode. When I came to this group to work on this, they had acquired some germanium, which every night they locked up in the safe in order not to lose it. Then we learned to make diodes from it. That work was for military purposes to start with. The diodes were very, very poor. These were point contact diodes, and they couldn't take more than a volt or two in the reverse direction. We knew that they should take a hundred volts at least. The electrical characteristics were very soft and very hopeless to work with. The group couldn't cope with that--they had an order to make five hundred. I started to work with it for a while, and then I discovered that if I overloaded them in the reverse direction, they got better and better. We learned the hard way the technique of forming the point contact by welding the wire to the germanium, making a much firmer connection, and at the same time forming a junction at the point. In this way, we were successful in meeting the orders for making diodes. This was our entrance in the semiconductor business.

Nebeker:

So that group actually produced diodes?

Wallmark:

Yes. After that we worked on silicon and tried to make high power rectifiers. Again, there were a lot of difficulties catching up with the Americans.

Nebeker:

I'm wondering about the technology transfer in that early stage. I suppose you were reading whatever patents or publications you could get hold of in those years.

Wallmark:

Actually there was no useful information around in those days. The only alternative was to go ahead and try out the best ideas you could think of.

Nebeker:

Each group had to work their way themselves. So you said first at RCA you did some tube work?

Wallmark:

Yes early in the year. When I came back I was put on Magnetrons, but I could not make any real contribution there. I had a predecessor on that, and he had a predecessor, but they were not very successful in doing what RCA wanted, which was to modulate the frequency of the Magnetron while the output power was varied.

Patents

Nebeker:

I wanted to ask you about some of these early patents of yours about electron tubes. Which were particularly significant? You have a list.

Wallmark:

Yes. Those were tube inventions and they were made at RCA.

Nebeker:

All five of those first five?

Wallmark:

Yes.

Nebeker:

I see the third one was patented in a number of countries. Can you tell me a little about that?

Wallmark:

That's the one that I made into my thesis. It was judged to be very interesting, and RCA applied for patents in many countries.

Nebeker:

I see. Were any of these others particularly important?

Wallmark:

They were interesting. The Power Supply Circuit was a way of stabilizing the supply voltage to the secondary emission electrode in the tube that I was working on. It was my first patent actually, and the first time that I realized what a patent was. Then I made five more suggestions like that during the rest of the time I was at RCA. I don't think any of them became very important because the transistor overwhelmed the tube work.

Nebeker:

I see. What was Patent 2,875,384, this semiconductor devices patent in '56? The sixth one on your list.

Wallmark:

<flashmp3>303 - wallmark - clip 2.mp3</flashmp3>

This had to do with transistors. Transistors had difficulties with their surface properties. We knew that silicon had a very well characterized oxide on top of it that germanium did not really have. Except there was a germanium monoxide that was interesting. It has a dark color and had many characteristics of the silicon dioxide. It was stable, thin layers could be made, and I treated the transistors electrolytically to provide a layer of monoxide, thinking that this might be similar to silicon dioxide in the stability and the way it protected the surface. It made the electrical characteristics more stable, and as it turned out, the transistors that I tried with this technique were very good. The transconductance became stable and very high. The life tests looked very good. The difficulty was that it was made on finished transistors, so that the conditions were not well controlled. I was never able to make this very reproducible and very repeatable and gradually I had to give up on it, because I couldn't control the conditions well enough.

Nebeker:

It wasn't clear that silicon was going to be the dominant substrate?

Wallmark:

No, this was during the time when germanium seemed promising.

Nebeker:

Do you know if the germanium monoxide has since been used?

Wallmark:

I'm not aware of any application like it. The patent became important when silicon came out because the patent was written in such a way that it covered a semiconducting device with a single element and with an oxide made from the single element itself. This applied also to silicon. So it happened to become a very important for the silicon case.

Nebeker:

RCA owned this patent?

Wallmark:

Yes. But then the MOS transistor came about, Steve Hofstein came around and said he had gone to the Patent Department trying to patent this new transistor and found that I had the patents on one of the basic parts--the native oxide layer on the silicon. The germanium patent covered this and so it happened to become important.

Nebeker:

Was that your idea at the time, that this might apply to other elements?

Wallmark:

No. I was working on the germanium transistors at that time. So I think it was really the skillful patent attorney who made it very generally applicable.

RCA research groups, transfer of technology

Nebeker:

Who were you working for? Can you tell me about your work at RCA? You were first in tubes, and then you got into semiconductor work?

Wallmark:

Yes. This was a group with many people in it. Charles Mueller was of course a part of this work and Ed Herold was still the head of the activities. There were many others. Steve Hofstein became very important. Fred Heiman of the same group also contributed.

Nebeker:

Did you ask to be transferred to begin working on semiconductors?

Wallmark:

Yes.

Nebeker:

Did you stay in the same group there?

Wallmark:

Yes.

Nebeker:

How much transfer of techniques or knowledge was there from tube work to transistor work?

Wallmark:

The basic principles of radio transmission and reception still apply. The active elements still amplify, rectify, modulate, and so on.

Nebeker:

What I mean is, all the knowledge and expertise that people had gained working with tubes, one might think that you do not have to start over from scratch doing this...?

Wallmark:

You had to, to a large extent. The materials and fabrications methods were very different and to a large extend unknown.

Nebeker:

So it really was starting a new area of work?

Wallmark:

Yes. Sure. The single crystal requirements, the surface and bulk cleanliness; The absence of glass and [inaudible].

Significance of transistors

Nebeker:

Were there some clever things that had been done with tubes that could be mimicked in transistors? Or was it pretty much just a new realm?

Wallmark:

I think it was completely new from a fabrication point of view, but the use of semiconductor devices was similar to that to tubes.

Nebeker:

You and others could see that this was the way of the future?

Wallmark:

Transistors were not very impressive to start with. So there was some room for certain doubts about their importance. They had very poor characteristics-- they were very unstable, had an unknown life, operated only at very low frequencies, and were very noisy. They required a number of improvements in order to make them practical.

Nebeker:

Weren't there a lot of people who thought transistors would be important in miniature devices and expensive military devices, but nothing else?

Wallmark:

Right, because that's where their advantages were particularly strong. But it was difficult to foresee that they could be used at thousands of volts or thousands of megacycles.

Nebeker:

From what you told me about that Swedish article you wrote, you believed from an early point...

Wallmark:

Yes, and I was pressed into action to defend my belief.

Semiconductors at RCA, lateral photocells

Nebeker:

So tell me more about that work then at RCA in semiconductors. What exactly did you do?

Wallmark:

Well, of course, I had a learning period also. One of my more interesting projects was the new focus that came out of work on the surface properties of transistors. Semiconductor devices at that time were thought to benefit from having an enriched layer on the surface. But there was not very good knowledge about how to do this. By having an enriched layer on the surface, you could push the charges towards the interior. They would not reach the surface, and in this way you could reduce the influenced surface properties.

[end of tape one, side one]

Wallmark:

This was Bill Webster's idea, and he later became the head of the RCA Laboratories. To apply the idea involved studying these layers. In doing that, I used an optical method to induce excess charges. And from the measurements that I made, I stumbled upon a way of making a new photocell with characteristics that not only registered the amount of light hitting the silicon, but also the direction of light hitting the surface of the semiconductor. I had all of a sudden a photocell that could register direction to a light source.

Nebeker:

Meaning that if you had a beam of a given intensity, the voltage or current it induced was very dependent on the angle of incidence?

Wallmark:

Right. I could, for example, measure the amount of light coming from the closest direction to a star. It could be used, perhaps, for military purposes to register the flash of light from a gun or something like that and get the exact direction so that counter measures could be taken against it. I have a patent on this. It was taken up with great interest by the Russians, who did a lot of work on it, and there were many publications on such photocells. They're called lateral photocells. As it turned out, this didn't become a very large thing for RCA. They sold licenses for it, but didn't see much sales. It didn't fit into television, which I had hoped. Later on after I came back to Sweden in 1964, I met a fellow who had seen this development. He had made a new system for measuring the movement of robots by fastening light sources on robot arms and measuring direction and movements of robot arms, among other things. This system was taken up by several new companies, and at the present time we have three or four companies working on similar developments. One company was created to fabricate new and improved versions of these photocells. So when I came back to Sweden, it became a breeding ground for new companies, which I found very interesting.

Nebeker:

Were you the discoverer of lateral photocells?

Wallmark:

Yes, I have the patent. It's still being manufactured in some number for calculating distance.

RCA project management

Nebeker:

Were you able to choose your own projects there at RCA?

Wallmark:

Choosing projects means that you work on something that the company thinks is interesting, and while you're doing that you may discover something new and different. In order to get support, you have to sell it to the company.

Nebeker:

In your case, you had to sell it to whom?

Wallmark:

To my supervisors.

Nebeker:

But you were given some freedom there to choose your particular topics?

Wallmark:

There was a great deal of freedom. But it had to fulfill some criteria. It has to be of interest to RCA. Ultimately, it had to be patented and RCA had to sell the licenses for it. RCA Laboratories was a place that lived on selling licenses. They created a sizable income for the laboratories and for the company. In a way, the production divisions of RCA were also customers for the patent licenses.

Nebeker:

I see. Were you ever assigned to a project? Some supervisor said, "We really need you to work on this."?

Wallmark:

This was done many times, of course. Things came up and people were gathered around this and some people could contribute and that decided who would work on it. It came about more or less by interaction.

Nebeker:

I see. It was very much the case that everyone was part of a team and somebody had to step forward and work on things as they came up?

Wallmark:

Right. Ultimately of course the supervisor had to decide if we were working on something that was useful.

Nebeker:

You mentioned television. Was that something that much of the work was directed towards?

Wallmark:

All the time. Sure. Television was always very much in focus. But I was never really very active in television projects myself, other than one or two small details in television systems. There were other groups who were more active in television than I was. Then, gradually of course, came the computer era. There was much work being done on computer related inventions.

Nebeker:

You had that earlier year at RCA and then you were at RCA from 1953 to 1964. Were you happy with the work environment there?

Wallmark:

Oh Yes.

Nebeker:

Were you planning all along to go back to Sweden?

Wallmark:

People have roots. I had roots in Sweden and also in the United States. I was married to an American girl, and she was part of my desire to go back to the United States. I had a desire to go back to Sweden, so we vacillated back and forth quite a bit.

Nebeker:

I see. So you had married her in that earlier period?

Wallmark:

Right.

Integrated electronics, computer circuits

Nebeker:

Were there more things that you think of right now on that period up to 1964?

Wallmark:

Well, yes. The integrated circuit era came about during that time. I was lucky to be in on some of the very early developments on the first integrated circuit devices at RCA.

Nebeker:

I was very surprised to see in this list of publications a 1958 publication about integrated electronics.

Wallmark:

I think it was probably the first use of the term.

Nebeker:

In 1958 or 1959, you had an article called "Integrated Electronic Devices." Is that what we now mean by integrated circuits?

Wallmark:

Yes.

Nebeker:

You were using it in that sense? That must be one of the earliest.

Wallmark:

Must be. This was based on our work on shift registers and also thinking about other computer circuits.

Nebeker:

Can you explain how that started?

Wallmark:

The time was right for introducing very small dimensions in transistors more extensively than earlier by putting them together very closely and getting rid of as many of the leads as possible. The ultimate expression of this was to put them on the same substrate. We made some shift registers that worked like that.

Nebeker:

That had a number of transistors on a single crystal?

Wallmark:

Yes, ten or more on a single crystal. Well, we didn't have the technology that ultimately became the important part--the silicon planar technology. We were hamstrung a little bit by RCA being involved in the Micro-Module work. It was what we today would call a hybrid circuit, with units put on the same substrate and connected more conventionally on the substrate. It had small ceramic wafers with two or three devices mounted on them. Then the wafers were assembled automatically to form cubes. The Micro-Module cube system was something that RCA got a large military order for. Therefore, work on new techniques that didn't fit into the Micro-Module thinking were not favored very much. So we got hamstrung with support for this, and progress was very slow.

[end of tape one, side two]

Nebeker:

You were telling me about this RCA system of Micro-modules.

Wallmark:

It meant that we couldn't quite keep pace with development elsewhere on the integrated circuit. Even though we had the concept early, we couldn't really follow it up. The fabrication system that we used for integrated devices was largely based on work by Herbert Nelson for tunnel diodes, using a mechanical approach for forming the elements. This was a very good method.

Nebeker:

Who else was working on integrated electronics at RCA together with you?

Wallmark:

Not very many in the beginning. I started it, and for a long time I was the only one working on this particular aspect. I had then a young fellow by the name of Stanford Marcus. Herbert Nelson was also working on it. Harwick Johnson was the group leader.

Nebeker:

Have any of these patents proved to be very important?

Wallmark:

I don't really know. That's one of the difficulties actually. Companies aren't very eager to spread information about which patents are commercially valuable and which aren't.

Nebeker:

I see that number thirteen on your list is for a "non-linear tunnel resistor." What was that?

Wallmark:

That was a way of biasing tunnel diodes.

Devices research at RCA, MNOS

Nebeker:

What people most impressed you at RCA in those years?

Wallmark:

Well, there were remarkable people all around. It was really a beehive for people interested in new developments in devices. RCA was very strong in devices. They had many strong people on television, and then computers when those came along. Ed Herold was certainly the leader of the device activities. Jan Rajchman was certainly a pioneer in memory and other computer devices.

Nebeker:

Did you have interactions with him?

Wallmark:

<flashmp3>303 - wallmark - clip 3.mp3</flashmp3>

Yes, very often. I became a member of his group. It was later, particularly after I came back in August 1966. I was working on computer memories. During that time I invented an important device called the MNOS transistor, which was a memory transistor. It was based on the theory that with a silicon nitride layer on a transistor, you might get a better transistor, because a silicon nitride layer has a high dielectric constant. You get a more sensitive transistor using it. At the same time, silicon nitride is a very stable insulator. The difficulty had to do with electrical instability.

Many laboratories other than RCA tried using nitride as a gate insulator. Some of these transistors were made by Joe Scott at RCA. He made a patent application for using the instability for memory purposes, but the Patent Department said they had "no interest." They put it in the inactive file because of the very serious electrical instability and the very low functioning. The characteristics were gliding all the time. You could a write a single bit of information to them, and it would stay for awhile but then it would glide away. You could never get them stable. Other companies found the same thing. But I thought the idea was interesting, so I succeeded in getting some transistors from Joe Scott and I studied them. I believed that I could improve these very, very poor devices and make them faster and more stable. Then I realized that under the nitride there must be a layer of oxide, however thin. I thought the first thing I should do was to control the oxide. At that time, there was a way of measuring the thickness of the oxide using an ellipsometer that we had at RCA. I successfully made a very thin, stable, reproducible oxide layer. I put the nitride on, and then I measured the devices. They were very stable. So then from this came a way of making stable memory transistors, and they are still in use.

Nebeker:

I see the nineteenth patent on your list here by Scott and Wallmark. That's an RCA patent?

Wallmark:

Yes. It was never accepted in the United States because there was an earlier patent on that invention by Raytheon. However, Raytheon did not believe that it was a very important idea, so they did not patent it elsewhere, only in United States. RCA thought it was a good idea, and applied for patents in all countries.

Nebeker:

That was then something that RCA earned money on in many other countries?

Wallmark:

Yes.

Nebeker:

Well that's very interesting.

Wallmark:

It was accepted in nineteen countries.

Nebeker:

Was it only in '66 or '67 that you were working on memory devices?

Wallmark:

To start with, we were working on integrated magnetic memories. It was Jan Rajchman's idea to integrate not only transistors but also the magnetics in the same circuit. But in melding these two technologies together, you had to sacrifice either on the semiconductor side or on the magnetic side because the fabrication processes were not compatible. So that combination was not practical. It was gradually abandoned, but during that time I hit on the MNOS idea, and that was what I was working on at this time.

Nebeker:

I see. Were you in Rajchman's group there that whole period?

Wallmark:

Yes.

Nebeker:

How was he to work for?

Wallmark:

Oh, great.

Nebeker:

Can you tell me about him?

Wallmark:

Yes, he is a "famous name" particularly in his field, which was magnetics. He found that he had to use semiconductors in combination with the magnetics. The idea of integrating magnetics and semiconductors was good, except that when you try to do it, you find that you cannot meld the two technologies together. The temperature ranges of the materials are different. You had to fabricate one separate from the other-- you could not do them both at the same time. It required one process to make the semiconductor devices for one side and another process for the magnetics on the other. And then you had to connect them. It meant connection wires to join the different semiconductor and magnetic arrays.

In the period I was there we just decided that that was not the way to go. We decided that what we should do was to cash in as fast as possible and then go to the next stage. Semiconductors were taking over a large part of the memory market. Then what was left were the very compact disc memories.

Chalmers University

Nebeker:

What about this period in between '64 and '66, when you were here at Chalmers University?

Wallmark:

After I came back to Sweden with all this background, I was very intent on working on semiconductor technology. At that time semiconductors had been accepted, even in Sweden. To get started, I approached Swedish industries to see if I could get some ties into what had happened in Sweden, and to find out what the interest was in Sweden. I found that Sweden was really different from the States. I was struck by the secrecy barriers that were present between industry and the universities. I had a very hard time. I met the research head for Ericsson, and he said they had a very interesting semiconductor work going on based on an American idea that he couldn't describe because it was very secret. It had to do with a negative characteristic in the device that could be used for memories. It was obvious what that was from what he told me. I knew the background, and I had a lot of knowledge that could have been of use to him, but he couldn't discuss it and he later wasted a lot of money on this development. I thought that it was very unfruitful to continue, and if I was to do work on semiconductors, I had to work on the device, the industry would not be interested. I had to produce the devices myself. Still, industry was not interested. I thought that I would have to design the "boxes" -- the applications for the devices, and sell them on the market. If it was a successful market, then maybe industry would become interested. This was the thinking I followed at the university. I then started spin-off companies, to commercialize my inventions. I very quickly built up a group of students, and they were very productive inventors. Quickly my laboratory became the most productive in new patents and inventions. At the same time I was creating spin-off companies.

Nebeker:

That was your idea from the beginning?

Wallmark:

I had to do that because I could not get any response from simply working on very sophisticated devices.

Nebeker:

I'm sure it's the feeling of inventors in all cultures, that they have wonderful ideas that industry won't pay attention to.

Wallmark:

Right.

Semiconductor research in the U.S. and Sweden, 1960s

Nebeker:

Was there a difference between Sweden and the United States in the '60s in the semiconductor field?

Wallmark:

What you're saying is true. Inventors have always had this difficulty. You cannot expect it to be different. If inventions are not ready for market, you have to answer questions. If you cannot answer, it's not going out on the market. You have to go on the market yourself. So the inventor also has to be an entrepreneur. This has become very well accepted at universities.

Nebeker:

Was there more secrecy in Sweden than the United States about this new technology?

Wallmark:

Yes. In Europe, secrets are held more tightly than in United States. In United States, people are moving. With people, ideas move. One has to accept this, and therefore secrecy barriers are more relaxed. You try to keep very few secrets, only the most sensitive ones, but the rest you have to let go. In Europe, the barriers are still very much active. The same old sick European tradition has persisted, and it has prevented interaction between industry and universities to this day.

Nebeker:

So you're saying that if you were at an American university, you wouldn't perhaps feel as great a need to carry the ideas so far toward the marketplace as you do here?

Wallmark:

It has to be done in the States also, but you have a more responsive infrastructure helping you to do this. Here you are on your own. Although I'm generalizing.

University work environments; teaching

Nebeker:

So you came back and got a professorship here?

Wallmark:

Right.

Nebeker:

Why then did you return to RCA for a couple of years?

Wallmark:

It was because of family matters. I had two boys and they were still in school, so I wanted to see them through until they were almost grown.

Nebeker:

Did you obtain some leave of absence from the university?

Wallmark:

Yes. And I had leave of absence from RCA.

Nebeker:

Now why did you think that a university setting was the right one for you? It seems like you're pretty much the entrepreneur type, maybe you should have been setting up your own companies?

Wallmark:

I probably wasn't a good enough entrepreneur; I was probably better at the university. Instead of starting one company myself, I now started a hundred companies through students and collaborators. I had of course the background of being at the university from my work with Alfven. I liked it. It's matter of personal preference.

Nebeker:

Do you enjoy teaching?

Wallmark:

Yes. Particularly I like having a group of youngsters around with all kinds of ideas. Following something that you can see growing and becoming successful.

Companies and patents to commercialize Chalmers research

Wallmark:

At the present time we have a couple of hundred companies spun off from the university and about four hundred patents that have come out of the university.

Nebeker:

Yes, I'm very interested in this topic. That was your idea when you came here?

Wallmark:

I had the background, but I was forced into it because I couldn't work in the same area as at RCA. There was no response. There was no interest. I couldn't talk to people that I really should have talked to.

Nebeker:

So you'd have an idea that you thought had great market potential. You had to carry it further here?

Wallmark:

Yes. All the way. It had to go out on the market.

Nebeker:

You would do that not in the university, but by spinning off some company to actually market the product?

Wallmark:

That's right. At the university, you're a teacher and you have young students. You shouldn't do it yourself. You have to have young people developing and commercializing

Nebeker:

I know of a few examples at American universities where at the university itself they went into production. They started producing some microwave attenuators at Polytechnic University for example. But that's pretty rare. Do you feel that there's some kind of a line, that university work is the research to a certain point, and then the production must be done outside the university?

Wallmark:

That's the tradition that was prevalent in Sweden in those days. You work on the development and then industry should take over. But the experience from the Trochotron work indicated that industry was not at all ready to take over until you could answer many questions, which you couldn't yet answer.

Nebeker:

What I'm trying to get at is, would the university itself undertake the production?

Wallmark:

No. The university is a government-run operation here. You could not do that at all. It had to be done by the inventors themselves.

Nebeker:

So they would set up a company to produce the product?

Wallmark:

Right. It had to be entirely different. But there is a transition period after the research which is government-run and paid for by private enterprise and private money through taxes. If you make a device, all the way up until you have the device in your hand, it is still in the research phase. Then the thing that always happens is that you have the device and you want to sell it, and all of a sudden it's private.

Nebeker:

I see. What about patent rights? Are they held by the university?

Wallmark:

In Sweden, the patents made at universities belong to the university researchers. This is done in order to stimulate inventions.

Nebeker:

So the inventors are the sole owners of the patents?

Wallmark:

Inventors are the sole owners. This has some advantages, and some disadvantages. We discuss this issue all the time. Very recently we had a major discussion of this again. The outcome was that it's been recommended that we retain this policy. The inventor gets to own it, but also has to pay for it.

Nebeker:

I see.

Wallmark:

He has to pay the cost of the patent. That can cost a tremendous amount.

Nebeker:

Yes. I see.

Wallmark:

What the university can do is to back it up. By teaching, by advising, by research support. But when it comes to business... The lateral photocell was picked up by a graduate student here. Another graduate student started a company that made a technology to transfer ECG information from the scene of an accident to the hospital. Then advice can be sent to the ditch where the fellow was found. This came too early, however.

Nebeker:

What period was this roughly?

Wallmark:

This was 1960. It was too early. It wasn't ready. It didn't fit. The company nearly failed. But the fellow running the company picked up the idea of the lateral photocell from a fellow student in the research department. He showed that this could be used to measure many things. He picked up this idea, started a new company, a photocell system "Selcom." This still exists. He has left it, he has been without. This company employs about thirty people, selling equipment all over the world. Another company was started, making photocells, and also selling all over the world. This company has been bought by the Japanese. This year they will make three million photocells. Three million photocells! It is fantastic. They are used in cameras for measuring distance.

Nebeker:

For automatic focusing?

Wallmark:

Yes. The original starter of the company made another one, competing with its former company. It just got in on the stock market. In the mean time, the Gothenburg area has become a seed bed for all kind of companies, working on things like measuring systems. It's very successful. There are three or four or five companies floating around based on ideas like that.

Nebeker:

At a university there are some people who always intend to work in industry. There some people who see themselves as entrepreneurs. But there are also a lot of people who see themselves as academics, who want to do research, but are not at all interested involving the applications.

Wallmark:

Most of them are like that.

Nebeker:

Most of the students here are like that as well?

Wallmark:

Most of the university staff are like that, also.

Nebeker:

But, you've always encouraged your students to carry the ideas further?

Wallmark:

Right.

Innovation Center; collaborations of university and industries

Nebeker:

Now I see that you're presently in this innovation center. When did that start?

Wallmark:

This started gradually after about 1964. I wrote papers on spin-offs from universities about that time. Then, only gradually, it caught on. It's a shift over thirty or forty years.

Nebeker:

I see. But at some point, there's a unit at the university called the Innovation Center

Wallmark:

Yes. The Innovation Center formally started in 1979. It tried to pull the interests in the university into some kind of a structure. The Innovation Center was tied to my Chair. I started an Innovation Building in 1983. It has gradually grown, and now a much more ambitious, incubator activity has been started in the last couple of years. The Incubator is a place for commercial enterprises connected to university research on campus. It is tied in all kinds of ways to university research. But it also allows us to separate private enterprise from the university.

Nebeker:

So a private enterprise could be situated there at the Innovation Center?

Wallmark:

Yes.

Nebeker:

I see. Do they rent space? Are they given space?

Wallmark:

Yes, they rent space. We have about two hundred companies that we have kept contact with, all spun off from the university involved in all kinds of activities originating in university research.

Nebeker:

I wonder if you could just explain to me what's in it for this new company. The Incubation Center, what is it that they gain?

Wallmark:

The Incubation Center is a very small part of the Innovation Center. The Incubator is only for about eight companies. As they grow, there's no room for them, they move out. More will come in. The two hundred, or nearly all of them, are in the Gothenburg region, within reach. They retain ties with the university in various ways. They are dependent upon the interaction with the university, such as financing functions, seminars, marketing activities, and particularly research support.

Nebeker:

Presumably, at least in many cases, the initial idea for this product comes from the university.

Wallmark:

In some ways. Very often directly from research. Somebody invents a new instrument for measurements and then we say, "Yes, let's see if we can sell it."

Nebeker:

How exactly are they able to get research support?

Wallmark:

Indirectly as they come from the group doing research on this type of measurement. So naturally through personal relationships, information goes back and forth. The company gets some market demands, some market feedback, and from the measurement activities at the university research, the company gets further knowledge and ideas to improve the instrument.

Nebeker:

Are the university staff ever hired as consultants to help solve particular problems?

Wallmark:

Sometimes, but mostly the interaction is less forma.

Nebeker:

Just being part of the community.

Wallmark:

Yes. The administration of the university has, of course, rules about what this interaction should be. It has to benefit university research, otherwise the university is not interested. And the same for the company.

Nebeker:

Now, it's very common I know in the United States for people already employed in industry to get an advanced degree. That means some work on work-related projects. Does that sometimes function here? Are people who are working on a degree at the same time setting up a company?

Wallmark:

Yes. And, of course then, many times, if the idea is successful, the desire for the degree might wane. I often get into discussions about, "Well should I really continue studies or should I quit and devote my time to the company?" There are some extremely well-known people like Bill Gates with Microsoft, for example, who left his studies very early. And then you have some that continue all the way and take a degree. It depends upon the idea. If the idea is good, it might be worth doing that instead. So we have people quitting all the time, and some of these students are really very good.

Nebeker:

Is it ever the case that these companies suggest research projects for graduate students who are looking for research projects?

Wallmark:

Right. An engineering degree here requires that you do three or four months in what is called thesis work, which is an engineering project. Nowadays, it's also done, paid by a company, maybe a spin-off company. So then the company says, "I would like this to be done. Can you do this for three or four months?" At the same time, the students use this as thesis work. This is a very good type of interaction.

Nebeker:

Okay. In addition to these two hundred spin-off companies, are there other high-tech companies in the Gothenburg area that have close ties to Chalmers?

Wallmark:

Yes. Sometimes the work in other spin-off companies is of a research nature, and they could benefit from interaction with us. They are welcome in our educational group, in which we do courses for company leaders or potential company leaders. We do courses for inventors. Or else, we have open lectures, and we invite people rather broadly to come and see if they get something out of it.

Nebeker:

And do you get many attendees from industry?

Wallmark:

Yes. We usually invite people from the outside to the seminar, to talk about new legislation with exports, about new tax rules or new support, and various other things.

Nebeker:

These examples sound like you're almost a business school.

Wallmark:

Yes.

Nebeker:

But you see that as part of the mission of the Innovation Center here?

Wallmark:

Right. My successor comes from industrial economy department, in the business part of the university .

Nebeker:

Well this is a very nice example of university inspired economic activity.

Wallmark:

It has been quite successful. The attitude of the university researchers has eventually shifted, and now they devote more thinking to the applications of whatever is going on, rather than waiting for industry to find them. Even in mathematics that kind of stuff happens. We have a fellow in the Data Logic Department who has made an algorithm for steering the work of airplane companies. You can over time adjust people's working time to their desires and also the company's desires. And this algorithm is sold all over the world now. This is a small company. But he doesn't want to run companies, so his father runs the company activities and he's continuing his research on new aspects of this problem. He's doing the research part of it.

Nebeker:

Do you see any changes in Sweden or elsewhere in Europe of better ties between industry and university, or more openness?

Wallmark:

Oh, yes. It's gradually improving. You can see changes over time.

Federal Board of Technical Development

Nebeker:

We've gotten away from your career a little bit, and I want to be sure to get through it, at least the main points. Let's see, you also served a good deal on this Federal Board of Technical Development?

Wallmark:

The Federal Board of Technical Development is where I was secretary very early.

Nebeker:

Since then, you have been a member of that board?

Wallmark:

No. I've been named on the advisory group with that board several times in various aspects.

Advisory group activities; invention funding

Nebeker:

I see. Maybe I should just ask you what other principle activities you've had the last couple of decades?

Wallmark:

The most interesting ones, I think, are the ones dealing with inventions, spin-off activities, and things like that. I have been a member of several advisory groups in industry, having to do with diversification or new inventions. I've been with Ericsson, Peritorp, Frico, SKF and others Then I've been chairman of a group giving financial support to inventors, rather large sums yearly. We have had about a thousand applicants each year to get these ten awards we give out each year.

Nebeker:

This is to allow people to further develop an idea?

Wallmark:

Yes. We try to get ideas very, very early. We provide a lot of helping in the early phase, then other support can be found from other groups. At the present time, I am chairman of a group on environmental research.

Nebeker:

I see. So before people could go to the venture capitalists or to a company, because the idea isn't far enough along, you're providing money?

Wallmark:

Usually most agencies that lend money for developing a product like to provide support late in the development process. They like to come in when the invention is ready and all questions are answered. I've been concerned with the very, very early phases where you don't know what's going to happen. Many things still have to be done. In order to find the answers to the problems and be able to carry the invention further and get support from financing groups, the university has to help.

Nebeker:

Do you think this inventor support has been successful? Has it been going on long enough to be able to say really?

Wallmark:

I think that the number of mature successes has not been very large yet . We have been so early that the risk for failure has been quite high and progress is sometimes slow. But it's important to try out ideas early, even if you get a negative answer, because you will get an answer before costs have risen too high

Nebeker:

Have you yourself started any companies?

Wallmark:

Yes, but only as a smaller "wheel" in the business. I am not really into running companies now. I like to see my students running the companies. That way I can work with many of them simultaneously. Running a company is an all-day business.

IRE and IEEE

Nebeker:

One thing I did want to ask you about was your involvement with IEEE and why you bothered to be a member. What has been valuable to you?

Wallmark:

I started out in my student days around 1945. I had the choice to either to join the IRE or one of the European counterparts. But the European counterparts were very closed. The work that they did was very much concentrated on standards. The IRE was much broader and more open. The attitude was different, and it was much more to my liking. You can see now the European counterparts have not developed much. But the IEEE has grown from maybe ten thousand members to over three hundred thousand. It has had fantastic development. It is very well-rounded, very well-done.

Nebeker:

Did you value the IRE, and then later IEEE, publications?

Wallmark:

That has been the main selling point. Europeans have nothing like that.

Nebeker:

What about conferences? There are some in Europe that IEEE organizes.

Wallmark:

IEEE organizes mostly in the United States, which is a gathering point for all Europeans also.

Nebeker:

What about the IEEE standards activities? Has that played any role in the areas you worked?

Wallmark:

I have not been very interested in standards. They are very active? It's not my prime interest. I think the IEEE history arrangement is a very good one. It is well conceived and well run.

Nebeker:

Now I did want to ask a couple of more general questions. But before doing that, I probably have neglected to ask about some aspects of your career. Well, one thing, you have retired now, is that right?

Wallmark:

Yes. Not quite, but almost.

Nebeker:

Was that fairly recent?

Wallmark:

Yes. I formally retired in 1989. I'm still spending time following up on some pet projects.

Nebeker:

I see. You still have an office here. Do you still have any students?

Wallmark:

Petering out.

Nebeker:

Are there any other things that you care to mention? Maybe people who have meant a great deal to you in your career?

[end of tape two, side two]

Publications

Nebeker:

Are there any books that have meant a lot to you?

Wallmark:

I've been active in writing books, of course. This is part of a professor's work. I wrote about a dozen or so, most of them in Swedish. But there are a couple of them that have been important internationally also. I was in on the first book on Microelectronics, edited by Ed Koonjian in 1963. I wrote the introductory part on micro-electronics theory. I am particularly proud of the very early predication of the so-called soft errors in space electronics. I was also an editor of Field-Effect Transistors, the first book on the subject, in 1966. This was an RCA venture, and the authors were taken from the RCA groups, which at that time had the leading position on FETs Then I shifted to inventions. I have another book on field effect transistors written together with one of my students here. I was writing books in Swedish on everything-- Thyristors, photocells, tunnel diodes, integrated circuits, and so on.

Nebeker:

Have these been used at Swedish universities?

Wallmark:

Yes.

Cultivating innovation; roles of universities, government, R&D

Nebeker:

One question I had is, there was recently established at the Smithsonian a Lemelson Center for the Study of Invention and Innovation, and we're working with them on a project. The idea, the reason Lemelson gave money to the Smithsonian is to better understand how innovation occurs, technological innovation. You've already given me a lot of your ideas and experience in this area. So, I mean this is just completely general, but are there any thoughts you care to express now about how one can cultivate innovation and create an environment that will stimulate invention and the carrying of that through to innovation?

Wallmark:

I have some thoughts, particularly on the university as a place for this. What you do at the university is important, because inventions at universities are on the average of a higher quality and really different from low technology inventions made elsewhere. There's a great deal of things one can do at a university in order to stimulate this and to improve the flow of inventions out of the university and the quality of this flow.

Nebeker:

There's a difference between fifty or a hundred years ago and today, in almost all areas of technological innovation. It's very difficult for the lone inventor to make an impact. Much more knowledge is required. Much more teamwork in the development stage is required. And maybe this is where the university has a particularly important role?

Wallmark:

It has always been difficult for the inventor, there's no doubt about that. And the scope of teamwork has certainly increased. There's more to technologies than there used to be and obviously a larger interest by society. Are you thinking about what kind of things society should do for this?

Nebeker:

Yes. One could argue about what's best for innovation in a society. These are very general questions, but you might say that what's best is to have some very large companies like AT&T and General Electric that can support this very expensive R&D activity. Other people might say, "That's wrong. These companies stagnate. What we need are hundreds of small start-ups."

Wallmark:

I think there are some inventions, some innovations, which cannot be done, except in very large companies. When you think about the telephone system, it is impossible for a lone inventor to come in and do something except a little piece of the work. The system approach has to be done by companies. And many of the systems these days are sponsored by large companies. And then, as you say, those are very complex inventions.

Nebeker:

Yes.

Wallmark:

So, there again, of course universities can still take part in the theoretical framework of this, inventing and new components, and various ways for applying or modifying system needs. But companies do the main work.

Nebeker:

What about the role of government in paying for R&D? In the United States there's been a lot of debate about how large that should be, whether the government is guessing where to put money.

Wallmark:

Government is very good in supporting the earliest phases of inventions, where companies aren't ready to do anything. That is the place for government to see to it that ideas are nurtured along until they can be judged. After that, government should not be involved.

Nebeker:

So government shouldn't try the sort of marketing end...

Wallmark:

Government has a part in marketing from the point of view of export activities, for example, or from the point of view of furthering the industrial structure of the country. But with individual inventions, the government is most effective, in giving support that is spread over many small cases in their early phase. But determining which ones should get the full-blown support must be done by industry or private inventors.

Nebeker:

Yes. What about the field of microelectronics? Has there been very good communication and teamwork between physicists and engineers over the last forty or fifty years now in microelectronics?

Wallmark:

Well, the main part of the work has been made at Bell Telephone Laboratories and RCA Laboratories in the large, industrial laboratories. Government support for microelectronics came through orders from industrial and military needs that the government had. But the key inventions were made in the large laboratories, and only a few came from outside.

Nebeker:

One criticism I've heard of some areas of academic Electrical Engineering is that the research at universities tends to go its own way, sometimes, far from the applicable technology. Is there anything I haven't asked about you'd care to comment on?

Wallmark:

Well, let me point out that my work at the university on innovations stems of course from pioneering work done in the United States. The U.S. has set examples that we all follow, such as Silicon Valley, which is a model for the entire world. Everyone tries to make a Silicon Valley at home. But also for example the pioneering work by Dr. Fred Terman, who was in the foreground when it comes to university research leading to practical applications. There are several others in the United States that also have provided good models. Also the phenomenon of creating spin-offs from the RCA Laboratories was very interesting, and that was going on all the time. Many were successful. But again, you see the European countries doing similar things now. Much more than they used to. So yes, there are lead people, but also lead companies and lead activities that one can study to imitate in some way.

Nebeker:

Well thank you very much.