IEEE
You are not logged in, please sign in to edit > Log in / create account  

Oral-History:Berthold Bosch

From GHN

(Difference between revisions)
Jump to: navigation, search
(14 intermediate revisions by 4 users not shown)
Line 1: Line 1:
 
== About Berthold Bosch  ==
 
== About Berthold Bosch  ==
  
Berthold G. Bosch was born in Bonn, Germany, on May 30, 1930. He received a degree in electrical engineering in 1956 from Aachen Technical University, Germany; a Ph.D. degree in 1960 from Southampton University, England; a Habilitation from Karlsruhe University, Germany, in 1969; and a D.Sc. degree from Southampton University in 1976. From 1956 to 1957, he held an AEG Foreign Scholarship at the Electronics Department of the University of Southampton. During 1958-1960, as a Research Assistant at the same department, he was engaged in work on microwave-tube noise. From 1960 to 1972, he was with AEGTelefunken in Ulm, Germany, where he occupied various posts in the Tube Works and in the Research Institute, eventually becoming Head of the Electronics Department in the latter institution.
+
[[Image:Berthold Bosch 1462b.jpg|thumb|left|Berthold Bosch]]
  
During that time, he carried out and was responsible for work on microwave tubes, parametric amplifiers, microwave semiconductors, and high-rate PCM circuitry. In 1969, he was made an Adjunct Staff Member (Privat-Dozent) in the Faculty of Electrical Engineering of Karkmhe University. In 1972, he became Professor of Electronics at the Ruhr-University, Bochum, Germany, and simultaneously Joint Director of the Institute of Electronics. During the academic year 1973-1974, he served as Dean of the Faculty of Electrical Engineering. His present research interests include devices and integrated circuits for high-speed electronics, integrated optics, and optical communications. He is coauthor (together with R. W. H. Engelmann) of ''Gunn-Effect Electronics''. Dr. Bosch was awarded the A.F. Bulgin Premium of the British IRE in 1962 (jointly with W. A. Gambling), and in 1969, he received the Annual Prize of the Nachrichtentechnische Gesellschaft.  
+
Berthold G. Bosch was born in Bonn, Germany, on May 30, 1930. He received a degree in electrical engineering in 1956 from Aachen Technical University, Germany; a Ph.D. degree in 1960 from Southampton University, England; a Habilitation from Karlsruhe University, Germany, in 1969; and a D.Sc. degree from Southampton University in 1976. From 1956 to 1957, he held an AEG Foreign Scholarship at the Electronics Department of the University of Southampton. During 1958-1960, as a Research Assistant at the same department, he was engaged in work on microwave-tube noise. From 1960 to 1972, he was with AEG [[Telefunken]] in Ulm, Germany, where he occupied various posts in the Tube Works and in the Research Institute, eventually becoming Head of the Electronics Department in the latter institution.
 +
 
 +
During that time, he carried out and was responsible for work on microwave tubes, parametric amplifiers, microwave [[Semiconductors|semiconductors]], and high-rate PCM circuitry. In 1969, he was made an Adjunct Staff Member (Privat-Dozent) in the Faculty of Electrical Engineering of Karkmhe University. In 1972, he became Professor of Electronics at the Ruhr-University, Bochum, Germany, and simultaneously Joint Director of the Institute of Electronics. During the academic year 1973-1974, he served as Dean of the Faculty of Electrical Engineering. His present research interests include devices and integrated circuits for high-speed electronics, integrated optics, and optical communications. He is coauthor (together with R. W. H. Engelmann) of ''Gunn-Effect Electronics''. Dr. Bosch was awarded the A.F. Bulgin Premium of the British IRE in 1962 (jointly with W. A. Gambling), and in 1969, he received the Annual Prize of the Nachrichtentechnische Gesellschaft.  
  
 
== About the Interview  ==
 
== About the Interview  ==
Line 9: Line 11:
 
Berthold G. Bosch: An Interview Conducted by Frederik Nebeker, IEEE History Center, 31 August 1994  
 
Berthold G. Bosch: An Interview Conducted by Frederik Nebeker, IEEE History Center, 31 August 1994  
  
Interview # 226 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc. and Rutgers, The State University of New Jersey
+
Interview # 226 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.  
  
== <br>Copyright Statement  ==
+
== 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.  
 
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, Rutgers - the State University, 39 Union Street, New Brunswick, NJ 08901-8538 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.  
+
Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, 39 Union Street, New Brunswick, NJ 08901-8538 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:  
 
It is recommended that this oral history be cited as follows:  
  
Berthold Bosch, an oral history conducted in 1994 by Frederik Nebeker, IEEE History Center, Rutgers University, New Brunswick, NJ, USA.  
+
Berthold Bosch, an oral history conducted in 1994 by Frederik Nebeker, IEEE History Center, New Brunswick, NJ, USA.  
 +
 
 +
== Interview  ==
 +
 
 +
INTERVIEW: Berthold Bosch
 +
 
 +
INTERVIEW BY: Frederik Nebeker
  
== <br>Interview  ==
+
DATE: August 31, 1994
  
INTERVIEW: Berthold Bosch<br>INTERVIEW BY: Frederik Nebeker<br>DATE: August 31, 1994<br>PLACE: Ruhr-University, Bochum  
+
PLACE: Ruhr-University, Bochum  
  
 
=== Childhood and Education  ===
 
=== Childhood and Education  ===
Line 53: Line 61:
 
'''Nebeker:'''  
 
'''Nebeker:'''  
  
You built a crystal radio —  
+
You built a crystal [[Radio|radio]] —  
  
 
'''Bosch:'''  
 
'''Bosch:'''  
Line 171: Line 179:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Well, it used to be a fairly general training including electrical machinery, all this heavy equipment stuff. Only after the basic studies and the intermediate exam, could one choose certain majors in the Hauptstudium. To my dislike, I was disappointed to see that at Aachen the chair of Radio Frequency Engineering and Microwaves was not filled because that particular professor at Aachen had been with the Nazi regime, and he had been dismissed, and now there was an empty chair. But just in time for me, somebody came — Professor Döring.  
+
Well, it used to be a fairly general training including electrical machinery, all this heavy equipment stuff. Only after the basic studies and the intermediate exam, could one choose certain majors in the Hauptstudium. To my dislike, I was disappointed to see that at Aachen the chair of Radio Frequency Engineering and Microwaves was not filled because that particular professor at Aachen had been with the Nazi regime, and he had been dismissed, and now there was an empty chair. But just in time for me, somebody came — [[Oral-History:Herbert Doring|Professor Döring]].  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 205: Line 213:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Well it was this sort of happy, fascinating task during school days. I mean, somehow it interested me, and of course during the war with radar, and this and that, it was very fascinating. Getting to shot-down planes, which of course wasn't allowed, and getting pieces of equipment out of them was very fascinating. I recall that the equipment of the American Air Force was quite different from the German.  
+
Well it was this sort of happy, fascinating task during school days. I mean, somehow it interested me, and of course during the war with [[Radar|radar]], and this and that, it was very fascinating. Getting to shot-down planes, which of course wasn't allowed, and getting pieces of equipment out of them was very fascinating. I recall that the equipment of the American Air Force was quite different from the German.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 303: Line 311:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
No. It was a scholarship offered by AEG, the big German firm. In those days, they provided that scholarship every year for one year. It was a lump sum of 4,000 Marks. And you could choose yourself where to go. I applied, and well I was surprised then to get it. One scholarship — in Germany everywhere displayed, apparently not too many people got around to apply there! Of course I would have liked to go to the States then — it was a first choice in those days of course, and today I should think as well — but the money was not sufficient for this. And so, the UK, England, was the next choice, and I had to find a place myself. Then I discovered somehow that there apparently was a department of electronics — the new word "electronics" appeared there. The general description then was electrical engineering. And there was a department of electronics at Southampton University. I wrote to the Professor and he said, "Okay, come."  
+
No. It was a scholarship offered by AEG, the big German firm. In those days, they provided that scholarship every year for one year. It was a lump sum of 4,000 Marks. And you could choose yourself where to go. I applied, and well I was surprised then to get it. One scholarship — in Germany everywhere displayed, apparently not too many people got around to apply there! Of course I would have liked to go to the States then — it was a first choice in those days of course, and today I should think as well — but the money was not sufficient for this.  
 +
 
 +
<p><flashmp3>226_-_bosch_-_clip_1.mp3</flashmp3></p>
 +
 
 +
And so, the UK, England, was the next choice, and I had to find a place myself. Then I discovered somehow that there apparently was a department of electronics — the new word "electronics" appeared there. The general description then was electrical engineering. And there was a department of electronics at Southampton University. I wrote to the Professor and he said, "Okay, come."  
  
 
So I went there in 1956, hardly being able to speak any English. I'd had two years of grammar school English, with Shakespeare and this sort of thing. It was important, but we weren't taught to buy a railway ticket or things like that! Well, I went then, and I got on the ferry boat across the channel, and when I went to Professor Eric Zepler to show myself to him, he said after a few sentences of my English, "Let's speak German," in fluent German. And then I heard from him that he was a born German.  
 
So I went there in 1956, hardly being able to speak any English. I'd had two years of grammar school English, with Shakespeare and this sort of thing. It was important, but we weren't taught to buy a railway ticket or things like that! Well, I went then, and I got on the ferry boat across the channel, and when I went to Professor Eric Zepler to show myself to him, he said after a few sentences of my English, "Let's speak German," in fluent German. And then I heard from him that he was a born German.  
Line 387: Line 399:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Well it was not so much in particular tubes but more general, what is it in reflex klystrons, what in backward wave oscillators, and so on, and why do these noise sources appear and what is their correlation and how far are they spreading out in frequency? Also the background noise, modulation noise, and spikes, ion oscillations, and this sort of thing. Of main interest of course was then the intermediate frequency range of the radar equipment, 30 megacycles per second in those days — a few tens of megahertz away from the carrier. We had to develop then also the measuring equipment, how to measure it. I made a panoramic noise spectrolyzer that later on was taken up and perfected by a British firm, Allscot, a Scottish firm.  
+
Well it was not so much in particular tubes but more general, what is it in reflex [[Klystron|klystrons]], what in backward wave oscillators, and so on, and why do these noise sources appear and what is their correlation and how far are they spreading out in frequency? Also the background noise, modulation noise, and spikes, ion oscillations, and this sort of thing. Of main interest of course was then the intermediate frequency range of the radar equipment, 30 megacycles per second in those days — a few tens of megahertz away from the carrier. We had to develop then also the measuring equipment, how to measure it. I made a panoramic noise spectrolyzer that later on was taken up and perfected by a British firm, Allscot, a Scottish firm.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 419: Line 431:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Well, I was able to determine what amount the modulation noise was in that particular band, AM and FM, and what was background noise, which gave a hint on how to design the tube to lower the noise level. A little bit later, then, a special scheme was developed in the United States and in Europe, more in those days for amplifiers, for reducing the noise on the beam, using the so-called multivelocity region one had to introduce between the actual gun and the start of the traveling wave interaction section — and there one had to do a noise matching of the space charge waves on the beam, and had to arrange it so that there was a minimum of the noise modulation on the beam where it entered the interaction region. This had to do with the sort of thing I had investigated as far as oscillators were concerned. Also I found that there was still a lot, in the tubes of those days, of ion oscillations, suddenly at certain frequencies a large peak in noise occurred, more discrete oscillation. So one could not speak of noise in the actual sense of the word, and there one certainly had to take more care in avoiding this sort of thing when making oscillator tubes.  
+
Well, I was able to determine what amount the modulation noise was in that particular band, AM and [[FM Radio|FM]], and what was background noise, which gave a hint on how to design the tube to lower the noise level. A little bit later, then, a special scheme was developed in the United States and in Europe, more in those days for amplifiers, for reducing the noise on the beam, using the so-called multivelocity region one had to introduce between the actual gun and the start of the traveling wave interaction section — and there one had to do a noise matching of the space charge waves on the beam, and had to arrange it so that there was a minimum of the noise modulation on the beam where it entered the interaction region. This had to do with the sort of thing I had investigated as far as oscillators were concerned. Also I found that there was still a lot, in the tubes of those days, of ion oscillations, suddenly at certain frequencies a large peak in noise occurred, more discrete oscillation. So one could not speak of noise in the actual sense of the word, and there one certainly had to take more care in avoiding this sort of thing when making oscillator tubes.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 437: Line 449:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Yes. Of course, when I finished at Southampton there was the big question of where to go now. When I left for England with this scholarship, AEG wanted to have me when coming back. They wanted to join me the just starting nuclear power plants business, on the instrumentation side. But I said that might be interesting, but not so much for me; microwave engineering was something that I wanted to do. They said, "Well, of course, within our company we have Telefunken, and we sent you there," and they arranged for a meeting, and it was more or less agreed that when I was to go back I would go there, to Ulm. But then I was offered, as we just talked about, this job to stay on at Southampton. But then in 1960, there was a strong wish to go now to the States. I had actually established connections and got an offer by the W.W. Hansen Lab at Stanford. They offered me a research associateship. But certain private reasons and a good offer by Telefunken — which was arranged by Professor Döring of Aachen — made me to return to Germany and join the Microwave Tube Development Lab at Ulm. For two years I worked in designing mainly traveling wave tubes — low noise traveling wave tubes in particular.  
+
Yes. Of course, when I finished at Southampton there was the big question of where to go now. When I left for England with this scholarship, AEG wanted to have me when coming back. They wanted to join me the just starting nuclear power plants business, on the instrumentation side. But I said that might be interesting, but not so much for me; microwave engineering was something that I wanted to do. They said, "Well, of course, within our company we have Telefunken, and we sent you there," and they arranged for a meeting, and it was more or less agreed that when I was to go back I would go there, to Ulm. But then I was offered, as we just talked about, this job to stay on at Southampton. But then in 1960, there was a strong wish to go now to the States. I had actually established connections and got an offer by the [[W. W. Hansen|W.W. Hansen]] Lab at Stanford. They offered me a research associateship. But certain private reasons and a good offer by Telefunken — which was arranged by Professor Döring of Aachen — made me to return to Germany and join the Microwave Tube Development Lab at Ulm. For two years I worked in designing mainly [[Traveling Wave Tube|traveling wave tubes]] — low noise traveling wave tubes in particular.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 521: Line 533:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Yes, I think in England they also started very early at STC, Standard Telecommunications, and then we at Telefunken. We were able to make oscillators and so on, but soon all over the world they started investigating this. Still today one can buy Gunn diodes, as they are called, but in a strict sense if one looks at the definition of a diode, it's not a diode, because it's not a unidirectional electron flow.  
+
Yes, I think in England they also started very early at STC, Standard Telecommunications, and then we at Telefunken. We were able to make oscillators and so on, but soon all over the world they started investigating this. Still today one can buy Gunn diodes, as they are called, but in a strict sense if one looks at the definition of a [[Diode|diode]], it's not a diode, because it's not a unidirectional electron flow.  
  
The application is very limited today: it's used for low power oscillators where the noise level must be quite low. This is a very limited field of application. In those days, one was very excited by it, and we wrote a lot of papers, made patent applications, using it as amplifiers and oscillators or logic devices. The work of today in the field of injection tunnel diodes reminds me of what was going on there. In the move to nanoelectronics where discrete energy levels in the band structure play a part, you can move those levels in various materials by certain bias levels. When they are on the same level, you can get a tunnel current. Again, a one port, or two terminal, device, as the Gunn device was, and again they make a lot of proposals for wonderful things one can do with it, but I myself, because of the experience with the Gunn effect and the Gunn diode (also the tunnel diode), I am doubtful whether much will come out of it.  
+
The application is very limited today: it's used for low power oscillators where the noise level must be quite low. This is a very limited field of application. In those days, one was very excited by it, and we wrote a lot of papers, made patent applications, using it as amplifiers and oscillators or logic devices. The work of today in the field of injection tunnel diodes reminds me of what was going on there. In the move to [[Nanoelectronics|nanoelectronics]] where discrete energy levels in the band structure play a part, you can move those levels in various materials by certain bias levels. When they are on the same level, you can get a tunnel current. Again, a one port, or two terminal, device, as the Gunn device was, and again they make a lot of proposals for wonderful things one can do with it, but I myself, because of the experience with the Gunn effect and the Gunn diode (also the tunnel diode), I am doubtful whether much will come out of it.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 691: Line 703:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Yes. Vergleich we call it in German. They get on an agreement, have to lower their demands and then they can move on in a limited way. That was the case with AEG. But they had to sell a lot of factories, departments, or close some of the departments. The department which was supposed to take over or finally take up the fiber optic work never was able to do it properly after economical breakdown. There was a part at the subfactory which is now A&amp;T at Backnang near Stuttgart that did some work and now belongs to the Bosch Telecommunications Group. But Telefunken itself disappeared more or less completely. In Ulm they carried the name for a while, Telefunken System Technick, but now it's within Daimler-Benz and they use the name DB Deutsche Aerospace AG. So the famous, at least the old famous name of Telefunken has more or less disappeared.  
+
Yes. Vergleich we call it in German. They get on an agreement, have to lower their demands and then they can move on in a limited way. That was the case with AEG. But they had to sell a lot of factories, departments, or close some of the departments. The department which was supposed to take over or finally take up the [[Fiber Optics|fiber optic]] work never was able to do it properly after economical breakdown. There was a part at the subfactory which is now A&amp;T at Backnang near Stuttgart that did some work and now belongs to the Bosch Telecommunications Group. But Telefunken itself disappeared more or less completely. In Ulm they carried the name for a while, Telefunken System Technick, but now it's within Daimler-Benz and they use the name DB Deutsche Aerospace AG. So the famous, at least the old famous name of Telefunken has more or less disappeared.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 725: Line 737:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
No. Now, when I came here in 1972, the question was what to take up, because there was in those years a certain loss of interest in III-V compounds. One was asking, "Are III-V compounds actually the thing one should look after?" Because tunnel diodes, or Gunn devices, didn't show up or didn't show what actually one expected of them, and the III-V galliumarsenide (and so on) transistor hadn't appeared definitely on the scene yet. There was a decline here in this sort of thing, and not yet sufficient work going on III-V transistors.  
+
No. Now, when I came here in 1972, the question was what to take up, because there was in those years a certain loss of interest in III-V compounds. One was asking, "Are III-V compounds actually the thing one should look after?" Because tunnel diodes, or Gunn devices, didn't show up or didn't show what actually one expected of them, and the III-V galliumarsenide (and so on) transistor hadn't appeared definitely on the scene yet. There was a decline here in this sort of thing, and not yet sufficient work going on III-V [[Transistors|transistors]].  
  
 
And so I decided, also considering the amount of funding and expenses, to switch back to silicon, where I had some experience with the early silicon detector diodes for parametric amplifiers and so on, to get away from galliumarsenide and take up silicon. Silicon bipolar transistors, to make them fast and make circuits for gigabit logic circuitry, what we had started to work on at Telefunken. But now relying on silicon for an intermediate stage we used step-recovery diodes for a while, for making pulse amplifiers and multiplexers and got on in the bit rate to three, four, five gigabits per second. But then the aim was building up here a silicon bipolar process line, being able to make now monolithic small ICs with fast high bit rate, using fast bipolar transistors. Again, concerning the motif, with the now as it appeared becoming more and more important optical fiber communications as the background.  
 
And so I decided, also considering the amount of funding and expenses, to switch back to silicon, where I had some experience with the early silicon detector diodes for parametric amplifiers and so on, to get away from galliumarsenide and take up silicon. Silicon bipolar transistors, to make them fast and make circuits for gigabit logic circuitry, what we had started to work on at Telefunken. But now relying on silicon for an intermediate stage we used step-recovery diodes for a while, for making pulse amplifiers and multiplexers and got on in the bit rate to three, four, five gigabits per second. But then the aim was building up here a silicon bipolar process line, being able to make now monolithic small ICs with fast high bit rate, using fast bipolar transistors. Again, concerning the motif, with the now as it appeared becoming more and more important optical fiber communications as the background.  
Line 873: Line 885:
 
'''Nebeker:'''  
 
'''Nebeker:'''  
  
Right, if you were a radar communication engineer —  
+
Right, if you were a [[Radar|radar]] communication engineer —  
  
 
'''Bosch:'''  
 
'''Bosch:'''  
Line 977: Line 989:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
No, it was clear. One could imagine certain applications, somewhere, but the bits and pieces — for instance, the laser, the semiconductor laser hadn't developed that far, so one could say this was a good advance for using it in every home and so on.  
+
No, it was clear. One could imagine certain applications, somewhere, but the bits and pieces — for instance, the laser, the [[Semiconductor Laser|semiconductor laser]] hadn't developed that far, so one could say this was a good advance for using it in every home and so on.  
  
 
=== Silicon Heterojunction Transistors  ===
 
=== Silicon Heterojunction Transistors  ===
Line 987: Line 999:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Well, the change we had here, something which was very fascinating to me, was the move from the silicon bipolar transistor, which we advanced — of course not only we, but other people of course at first. We adopted it for these particular applications: the change from the silicon homojunction bipolar transistor to the silicon germanium heterojunction bipolar transistor. It was Shockley, in his first patent — 1949, I think — who patented the junction transistor, the bipolar transistor, and put in a paragraph that it would be particularly useful if one could make the bipolar transistor so that the band gap in the middle would be larger than in the base. In this way, one should be able to have an injection of carriers, in case of npn sequence of electrons, into the base, and then going on through the collector, but one would avoid a back injection of holes from the base, which lowers the transistor efficiency and so on.  
+
Well, the change we had here, something which was very fascinating to me, was the move from the silicon bipolar transistor, which we advanced — of course not only we, but other people of course at first. We adopted it for these particular applications: the change from the silicon homojunction bipolar transistor to the silicon germanium heterojunction bipolar transistor. It was [[William Shockley|Shockley]], in his first patent — 1949, I think — who patented the junction transistor, the bipolar transistor, and put in a paragraph that it would be particularly useful if one could make the bipolar transistor so that the band gap in the middle would be larger than in the base. In this way, one should be able to have an injection of carriers, in case of npn sequence of electrons, into the base, and then going on through the collector, but one would avoid a back injection of holes from the base, which lowers the transistor efficiency and so on.  
  
 
But it was then completely out of discussion concerning a technological realization. It meant that one had to join two different materials somehow. This later was then possible to put into practice with the III-V compounds, thereby adjusting the combination of the compounds, as we know. One was able to have different materials, compounds, having more or less the same crystal lattice, so they could be grown on top of each other. This led to the III-V bipolar transistor, galliumarsenide—gallium-aluminiumarsenide, for instance, as a combination. And that III-V bipolar heterotransistor is a very successful device. The bipolar transistor as its main feature can deliver much more current which is to be used as a higher power device, or on the other hand can discharge and charge up capacitors or whatever it is in a shorter time, meaning to produce sharper or steeper transients, to go to higher bit rates. Of course III-V technology is expensive and so on; particular applications might justify it, but a certain drawback is the expense. Then, with the improvement in semiconductor technology it was made possible by new epitaxial processes to grow materials one on the other which differed in lattice — having a lattice mismatch to a certain extent — at least growing up to a certain height.  
 
But it was then completely out of discussion concerning a technological realization. It meant that one had to join two different materials somehow. This later was then possible to put into practice with the III-V compounds, thereby adjusting the combination of the compounds, as we know. One was able to have different materials, compounds, having more or less the same crystal lattice, so they could be grown on top of each other. This led to the III-V bipolar transistor, galliumarsenide—gallium-aluminiumarsenide, for instance, as a combination. And that III-V bipolar heterotransistor is a very successful device. The bipolar transistor as its main feature can deliver much more current which is to be used as a higher power device, or on the other hand can discharge and charge up capacitors or whatever it is in a shorter time, meaning to produce sharper or steeper transients, to go to higher bit rates. Of course III-V technology is expensive and so on; particular applications might justify it, but a certain drawback is the expense. Then, with the improvement in semiconductor technology it was made possible by new epitaxial processes to grow materials one on the other which differed in lattice — having a lattice mismatch to a certain extent — at least growing up to a certain height.  
Line 1,071: Line 1,083:
 
'''Nebeker:'''  
 
'''Nebeker:'''  
  
I see. I wanted to ask a fairly general question of what books, if any, have had a very large influence on you in your career. Sometimes someone will say, "Frederick Terman's Radio Engineer's Handbook was my bible for many years." Are there any books which have really played a large role in your life?  
+
I see. I wanted to ask a fairly general question of what books, if any, have had a very large influence on you in your career. Sometimes someone will say, "[[Frederick Terman|Frederick Terman's]] Radio Engineer's Handbook was my bible for many years." Are there any books which have really played a large role in your life?  
  
 
'''Bosch:'''  
 
'''Bosch:'''  
Line 1,083: Line 1,095:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Well, the early days it was the MIT Rad Lab series, and Terman of course — I recently bought again an early copy — I am buying old books now by the way.  
+
Well, the early days it was the [[MIT Rad Lab|MIT Rad Lab]] series, and Terman of course — I recently bought again an early copy — I am buying old books now by the way.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 1,107: Line 1,119:
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Rad Labs, also the Ramo-Whinnery.. Then of course there were some German books as well.  
+
Rad Labs, also the [[Simon Ramo|Ramo]]-[[John R. Whinnery|Whinnery]].. Then of course there were some German books as well.  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 1,135: Line 1,147:
 
'''Nebeker:'''  
 
'''Nebeker:'''  
  
Maybe Shockley's transistor book...?  
+
Maybe [[William Shockley|Shockley's]] [[Transistors|transistor]] book...?  
  
 
'''Bosch:'''  
 
'''Bosch:'''  
  
Yes, but this came later to me, because I was first a microwave tube engineer; of course then Kompfner and Pierce. Naturally books appeared too late to be of use in the actual development. Like Parametric Amplifiers by Blackwell-Kotzebnei, one of the first books published on parametric amplifiers. From Bell Labs later on Sze’s ''Integrated Circuits'' or again Sze’s VLSI book, still later on…  
+
Yes, but this came later to me, because I was first a microwave tube engineer; of course then [[Rudolf Kompfner|Kompfner]] and [[John Pierce|Pierce]]. Naturally books appeared too late to be of use in the actual development. Like Parametric Amplifiers by Blackwell-Kotzebnei, one of the first books published on parametric amplifiers. From Bell Labs later on Sze’s ''Integrated Circuits'' or again Sze’s VLSI book, still later on…  
  
 
'''Nebeker:'''  
 
'''Nebeker:'''  
Line 1,161: Line 1,173:
 
Well, thank you very much.  
 
Well, thank you very much.  
  
<br> ]]
+
[[Category:People and organizations|Bosch]] [[Category:Corporations|Bosch]] [[Category:Engineers|Bosch]] [[Category:Inventors|Bosch]] [[Category:Universities|Bosch]] [[Category:Business, management & industry|Bosch]] [[Category:Business|Bosch]] [[Category:Research and development management|Bosch]] [[Category:International collaboration|Bosch]] [[Category:International trade|Bosch]] [[Category:Culture and society|Bosch]] [[Category:Defense & security|Bosch]] [[Category:World War II|Bosch]] [[Category:Communications|Bosch]] [[Category:Optical communication equipment|Bosch]] [[Category:Optical fiber communication|Bosch]] [[Category:High-speed electronics|Bosch]] [[Category:High speed integrated circuits|Bosch]] [[Category:High-speed networks|Bosch]] [[Category:Multiplexing|Bosch]] [[Category:Demultiplexing|Bosch]] [[Category:Components, circuits, devices & systems|Bosch]] [[Category:Pulse circuits|Bosch]] [[Category:Solid-state circuits|Bosch]] [[Category:Electron tubes|Bosch]] [[Category:Semiconductor devices|Bosch]] [[Category:Thick film devices|Bosch]] [[Category:Integrated circuits|Bosch]] [[Category:Bipolar integrated circuits|Bosch]] [[Category:Integrated circuit noise|Bosch]] [[Category:Thick film circuits|Bosch]] [[Category:Engineered materials & dielectrics|Bosch]] [[Category:Compounds|Bosch]] [[Category:Dielectrics|Bosch]] [[Category:Dielectric materials|Bosch]] [[Category:Fields, waves & electromagnetics|Bosch]] [[Category:Microwave technology|Bosch]][[Category:Lasers, lighting & electrooptics|Bosch]] [[Category:Fiber optics|Bosch]] [[Category:Research and development labs|Bosch]] [[Category:News|Bosch]]
 
+
<br>
+
 
+
[[Category:People_and_organizations]] [[Category:Corporations]] [[Category:Engineers]] [[Category:Inventors]] [[Category:Universities]] [[Category:Business,_management_&_industry|Category:Business,_management_&amp;_industry]] [[Category:Business]] [[Category:Research_and_development_management]] [[Category:International_collaboration]] [[Category:International_trade]] [[Category:Culture_and_society]] [[Category:Defense_&_security|Category:Defense_&amp;_security]] [[Category:World_War_II]] [[Category:Communications]] [[Category:Optical_communication_equipment]] [[Category:Optical_fiber_communication]] [[Category:High-speed_electronics]] [[Category:High_speed_integrated_circuits]] [[Category:High-speed_networks]] [[Category:Multiplexing]] [[Category:Demultiplexing]] [[Category:Components,_circuits,_devices_&_systems|Category:Components,_circuits,_devices_&amp;_systems]] [[Category:Pulse_circuits]] [[Category:Solid-state_circuits]] [[Category:Electron_tubes]] [[Category:Semiconductor_devices]] [[Category:Thick_film_devices]] [[Category:Integrated_circuits]] [[Category:Bipolar_integrated_circuits]] [[Category:Integrated_circuit_noise]] [[Category:Thick_film_circuits]] [[Category:Engineered_materials_&_dielectrics|Category:Engineered_materials_&amp;_dielectrics]] [[Category:Compounds]] [[Category:Dielectrics]] [[Category:Dielectric_materials]] [[Category:Fields,_waves_&_electromagnetics|Category:Fields,_waves_&amp;_electromagnetics]] [[Category:Microwave_amplifiers]] [[Category:Microwave_bands]] [[Category:Lasers,_lighting_&_electrooptics|Category:Lasers,_lighting_&amp;_electrooptics]] [[Category:Fiber_optics]]
+
 
+
[[Category:People_and_organizations]]
+
[[Category:Research_and_development_labs]]
+

Revision as of 19:03, 8 May 2014

Contents

About Berthold Bosch

Berthold Bosch
Berthold Bosch

Berthold G. Bosch was born in Bonn, Germany, on May 30, 1930. He received a degree in electrical engineering in 1956 from Aachen Technical University, Germany; a Ph.D. degree in 1960 from Southampton University, England; a Habilitation from Karlsruhe University, Germany, in 1969; and a D.Sc. degree from Southampton University in 1976. From 1956 to 1957, he held an AEG Foreign Scholarship at the Electronics Department of the University of Southampton. During 1958-1960, as a Research Assistant at the same department, he was engaged in work on microwave-tube noise. From 1960 to 1972, he was with AEG Telefunken in Ulm, Germany, where he occupied various posts in the Tube Works and in the Research Institute, eventually becoming Head of the Electronics Department in the latter institution.

During that time, he carried out and was responsible for work on microwave tubes, parametric amplifiers, microwave semiconductors, and high-rate PCM circuitry. In 1969, he was made an Adjunct Staff Member (Privat-Dozent) in the Faculty of Electrical Engineering of Karkmhe University. In 1972, he became Professor of Electronics at the Ruhr-University, Bochum, Germany, and simultaneously Joint Director of the Institute of Electronics. During the academic year 1973-1974, he served as Dean of the Faculty of Electrical Engineering. His present research interests include devices and integrated circuits for high-speed electronics, integrated optics, and optical communications. He is coauthor (together with R. W. H. Engelmann) of Gunn-Effect Electronics. Dr. Bosch was awarded the A.F. Bulgin Premium of the British IRE in 1962 (jointly with W. A. Gambling), and in 1969, he received the Annual Prize of the Nachrichtentechnische Gesellschaft.

About the Interview

Berthold G. Bosch: An Interview Conducted by Frederik Nebeker, IEEE History Center, 31 August 1994

Interview # 226 for the IEEE History Center, 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, 39 Union Street, New Brunswick, NJ 08901-8538 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:

Berthold Bosch, an oral history conducted in 1994 by Frederik Nebeker, IEEE History Center, New Brunswick, NJ, USA.

Interview

INTERVIEW: Berthold Bosch

INTERVIEW BY: Frederik Nebeker

DATE: August 31, 1994

PLACE: Ruhr-University, Bochum

Childhood and Education

Nebeker:

It's the 31 of August 1994, talking with Professor Bosch; this is Rik Nebeker. So you were born in 1930 in Bonn, and I can see that your education was in four different cities. Was that the disruption of the war period?

Bosch:

Partly by the war, but my father was a civil servant before joining the university as a professor, and civil servants used to be sent about all over the place.

Nebeker:

I see that he was later a professor of education.

Bosch:

That's right. Psychology and Education — a definite difference from my later field and it's still today difficult to talk in common terms.

Nebeker:

Were you interested in science and technology as a young person?

Bosch:

Yes. The usual tinkering and parts building.

Nebeker:

You built a crystal radio

Bosch:

A crystal radio in the first place and so on.

Nebeker:

So had you decided by the time you reached the age to go to university that you would go into engineering?

Bosch:

Yes, definitely. My father sent me to a grammar school — you probably know what it is — you start learning Latin, a second language, the old Greek, and so on; a little bit of math and so on. But I shouldn't say it was the wrong education for me, because I find here that people who have had this sort of secondary school training have learned things that are useful to know as well. The Latin grammar, for instance, sharpens thinking or whatever. But in fact I was the only one in my school leaving class —

Nebeker:

Abitur?

Bosch:

With the Abitur, intending to go for engineering.

Nebeker:

I see. So that was a Humanistic Gymnasium?

Bosch:

Gymnasium, yes. And my principal there was shattered to have somebody like me in his class.

Nebeker:

So where did you go? I guess after the Gymnasium you had some practical work?

Bosch:

Yes, well, this is required before you can enroll for engineering study, half a year.

Nebeker:

I see, and where did you do that.

Bosch:

Well, in a small factory for farm equipment and where in the basic mechanical work you learned how to use a file, or the machine tool shop, and so on, on the lowest level, just using your hands.

Technische Hochschule, Aachen

Nebeker:

And then you entered the Technische Hochschule in Aachen?

Bosch:

In Aachen, which was not by choice, because in those days it was difficult after the war to get into university. The number of places was limited, and I first had admission to Karlsruhe University, but there I would have to do reconstruction work on the ruins, getting the bricks out and so on, for half a year before —

Nebeker:

That was required of all incoming students?

Bosch:

Yes, in general yes. Not everywhere, but in my case at Karlsruhe. Soldiers, dismissed soldiers could start school at once, but somebody like me was supposed to work for half a year. Then Aachen decided to enlarge — Aachen University — so I had a fairly late date concerning the semester. It had started. I could go to Aachen and take up electrical engineering.

Nebeker:

I meant to ask if your family was much affected by the war. Were you in bombed areas?

Bosch:

Yes. In fact, the last place we stayed, at least the family — my father of course being moved about — was Wiesbaden, and this was not bombed until February 1945, but then we were hit, as many others, at that fairly late date. It took off my detector antenna of course — this was certainly not the worst thing of it!

Nebeker:

But it was that close to your house?

Bosch:

Yes, indeed.

Nebeker:

So in 1949 then it was that you started at Aachen? In electrical engineering?

Bosch:

Yes, and in those days you had to pass a more general education test, most of it together with the mechanical engineers up to the intermediate exam.

Nebeker:

I see. Is that the first year or two?

Bosch:

First two years.

Nebeker:

First two years.

Bosch:

And I was supposed to be doing all sorts of things — designing mechanical machine pieces and so on, on the drawing board.

Nebeker:

Is your training in mechanical engineering useful to you?

Bosch:

In a way, yes. I mean, I am able to make a drawing for the workshop, which our students today hardly can do anymore. They need some help — have to talk a lot with the people from the shop. It was different in those days, of course — electrical engineering itself was restricted much more. Today we have different schemes, of course, in training our electrical engineers.

Nebeker:

So what kind of electrical engineering training did you get mainly?

Bosch:

Well, it used to be a fairly general training including electrical machinery, all this heavy equipment stuff. Only after the basic studies and the intermediate exam, could one choose certain majors in the Hauptstudium. To my dislike, I was disappointed to see that at Aachen the chair of Radio Frequency Engineering and Microwaves was not filled because that particular professor at Aachen had been with the Nazi regime, and he had been dismissed, and now there was an empty chair. But just in time for me, somebody came — Professor Döring.

Nebeker:

Yes, I'm going to be seeing him.

Bosch:

We awarded an honorary doctorate last semester to him. He is now 84, I think. It was he who drew my attention or interest to tubes and in particular microwave tubes. You see there is still a piece of it there. My interests since 1960-62 have been in solid-state electronics, but tubes were my first love. I finished my study there in electrical engineering with a certain emphasis on communication engineering. However, it was not very much — the emphasis was not very strong. It was a fairly general education, the whole story of electrical engineering.

Nebeker:

I see. Was it Professor Döring you worked most with in this period?

Bosch:

In fact, yes, also because I was sort of an auxiliary student assistant to him, as the young girl you saw is one here. That is that students in their last years are employed by the department or chair, whatever it is, eight hours or sixteen hours for certain tasks in the lab or doing assistance, perhaps even to the secretary. I was such a person with Professor Döring, I was his personal student assistant, so to speak, who did proofreading, for instance, of his articles — that was one thing.

Nebeker:

I see. And he was working on microwave tubes?

Bosch:

Tubes, yes, he did some rather good research work in the pre-war period, but then also in the war years. The basis was the work by Oscar Herl. This was before the breakthrough by the Varian brothers. During the war, he joined the Lorenz Company, now a subsidiary of ITT. Lorenz was in those days a separate company but later on, after the war, taken over by ITT, and he was head of the tube development at Lorenz before, in 1952, he came to Aachen University. I had waited for high frequency engineering, and then he was there, still in time for us.

Postwar Hardships & Technology Scavenging

Nebeker:

How did you happen to decide on high frequency engineering?

Bosch:

Well it was this sort of happy, fascinating task during school days. I mean, somehow it interested me, and of course during the war with radar, and this and that, it was very fascinating. Getting to shot-down planes, which of course wasn't allowed, and getting pieces of equipment out of them was very fascinating. I recall that the equipment of the American Air Force was quite different from the German.

Nebeker:

You had gotten hold of some of this equipment?

Bosch:

Yes, some pieces of it with tubes —

Nebeker:

That you yourself retrieved from a shot-down plane.

Bosch:

Yes, in one case. One had to be careful, of course.

Nebeker:

It was prohibited?

Bosch:

Yes, surely! It should have — understandably — been handed in for investigation.

Nebeker:

During the war! I see. But then after —

Bosch:

Afterward it was more easy, of course. We were intrigued by the electrical construction, but also the mechanical, because it was lightweight aluminum casing and so on, whereas the German equipment was very heavy stuff, even in the planes and so on.

Nebeker:

I know that some of these planes had very sophisticated analog computers for bombing radars.

Bosch:

Oh? This was beyond my reach —

Nebeker:

So they were very sophisticated for the time: not all electronics, partly mechanical in fact. Well, that's very interesting. So that was an influence on you, seeing these aircraft.

Bosch:

Yes. And then of course after the war there was a lot of surplus equipment around. There was one famous tube — the RV12P 2000 German military tube — then used by every amateur making radio circuits.

Nebeker:

And that was a hobby of yours, to construct circuits?

Bosch:

Yes. Of course it was also of value in exchanging, getting food for radios from farmers or other people who still had flour...

Nebeker:

So you would build radios to trade?

Bosch:

Yes. That's right, in particular for food.

Nebeker:

Were those hard years in that way for your family?

Bosch:

In a certain way, yes, but not to the degree that it was for other people — refugees — my father had a job relatively soon again, but the main difficulty was getting something to eat until 1948 or so.

Nebeker:

Yes. I've heard it was especially hard in the cities.

Bosch:

Yes, of course, in the cities, but we moved in 1946 to a small town; there was a small college where my father was teaching then. But it was a closed circle then. We were newcomers, coming from a different part of Germany, and it was hard there. Although the countryside was around with farmers and so on — It was moderately hard, I should say, not so particularly.

Southampton University & Eric Zepler

Nebeker:

And then after you completed your degree, you received this fellowship — Stipendiat — to work at Southampton. How did you decide on that?

Bosch:

Well, I was in contact with certain companies in Germany to join there, for instance the Lorenz radio and later also television factory. But then I had applied for this scholarship just for fun; I saw the display and thought, "Well, I'll send my application in."

Nebeker:

That was specifically for Southampton?

Bosch:

No. It was a scholarship offered by AEG, the big German firm. In those days, they provided that scholarship every year for one year. It was a lump sum of 4,000 Marks. And you could choose yourself where to go. I applied, and well I was surprised then to get it. One scholarship — in Germany everywhere displayed, apparently not too many people got around to apply there! Of course I would have liked to go to the States then — it was a first choice in those days of course, and today I should think as well — but the money was not sufficient for this.

And so, the UK, England, was the next choice, and I had to find a place myself. Then I discovered somehow that there apparently was a department of electronics — the new word "electronics" appeared there. The general description then was electrical engineering. And there was a department of electronics at Southampton University. I wrote to the Professor and he said, "Okay, come."

So I went there in 1956, hardly being able to speak any English. I'd had two years of grammar school English, with Shakespeare and this sort of thing. It was important, but we weren't taught to buy a railway ticket or things like that! Well, I went then, and I got on the ferry boat across the channel, and when I went to Professor Eric Zepler to show myself to him, he said after a few sentences of my English, "Let's speak German," in fluent German. And then I heard from him that he was a born German.

Nebeker:

You didn't know this —

Bosch:

No, I didn't know this. As a Jew he'd had to leave Germany, was forced to in 1937. In fact he had a major man at Telefunken in Berlin, in the Receiver Design department, and he immediately got a job then in England with Marconi just outside London, at Chelmsford. Actually, I recently wrote a short biographical paper on him, which is one of my hobbies. I think one should remember those people, and I admired him very much. He was responsible at Telefunken for the commercial and military receiver design up to 1932, 33. Then they divided the radio design staff and he was in charge of looking after the military stuff only, and so at once he could do the same at Marconi’s. A funny story is that in the war, in the Battle of Britain, 1940, the war in the air — the German planes carried equipment designed by him and also the British planes had equipment designed by him!

Nebeker:

These were the communications radios?

Bosch:

Yes.

Nebeker:

That's very interesting.

Bosch:

Well, perhaps we should not talk too much about him, but he was a very fascinating person, a very able chess player as well. He was a grand master of I don't know what, and such a title pleased him much more than an honorary doctorate, and in the years that I was in England he was also President of the British Institute of Radio and Electrical Engineers—(not the IEE).

Nebeker:

The IERE, yes. My goodness. And he was still active as a chess player in those years?

Bosch:

Oh yes. He played against himself somehow when he couldn't sleep at night. I don't know how this is possible, but I was told that he was able to do this.

Nebeker:

And then you stayed on obviously longer than the one year you'd planned?

Bosch:

Yes. After half a year they offered me to stay on and work for a Ph.D..

Nebeker:

They offered you a scholarship — a fellowship?

Bosch:

Yes. It was the research assistantship, and paid by the government. All of what I was doing then was on grants from the British Admiralty.

Noise in Microwave Tubes

Nebeker:

Were you working on a specific research project for the Admiralty?

Bosch:

Yes. First it was an investigation of noise in microwave mixer diodes, but this was only for not quite the first year. And then, staying on, I worked again on admiralty grants, on noise in microwave tubes.

Nebeker:

Just understanding...?

Bosch:

Yes, what happened, the sources of noise, different sources of noise, separating them.

Nebeker:

So this was more in the nature of a research grant paid by the admiralty rather than work on some device in particular.

Bosch:

It was a research grant investigating tubes — well, which were used of course, by the Navy.

Nebeker:

Oh, you're investigating noise in particular in tubes?

Bosch:

Well it was not so much in particular tubes but more general, what is it in reflex klystrons, what in backward wave oscillators, and so on, and why do these noise sources appear and what is their correlation and how far are they spreading out in frequency? Also the background noise, modulation noise, and spikes, ion oscillations, and this sort of thing. Of main interest of course was then the intermediate frequency range of the radar equipment, 30 megacycles per second in those days — a few tens of megahertz away from the carrier. We had to develop then also the measuring equipment, how to measure it. I made a panoramic noise spectrolyzer that later on was taken up and perfected by a British firm, Allscot, a Scottish firm.

Nebeker:

And marketed. I see. And you completed your doctorate in 1960.

Bosch:

Right.

Nebeker:

And what was your thesis?

Bosch:

Well, noise: "An Investigation of Noise in Microwave Oscillators."

Nebeker:

Yes, I see it here.

Bosch:

We had close collaboration with British firms in those days, making tubes. They did work for the Admiralty and we did things together, mainly with Mullards, but also with General Electric. They made in those days a very intriguing device — an electrostatically focused oscillator. Normally magnetic focusing, first with coils, solenoids, later on with a panel of magnets was used — but they had an electrostatically focused device which was very appealing. That somehow disappeared; I'm not sure what the reasons were.

Nebeker:

You already mentioned the noise analyzer that had practical importance. Did your thesis work have other practical consequences?

Bosch:

Well, I was able to determine what amount the modulation noise was in that particular band, AM and FM, and what was background noise, which gave a hint on how to design the tube to lower the noise level. A little bit later, then, a special scheme was developed in the United States and in Europe, more in those days for amplifiers, for reducing the noise on the beam, using the so-called multivelocity region one had to introduce between the actual gun and the start of the traveling wave interaction section — and there one had to do a noise matching of the space charge waves on the beam, and had to arrange it so that there was a minimum of the noise modulation on the beam where it entered the interaction region. This had to do with the sort of thing I had investigated as far as oscillators were concerned. Also I found that there was still a lot, in the tubes of those days, of ion oscillations, suddenly at certain frequencies a large peak in noise occurred, more discrete oscillation. So one could not speak of noise in the actual sense of the word, and there one certainly had to take more care in avoiding this sort of thing when making oscillator tubes.

Nebeker:

I see. So this was valuable for design of various —

Bosch:

I hope so, to a certain extent!

Beginnings at Telefunken

Nebeker:

And then you took a job at Telefunken.

Bosch:

Yes. Of course, when I finished at Southampton there was the big question of where to go now. When I left for England with this scholarship, AEG wanted to have me when coming back. They wanted to join me the just starting nuclear power plants business, on the instrumentation side. But I said that might be interesting, but not so much for me; microwave engineering was something that I wanted to do. They said, "Well, of course, within our company we have Telefunken, and we sent you there," and they arranged for a meeting, and it was more or less agreed that when I was to go back I would go there, to Ulm. But then I was offered, as we just talked about, this job to stay on at Southampton. But then in 1960, there was a strong wish to go now to the States. I had actually established connections and got an offer by the W.W. Hansen Lab at Stanford. They offered me a research associateship. But certain private reasons and a good offer by Telefunken — which was arranged by Professor Döring of Aachen — made me to return to Germany and join the Microwave Tube Development Lab at Ulm. For two years I worked in designing mainly traveling wave tubes — low noise traveling wave tubes in particular.

Nebeker:

In the design of those?

Bosch:

In the design, yes.

Nebeker:

What was the application?

Bosch:

Mainly radio links – radio telephone links at microwave frequencies.

Nebeker:

How did you like that work?

Bosch:

Not so much, in particular, so I left it after two years. A new experience — not so much now scientifically or engineering-wise — was the change of coming from an English University to a German manufacturing company. In England, for instance, we started at nine or even half past nine in the morning, and at eleven or so we had a pleasant tea or coffee break, whereas in Germany there at Ulm I was supposed to enter the factory at 7:10 exactly, and when I passed it at 7:12, the doorman shouted after me, "Your name please?" and he took a list and it went up to the superiors. But of course, after two weeks he knew me. But then I discovered they had a good research institute. And I transferred to this after two years.

Nebeker:

And where was that?

Parametric Amplifier

Bosch:

In the same location, even in the same area, a few buildings away. That was then my change to solid-state microwave electronics. And there the big thing was, besides MASERSI work on parametric amplifiers, also tunnel diode amplifiers and so on, and our aim there in particular was to make a parametric amplifier without having to use a circulator. A parametric amplifier is a one-port device, and you have to separate input-output, which was normally done by using a microwave-circulator. And then we devised what we called a four-frequency parametric amplifier, a directional device.

Nebeker:

So you were part of a team developing that?

Bosch:

Yes, it was a team; in particular we had a good theoretician in that team. He is the same age as I; he is now Professor in Hanover Technical University. He is going to be retired, like me, next year at the age of 65.

Nebeker:

Oh, that's mandatory?

Bosch:

Yes, indeed. And a third person — a Fachhochschule engineer — was very good on the practical side. Well, it worked quite well, but parametric amplifiers were superseded five, six years later by low noise transistors. But this again was a very exciting period — we got a new device, a new circuit, but unfortunately it was not of much interest after a few years when it needed to be made into a practical selling device.

Nebeker:

I see. So before it reached that stage to be manufactured it was superseded. Did it find any niche?

Bosch:

Well, they used it for a time at a radio telescope near Bonn, but I think not for very long. And well this of course happened with many things, but I think this is normal for somebody in the research place or even development department. Lots of things are made and only very few of them find their way into application.

Nebeker:

Right. If one thinks of it as a kind of a tree with all these little branches, many of them don't develop very far, but sometimes there's a particular application that some device can live on — And what project did you then go to?

Gunn Effect Research & Patents

Bosch:

After that, the main activity was in the Gunn effect. Gunn at IBM in 1964, I think it was, discovered the possibility to get oscillations from a piece of galliumarsenide, from a bulk piece of galliumarsenide, only two electrodes for applying a DC bias and then it started to oscillate at microwave frequencies, which puzzled people for a fairly long time. But after I think three years or so, Herbert Krömer was one of those who said, "Oh well, the principle underlying it has been proposed many years ago by Ridley and Watkins — it has to do with the special band structure, the conduction band structure of galliumarsenide, with sub-bands besides the main band. Electrons are transferred from the main valley into the conduction band, to the satellite valleys where the mobility is higher."

Nebeker:

Had Gunn discovered this sort of by accident or was he looking?

Bosch:

By accident, and of course this again is one of those examples where he was looking for something completely different, that he saw something strange and didn't dismiss it because he wasn't looking for it, but said, "Oh, what is this?" and this then was the “Gunn effect” which received very much attention in those days. I heard very early about it before he published results, because of a visit to Yorktown Heights. They didn't tell much about it, but from what they said one could imagine or at least try to repeat it. In those days it was in the first place hard to get a piece of galliumarsenide sufficiently monocrystalline and so on, but we were able to do it and to get also the oscillations and then it started.

Nebeker:

So your group immediately, as soon as you heard of the Gunn effect, tried to reproduce it?

Bosch:

Yes, I think in England they also started very early at STC, Standard Telecommunications, and then we at Telefunken. We were able to make oscillators and so on, but soon all over the world they started investigating this. Still today one can buy Gunn diodes, as they are called, but in a strict sense if one looks at the definition of a diode, it's not a diode, because it's not a unidirectional electron flow.

The application is very limited today: it's used for low power oscillators where the noise level must be quite low. This is a very limited field of application. In those days, one was very excited by it, and we wrote a lot of papers, made patent applications, using it as amplifiers and oscillators or logic devices. The work of today in the field of injection tunnel diodes reminds me of what was going on there. In the move to nanoelectronics where discrete energy levels in the band structure play a part, you can move those levels in various materials by certain bias levels. When they are on the same level, you can get a tunnel current. Again, a one port, or two terminal, device, as the Gunn device was, and again they make a lot of proposals for wonderful things one can do with it, but I myself, because of the experience with the Gunn effect and the Gunn diode (also the tunnel diode), I am doubtful whether much will come out of it.

Nebeker:

So the general phenomenon here, if we generalize from those two examples, is the new effect that one might use in a range in devices, and then there's a rush to patent the different devices. I see you have a very large number of patents, and many of those relate to these Gunn effects.

Bosch:

Most of them, yes.

Nebeker:

And that was Telefunken that wanted to protect its position —

Bosch:

Of those twenty or twenty-five patents concerning the Gunn effect, only one has come to practical use.

Nebeker:

Which one is that?

Bosch:

It's a tuning effect, a synchronization tuned Gunn oscillator.

Nebeker:

I would like to know more about it.

Bosch:

Sometimes the patent only says "Improvements in or Relating to — " It's one of these. Well, it's in fact also a US patent, so...

Nebeker:

If it's not possible to tell from those titles.

Bosch:

No, here it is: "Frequency Control and Synchronization..." This one here.

Nebeker:

I see. You also have a German patent?

Bosch:

Yes.

Bosch:

For us of course then the US patent was of more importance — it's a limited German market.

Nebeker:

And this is frequency control and synchronization of Gunn oscillators, and that's been licensed to the manufacturers?

Bosch:

Yes, it's licensed to a number of manufacturers but, well, it's expired of course by now.

Nebeker:

Were there any of the earlier patents here that were important in industry?

Bosch:

Not to that extent. I mean, the parametric amplifiers weren't, and the low noise electron gun was used in traveling wave tubes for a certain time, but only I should say for about five years or so. There is a US patent as well. Later on, all the Gunn effect patents, as far as I know, didn't bring much, get much importance.

Nebeker:

Did you continue to patent things after you left Telefunken?

Bosch:

Not much. I think one patent was there. But patenting at a university is a rather complicated affair; either you do it privately, and then it's expensive. Involving the University it's rather difficult... a lot of paperwork is involved.

Telefunken & Optical Communications

Nebeker:

Okay. So we had that period of work on the parametric amplifiers, then a period of work on the Gunn effect. Did that continue until or through 1972?

Bosch:

Yes. But in about '69 or 1970, we — well, my group took up another topic which now was the field of logic circuitry at very high bit rates. Bit rates increased continuously and reached now the spectral frequencies of the microwave range.

So they asked me — I was responsible for microwave electronics in the research institute — to look after this. It was in connection with the developing wide band optical communication. This was the fiber optic communications Telefunken was involved in very much, rather early. Now, on the electrical side, one has to provide signals with the sufficiently high bit rate. They asked on the transmitter side for laser modulators, for multiplexers combining the lower bit rate signals. And then on the receiving side, amplifiers of course, demultiplexers, decision circuits, and so on.

Nebeker:

I see. So Telefunken was trying to develop the whole system?

Bosch:

Yes. Yes, the whole system. It was done in the research institute, and it was difficult to convince the manufacturing plants that this was important. I think it's the usual thing — one has to do much work in persuading the people that this is something worthwhile. There were people working on the fibers, others on the lasers, on the photodiodes for detecting —

Nebeker:

Is this all your group or different groups?

Bosch:

No, there were different groups. In my group, the Microwave Electronics group, we took up the circuitry work for the high bit rates. Where I recall, in 1971 for instance, one gigabits per second, then an extremely high bit rate for us. And the circuits were made as hybrid circuits. Thick film or whatever, with discrete transistors then put in.

Nebeker:

And how much was the design of such devices constrained by manufacturing difficulties — the difficulties of constructing them?

Bosch:

Well, what we did of course was research work, single pieces and of course we had connections, close connections to the department which they called at Ulm the pre-development department. Each manufacturing plant used to have, besides the actual down-to-earth developments, departments for more basic work. Often they were not on good terms with the research institute of the overall company.

Nebeker:

Why is that?

Bosch:

Well, one reason for instance is that all the subfactories had to partly pay for the research institute, certain fixed dues. Whether something useful for them came out or not they had to pay, which was not such a good idea for them, of course. Would it be useful to them? Very often the meaning of what is useful varied. But this I think happens in every company, more or less, between the R & D department and the manufacturing department, leading to a certain amount of deadlock.

Nebeker:

Inertia?

Bosch:

Friction losses in that respect. But on the whole, the research institute transferred a lot of things into the factories, for instance with computers. AEG used to make large scale (large-scale in those days) computers with a plant at Konstanz. In our case, with the circuitry or in general with the fiber optic work, it was still too early in those days.

Nebeker:

Too far from a manufactured product.

Bosch:

Too far from real applications, and really felt needs in the manufacturing plants. They were very much impressed by what one could do with those optic fibers. And they used to say, "This is the system of the future, but the future is still well away. But you do work on it, perhaps later on we will come back to it; we have more immediate needs now." There was indeed a lot of research work, as I said, going on, which started in the mid- sixties. One of my colleagues, Manfred Börner, who now is Professor in Munich — just retired — holds a basic patent on it. It took a long time to be issued.

Nebeker:

Basic patent on what?

Bosch:

On the whole system, semiconductor laser, a possibly monomode fiber, then a semiconductor diode detector, for use as a communication system. ITT got the first patent, but later on Börner was recognized as well and also got his patent for AEG. But it was a long struggle until he came through.

Nebeker:

And how did the work at Telefunken contribute to the development of fiber optics?

Bosch:

There were I think quite a number of good special contributions besides proposing the basic system, but Telefunken and in particular the mother company AEG, soon after I left, got into difficulties, as you perhaps know. AEG almost disappeared; they — what is the term in English? — came near bankruptcy.

Nebeker:

The protection Chapter 11 before bankruptcy?

Bosch:

Negotiations between —

Nebeker:

Yes, the creditors.

Bosch:

Yes. Vergleich we call it in German. They get on an agreement, have to lower their demands and then they can move on in a limited way. That was the case with AEG. But they had to sell a lot of factories, departments, or close some of the departments. The department which was supposed to take over or finally take up the fiber optic work never was able to do it properly after economical breakdown. There was a part at the subfactory which is now A&T at Backnang near Stuttgart that did some work and now belongs to the Bosch Telecommunications Group. But Telefunken itself disappeared more or less completely. In Ulm they carried the name for a while, Telefunken System Technick, but now it's within Daimler-Benz and they use the name DB Deutsche Aerospace AG. So the famous, at least the old famous name of Telefunken has more or less disappeared.

Nebeker:

Like RCA.

Bosch:

Yes, indeed.

Nebeker:

So that was your last work there before taking a professorship here?

Bosch:

Yes. One of the last things still concerned with the Gunn effect was a semiconductor traveling wave amplifier using the Gunn effect as a gain-providing medium, which looked quite interesting for a while, but with the progress in transistor development, lowering noise figures, higher frequencies, this again was of no use after a few years. My staff people presented me when I left with a traveling wave amplifier with their names engraved on the pedestal, which I have down here — can you see? This was the last piece of hardware there.

Nebeker:

And what is that?

Bosch:

This is actually a piece of galliumarsenide made with electrodes and so on. A two-port device, microwave amplifier device using the Gunn effect — the space charge waves are excited, and amplified when they pass through the active medium there and can be coupled out at the end. I worked at three gigahertz or so. But again, this was more or less a research result only.

Nebeker:

It didn't reach the manufacturing stage.

Ruhr University and Silicon ICs

Bosch:

No. Now, when I came here in 1972, the question was what to take up, because there was in those years a certain loss of interest in III-V compounds. One was asking, "Are III-V compounds actually the thing one should look after?" Because tunnel diodes, or Gunn devices, didn't show up or didn't show what actually one expected of them, and the III-V galliumarsenide (and so on) transistor hadn't appeared definitely on the scene yet. There was a decline here in this sort of thing, and not yet sufficient work going on III-V transistors.

And so I decided, also considering the amount of funding and expenses, to switch back to silicon, where I had some experience with the early silicon detector diodes for parametric amplifiers and so on, to get away from galliumarsenide and take up silicon. Silicon bipolar transistors, to make them fast and make circuits for gigabit logic circuitry, what we had started to work on at Telefunken. But now relying on silicon for an intermediate stage we used step-recovery diodes for a while, for making pulse amplifiers and multiplexers and got on in the bit rate to three, four, five gigabits per second. But then the aim was building up here a silicon bipolar process line, being able to make now monolithic small ICs with fast high bit rate, using fast bipolar transistors. Again, concerning the motif, with the now as it appeared becoming more and more important optical fiber communications as the background.

Nebeker:

That was the main reason for developing these high rate circuits?

Bosch:

Yes. The electronic circuitry at both ends of the fiber. This was, I think, a period of ten, fifteen years, rather fruitful years, making more and more sophisticated circuits, going up in the bit rate up to twenty gigabits per second. The circuits became more complex within our framework of complexity, a maximum perhaps of three hundred, four hundred transistors on one chip. But even yet expanding to, for instance, fast AD converters as well, with sample frequencies of a few — one, two, and the last one three, gigasamples per second. When we reached about five years ago, the mark of twenty gigabit per second, we found that the possibilities of my clean room fabrication facilities were at an end; I would have had to install lots of things, invest very much money, which at a university is rather difficult to do; of course, in my younger days perhaps I would have had the energy and found the means, but there was not much point now in starting once again.

Nebeker:

But you did have facilities for doing these ICs.

Bosch:

Yes, we had and still have. I had in my group some very able people, who were able to make progress by — for instance, going from the non-self-aligned to the self-aligned transistor structures, lowering the design rules to 0.8 microns. This was the smallest structure we could do. But then we were forced to cooperate in the more intense way with outside partners, industrial companies outside.

Changes in University Funding

Nebeker:

May I ask before going on how it compared trying to do research here with doing research in industry before that, in terms of the assistance you had, in terms of the equipment and other resources?

Bosch:

The way of approaching targets, I think, in industry, in the research institute of Telefunken, was in a certain way up to the researchers in charge. There were certain guidelines because of the activity of the company as such. But within this framework one was able to choose goals to a good extent. One was not rigidly fixed; no one said, "Now, you do exactly this or that." The resources were fairly good, I should say, in those days; they were prosperous years and the company was growing. This was quite satisfying. When I came here, the equipment, or the possibilities were ok, the financial funding was good, because again in 1972 the money was there. This university was founded in 1965 as one of the big new universities in Germany, and money poured in from all sides. In contrast to what often was a problem at universities, I should say, I had sufficient technical staff. Including — and this is important – non-academic staff, technical staff for the labs and so on. And also it was not difficult to get research money from different sources.

Nebeker:

Industry?

Bosch:

Yes, but also the government ministries. The German Ministry for Research, BMFT — this is still our main source of supply. Times have become more difficult, as everywhere, I think; and there is now hardly any money coming from the university itself. Which in 1972 was not the case. Nowadays I am relying totally on what we call Drittmittel – third-party sourcing — either industry or governmental ministry funding.

Nebeker:

So all of the support of your laboratory has to come from outside.

Bosch:

That's right concerning recent years. Except for the staff. It's different with the academic staff. Most of them are paid from the grants, but with the permanent technical staff, they are paid by the university. The staff is paid from the “Haushalt.”

Nebeker:

The usual budget.

Bosch:

Yes, the usual budget.

Nebeker:

So you didn't feel much constrained then in your research having come here?

Bosch:

No, coming from an industrial research institute, no, not concerning money or staff. Of course, I had to build up this Chair, as we call it, from nothing: it didn't exist so far because it was a new university, and this took me about two and a half years until I could here publish my first paper. This building up and —

Nebeker:

That means getting other faculty members? Students?

Bosch:

Coming to this university, I had to prepare lectures, and this and that —

Nebeker:

Set up the laboratory.

Bosch:

— and so on. I wanted to establish a new field of activity, as I said I thought it better to go for silicon at that moment, silicon devices, bipolar transistors and so on, to improve them. This took time, building up the clean room facilities and so on, getting people trained to work on it who started new on it.

Publication on Gunn Effect

Nebeker:

Was it a delayed publication? Well, that's not the way to put it. Your 1975 book on the Gunn effect: can you —

Bosch:

It came out in 1975, yes.

Nebeker:

Right. You weren't researching that at the time, is that right?

Bosch:

Yes. This was still the outcome of what we did. Co-author Engelmann was my main investigator. I found him more or less at Hewlett-Packard. He worked there for a couple of years and wanted to go back to Germany. It took some time to write the book; particularly for me as I had moved to this place, so it was published somewhat late. But, this is the way...

Nebeker:

Putting aside any modesty, is that book a good characterization of that field, Gunn effect electronics?

Bosch:

Well, I of course have my own feeling, meaning on it — I think it's not bad. It also covers the fundamentals and should be readable for an engineer, apart from a solid-state physicist. It also covers the applications, it covers the making of devices. And we tried in all those parts to go a bit into the basic depths, not just the phenomenological surface. But in 1975, well, the Gunn was still of importance, but in limited fields. It's a book for a small group of interested people.

Sabbatical Travels in US

Nebeker:

Okay. I also wanted to ask you about some of this work you've done away from here since then. I guess it was first in 1976 that you took a sabbatical, if it's called that here, and did some work at Hewlett-Packard and Stanford and Texas Instruments. Is that Southern Methodist University? How did you do that? Those were laboratories you were interested in?

Bosch:

Yes. People who worked on similar lines; we had contacts established on conferences. Or with HP, there were two or three former Germans I knew. At TRW it was a bit difficult to get in because of the military funded work.

Nebeker:

What was that?

Bosch:

It was, well, fast bipolar circuitry. TRW at about that time was the first to be able to make a one micron-emitter bipolar emitter for ICs. They were of course a bit reluctant to let out too much of this knowledge. HP Labs was completely different of course, and universities of course also.

Nebeker:

So you arranged to spend a week or two or longer at these places?

Bosch:

Yes, two to three weeks was the most I think at HP, but mostly it was traveling about and having a look and talking to the people I knew there. It was in those days the gigabit electronics which interested me and a lot of other people.

Nebeker:

Was most all of that research open in the full public domain, or was there much of it like the TRW work?

Bosch:

Mainly it was not classified — well, mostly it was military sponsored work, but to a great extent open. I visited for instance Max Yoder at the Naval Research Lab on invitation, and who were interested in using it for sophisticated radar systems, where they of course didn't disclose much about the system in general, but special signal processing parts of it could be talked about.

Nebeker:

So it didn't seem to you, if you're studying gigabit circuitry, that these security concerns hampered communication?

Bosch:

No. Not in most cases in general. Concerning the applications one saw for the future —

Nebeker:

Right, if you were a radar communication engineer —

Bosch:

Radar is different, of course, but my interest was not in radar. On the other hand the radar people were interested in part of what was going on here. One simple application was just connecting an actual radar site to the processing unit by optical fibers, and getting over as much bits as possible. Or Max Yoder was interested in fast AD converters, having a sampling rate as high as possible, but he didn't need many bits; the resolution was fairly low...

Nebeker:

What was the application in mind?

Bosch:

Well, far ranging detecting of objects.

Nebeker:

What sort of detecting?

Bosch:

Planes or whatever. Radar applications on large ships, where they were even investigating using devices cooled by liquid helium.

Nebeker:

To reduce noise?

Bosch:

Yes. To reduce noise, but also to be able to use Josephson-junction for detecting. But again, these were research projects and I haven't heard of actual final applications of this sort of thing.

University of Newcastle and III-Vs

Nebeker:

You were also about this time, 1977, a guest professor at the University of Newcastle. How did that come about?

Bosch:

Yes. One summer. There was a Professor Hartnagel, a German, who was with me in the early days at Telefunken in my electron tube phase. How did he get to England? Well, I should say by meeting and marrying an English girl. Particularly his object then was to achieve a galliumarsenide oxide — it was considered and still is a shortcoming — that with galliumarsenide one hasn't got a native oxide, as in silicon. He was trying to get some artificial oxide which would do the job, but in the end that was not particularly successful.

Nebeker:

Why were you attracted to spending some time at Newcastle?

Bosch:

Well, for one point, it was to renew acquaintance with him, but then also at this moment as we were doing silicon work, to have an opportunity to look closer at the rival III-V, to talk to people working on the III-Vs, which all the time I used to watch, not to limit myself to looking only at silicon. But to be able to keep abreast of what was going on in the III-V field. I have kept this attitude, and for instance our Ministry of Research has me put on a III-V project, a national project they have, because they think I know enough also about III-Vs. But I have a skeptical view on it, knowing the silicon field. I think it is obvious that one knows what is going on in one’s own field. But for what is going on in the rival field, a lot of people know it only after a delay of two years or so. Until the publications come out, or the contributions at conferences — when it is presented there, it is, well, not the latest stuff. So they often get into the wrong position when judging what is going on in the other field, and perhaps might take the wrong decisions for themselves because they are not aware of the latest competition results. And that was also one of the reasons I went to Newcastle.

Progress in Gigabit Electronics

Nebeker:

I took a look at your very nice review article from 1979 on gigabit electronics. This is a tall order, but can you in short time give an overview of what's happened in the fifteen years since then in gigabit —

Bosch:

In the field? May I look at that? It’s a relatively long time ago.

Nebeker:

[Laughing] Have you forgotten exactly what you said in 1979?

Bosch:

The main advancement since is the development of ICs, monolithic integration. Here in 1979 are still hybrid integrated circuits, which have of course completely disappeared. Now we have ICs working up to forty gigabits per second. Also, as I mentioned, the bit rate having gone up from one or two gigabits per second in those days to thirty, forty gigabits per second.

Nebeker:

Has that been the result of one or two breakthroughs, or a more steady advance?

Bosch:

I think it's making use mainly of the steady advance in IC development and the logic gates. For instance, the optical communication applications I mentioned have now found also application even in computer circuits — high bit rates, or throughput and whatever it is, or AD converters with much higher sample rates.

Nebeker:

Right. I happen to have talked to some people in seismology, and these very fast AD converters with the very high bit rates were very important to get these wide band seismographs — to be able to get the whole or a very large part of the frequencies, the spectrum in just a single instrument instead of having specialized seismographs. I can just imagine that there must be quite a few areas of application where being able to have a very high bit rate would be important. You mentioned in that article, of course, radar, and sensing systems, and obviously communications with fiber optics. Are there other applications areas that you'd mention where the very high bit rate has been important?

Bosch:

Well, I think there are quite a number of very special applications, the processing of those seismic signals might be one. But here and now our industry partners are heading to or involved in the fiber optic work, so there is no definite connection to somebody who says, "I am looking for earthquakes," or whatever it is. This fiber work with the highest bit rates, two gigabits per second to every home and so on, is such a main topic, and something apparently which enables industrial companies to earn money, hopefully. This is more or less the only and the main application I am concerned with. I should say that at that time, 1978, there were a lot more applications mentioned which might have been using such circuitry; there are certainly some of them still making use of such circuitry, but it's completely overridden by this fiber optic communications, fiber optic communication application.

Nebeker:

So that's driving the field? Well, that's obviously of so great an importance, that's not surprising.

Bosch:

There are a lot of European Community projects everywhere all over Europe — there is much work going on in that field.

Nebeker:

So over the last fifteen years or so since that article was written you have had this increase that you've mentioned.

Bosch:

Increase in bit rate, and the important point of further development in IC technology to smaller dimensions —

Nebeker:

— and the establishment of fiber optics communications. Maybe at that time it wasn't clear that it was going to be —

Bosch:

No, it was clear. One could imagine certain applications, somewhere, but the bits and pieces — for instance, the laser, the semiconductor laser hadn't developed that far, so one could say this was a good advance for using it in every home and so on.

Silicon Heterojunction Transistors

Nebeker:

Can you sum up the way your own research here has gone over the last ten or fifteen years?

Bosch:

Well, the change we had here, something which was very fascinating to me, was the move from the silicon bipolar transistor, which we advanced — of course not only we, but other people of course at first. We adopted it for these particular applications: the change from the silicon homojunction bipolar transistor to the silicon germanium heterojunction bipolar transistor. It was Shockley, in his first patent — 1949, I think — who patented the junction transistor, the bipolar transistor, and put in a paragraph that it would be particularly useful if one could make the bipolar transistor so that the band gap in the middle would be larger than in the base. In this way, one should be able to have an injection of carriers, in case of npn sequence of electrons, into the base, and then going on through the collector, but one would avoid a back injection of holes from the base, which lowers the transistor efficiency and so on.

But it was then completely out of discussion concerning a technological realization. It meant that one had to join two different materials somehow. This later was then possible to put into practice with the III-V compounds, thereby adjusting the combination of the compounds, as we know. One was able to have different materials, compounds, having more or less the same crystal lattice, so they could be grown on top of each other. This led to the III-V bipolar transistor, galliumarsenide—gallium-aluminiumarsenide, for instance, as a combination. And that III-V bipolar heterotransistor is a very successful device. The bipolar transistor as its main feature can deliver much more current which is to be used as a higher power device, or on the other hand can discharge and charge up capacitors or whatever it is in a shorter time, meaning to produce sharper or steeper transients, to go to higher bit rates. Of course III-V technology is expensive and so on; particular applications might justify it, but a certain drawback is the expense. Then, with the improvement in semiconductor technology it was made possible by new epitaxial processes to grow materials one on the other which differed in lattice — having a lattice mismatch to a certain extent — at least growing up to a certain height.

So this opened up the way to make devices using compounds like silicon germanium (that's silicon with for example twenty percent germanium in it) with monocrystalline silicon. There the band gap is different; one can arrange it so as to realize the old Shockley proposal — a device now which is silicon based. It's almost all silicon with all the advantage of silicon oxide that you can use, all the experience one has gained over the many years of silicon technology, and having only at one place or two places a small silicon germanium layer. Now one has the possibility of making better transistors with a higher cut of frequencies, higher current density, and it is possible now to achieve a performance more or less equal to those III-V transistors, a performance that is two to three times better in transit frequency or whatever it is you are interested in – for example lower power, with is a tradeoff between the two. Silicon based devices achieving an effect which until then could only be achieved by III-V devices.

Nebeker:

And that's been a major part of your work?

Bosch:

Yes. We entered here in this field fairly early, in conjunction with Telefunken Research Institute, which is now Daimler-Benz Research Institute. They early entered the field of growing such semiconductor layers. I didn't enter this field of growing material here myself, but they provide us with layers. We then made transistors or chips from these layers here in our facility, or use other facilities outside when it's required to go below the level of, as I mentioned, 0.8 microns. So I had the pleasure, in my last period here, so to speak, to be able to still play around with a new device, which I think is gaining more and more importance. They not only useful, these heterojunctions, silicon-germanium, for bipolar devices — this was the first device we had found application for — but also for MOS transistors, where one can increase the hole mobility in the channel transistor. It also can be used for photodetector diodes; it is used even for optical wave guides, and a lot of other things. For instance, heterojunction field-effect transistors. There is a whole new field of applications with improved performance, now silicon based and not III-V based.

Nebeker:

I see. And was this in opposition to the trend of the eighties to go to the III-V devices?

Bosch:

Yes, of course the III-V are making progress continuously; there is a saying that the III-V will always be the material of the future! I mean, they were leading all the time — let's take the bit rate as one thing of interest. Silicon followed because of improved processing facilities, making areas smaller, putting them closer together, and avoiding delays and so on. It was a follow-on effect of two to three years behind. But they were cheaper than those III-Vs. But, suddenly something new appeared which exactly rivaled those applications. In the photoemission field, that's a different story: but even there the III-Vs of course might not stay unrivaled, because they try now to make also light-emitting devices on the basis of silicon-germanium material. Now suddenly this distance which was there all the time between silicon and the III-Vs seems to shrink or disappear, and the circuits as far as we are concerned based on the silicon-germanium compound are practically as fast as the III-Vs. This was something very interesting.

Nebeker:

And very important because of the advantages of the silicon.

Bosch:

It's important to see how the number of papers at conferences have been increasing – and the number of the companies, the industrial companies entering the field. There's hardly anybody left now who doesn't work also in this field.

Consulting Work and IC Design

Nebeker:

I wanted to ask about your consulting work. You've done a fair amount of that. Is that something that has been valuable to you for your research here?

Bosch:

Yes, indeed. Consulting to a few companies, and foremost to my old company. One thing is for them to become informed about latest trends, which is more general, and then wanting to have carried out particular tasks in connection with what we are doing here. Fast AD converters right now for a special application in the R&D department. Then I had for a couple of years a consultantship with Standard Electrik Lorenz, ITT, but about five years ago they were sold to Alcatel in France. This work dealt with fast circuitry for optical communication systems in which they were involved very heavily. This was industry consulting. The other part is consulting work with governmental agencies, to referee on research —

Nebeker:

— proposals —

Bosch:

Yes, and so on, and in the course of the work, having meetings, discussing what they have done... I have to visit labs to have a look at what the progress is. The third thing is a bit different from strict consulting. I founded with two young people who took a Ph.D. here, a small company which is situated on the campus.

Nebeker:

I see. What are the products?

Bosch:

Well, it now has about twenty people working for it; it is not very large, more a consulting firm. It has two fields of activities: one is special IC design, in particular fast circuitry —

Nebeker:

I see. So some company wanting that kind of circuitry would come to you for IC design.

Bosch:

Yes. And during the recession which I guess we do have at the moment but hope we will have gone through soon, this actually was a good period for us, because a lot of companies like Siemens, or others as well, were more inclined to give certain tasks away, not having the manpower themselves to keep them up. They then gave tasks to the outside.

Nebeker:

That's become a tendency in the US too.

Bosch:

The outsourcing — and so we managed to grow over the past three to four years. It's not an undertaking which makes a lot of profit, but we were able to grow from three people now to about twenty, and are able to support these people and to buy some new equipment and software. Software started with IC design software and now is spreading to other topics — I don't understand very much about the latter. One main thing for example is databases for computing in the language C. They actually have a partial sales activity for one US firm, where they sell their databases.

Nebeker:

I see, a German representative?

Bosch:

Middle European representative, yes. This works quite well, but it spreads to fields a bit outside of my scope. Our consultants train people on these software products, even customers like banks or insurance companies, which use these systems. We have just now on the first of January of this year separated the firms into separate companies: IC design and object-oriented databases.

Influential Books in the Field

Nebeker:

I see. I wanted to ask a fairly general question of what books, if any, have had a very large influence on you in your career. Sometimes someone will say, "Frederick Terman's Radio Engineer's Handbook was my bible for many years." Are there any books which have really played a large role in your life?

Bosch:

From the early days?

Nebeker:

From the early days.

Bosch:

Well, the early days it was the MIT Rad Lab series, and Terman of course — I recently bought again an early copy — I am buying old books now by the way.

Nebeker:

You have the first edition?

Bosch:

It's the second edition of 1937.

Nebeker:

Yes, that's all I could get either!

Bosch:

Well, what books then?

Nebeker:

So the MIT Rad Lab series.

Bosch:

Rad Labs, also the Ramo-Whinnery.. Then of course there were some German books as well.

Nebeker:

Are there any that stand out in your mind?

Bosch:

Yes: Küpfmüller. It's Theoretische Elektrotechnik and Vilbig’s Hochfrequenz-technik. Let's see where they stand. More practically, when I went to take my Ph.D. in England, there was a very thick U.S. book of practical use by the Cruft Electronics staff, the radar lab people, on circuits and tubes.

Nebeker:

What was the title of the book?

Bosch:

Electronic Circuits and Tubes, or something very similar. Let me see...also not to be overlooked... in the early days, the Radio Engineering Handbook by Henney — the first edition was in '35 I think and then it was revised almost every year.

Nebeker:

Of course you'd been working in a pretty new area so it's —

Bosch:

Yes, there weren't any books yet in our particular fields, but of course one had to rely on fundamental works.

Nebeker:

Maybe Shockley's transistor book...?

Bosch:

Yes, but this came later to me, because I was first a microwave tube engineer; of course then Kompfner and Pierce. Naturally books appeared too late to be of use in the actual development. Like Parametric Amplifiers by Blackwell-Kotzebnei, one of the first books published on parametric amplifiers. From Bell Labs later on Sze’s Integrated Circuits or again Sze’s VLSI book, still later on…

Nebeker:

That's very helpful.

Bosch:

Eberhard Spenke’s Electronic Semiconductors was very much used and still is a useful book, at least in Germany, containing much of the basic stuff. Spenke worked from the ‘30s, but his main activity then was with selenium and other heavy current semiconductor rectifiers. This is a good book. Concerning basic semiconductor work, it was a useful book. And here, look…

Nebeker:

Millman and Taub, Pulse Digital Waveforms. I see.

Bosch:

That is just a random sample from the shelf. I at least am somewhat surprised to be interviewed, because one doesn't quite feel part of that class who might particularly contribute to history, but on the other hand in twenty or thirty years it might look different. I think that is one of your aims, to get it on record as long as it is possible.

Nebeker:

Well, thank you very much.