Oral History Transcript — Dr. Nick Holonyak
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
See the catalog record for this interview and search for other interviews in our collection
Interview with Dr. Nick Holonyak
By Babak Ashrafi At Dr. Holonyak's Office in Urbana, IL
March 23, 2005
Nick Holonyak; March 23, 2005
ABSTRACT: Topics discussed include: family background, early education, his work at Bell Labs and General Electric, transistors, lasers, silicon, John Bardeen and the University of Illinois at Champaign-Urbana.
Ashrafi:Today is March 23rd. This is Babak Ashrafi in Nick Holonyak’s office in Urbana.
Holonyak:Whatever it is, it’s an office, but it’s not much of an office, (a corner at one end of a lab).
Ashrafi:And you were saying when you came here, you were working…
When Bardeen came here in 1951, I was in a group working on microwave vacuum tube problems, and all of my training had been on vacuum tubes. Actually, I had soft of a broad background. I had a certain amount of physics, certain amount of electrical engineering, and a certain amount of mathematics; it was sort of smeared out. When Bardeen came, I began to emphasize the physics part of it more, but I was actually in the Electrical Engineering Department. The physicists wanted me to transfer to physics, but I just stayed where I was because I felt I’d get behind otherwise and it was an issue of money and everything else.
Bardeen had an appointment in EE and physics, and he set up his semiconductor laboratory in the Electrical Engineering Department. Then he started the group that became the BCS Theory in his physics quarters, but he had, in essence, two places. The interesting thing about John, when you think about it, is he was a theoretician. But why would he run an experimental effort? John believed in working physics from data. In other words, he considered it sort of, not a useless exercise, but some kind of an exercise of vanity or something to try and just invent physics from your mind. John was trying to see how nature really constructed physics, and then trying to puzzle out why it did what it did. What would account for the data? He liked to have reality before him, and he liked to look for the key ideas and get rid of all the stuff that didn’t matter much.
Ashrafi:So I read some of your stuff, including all the stuff you sent me. Thank you very much.
Holonyak:I don’t know quite what Bobbie sent. The secretary is over in our other office and…
Ashrafi:She sent your list of publications, your biography, and the awards that you’ve won. I’ve also read the interview that you did with Fred Nebeker, and there’s a lot in there, but not a lot of “beef” about what you’ve actually done.
It was like this: It was a conversation. He (Nebeker) came here for some meeting, and he said that IEEE likes to have more voluminous material about people who have gotten IEEE medals. I had already received the Edison Medal, which was one of their medals. Their medals are in a different category than their awards. Medals are considered bigger. Actually since then, I got their biggest one. But they like to have some more extensive material than the records and a history account. So basically, we just had a conversation. We were in a different building then; we were in an old building which doesn’t exist anymore south of here.
Actually, I knew where everything was then because when you start off, you don’t have much stuff (papers, lab artifacts, ect.) As you start accumulating it over the years, it piles up. And I knew where everything was there. Then we moved here, then it gets scattered once. Now it gets scattered again. It’s like scattering, you know. It gets demolished, misplaced, and you don’t know where various things are. For example, the secretary that sent you that earlier material was sorting through some of my stuff in the other place and she found a folder I had of my dad’s stuff. He came from Europe and there were some old records. In his language, there were some things written that no one would be able to read but me.
Ashrafi:So he was a miner?
Holonyak:My dad was an immigrant from the Carpathian Mountains. Let me tell you a little story. Out east in the New York area, there’s a writer. Actually, I guess he’s an ichthyologist—a fish expert. But he writes for various magazines and newspapers and things like that, a guy whose name is Karas, Nicholas Karas. A guy I was in the Army with, a sergeant who’s out in New Jersey, sent me a brochure that said that this guy, this writer, Karas, was going to give a talk about a book he had written, a book called Hunky. So I looked at this account of this talk that this guy is going to give, and I read it and I thought, “What the hell is this? What’s this guy know about this? And using the word ‘hunky’. What does he know about hunky?”
Holonyak:I’ll explain it. So I get to the bookstore to get me the book and I start reading it, and I’m fascinated. I’m born in 1928; he’s born in 1931, so I realize, “Uh-oh, he’s about my age, and maybe he knows”—(Yeah, let’s try the coffee. Let’s see if it’s any good. It’s fair. It could be better.)
Ashrafi:It’s good for me.
You can drink it all day and it won’t bother you because it’s half decaf. It won’t make you vibrate. So I read this book and I think, “My god. Except for the fact that I would have different details and all that, I could be telling this story.” And so after I read it, I wrote him a letter. I write him a long letter. See, before they screwed this up (the building in construction), I had all that stuff here. And then in my folks’ language, I said to him, [speaks Slavic] “What does that stranger know about us, these people that the Americans call ‘Hunkies’?” Well, who are they? In Eastern Europe, it wasn’t Yankee Americans, it wasn’t the Anglo-Saxons that dug the coal. Just like now, there are a lot of poor immigrants who are doing a lot of the dirty work.
Well, it’s true around the whole planet. At the turn of the century, from the 1800s to the 1900s, there was a wave of immigrants from Eastern Europe who were coming in to mine coal in West Virginia and Pennsylvania and all that, and those were Eastern Europeans mainly. Some of them might have been Hungarians that Americans got confused with, and came up with the word “Hunky,” but most of them were really Slavic people from Eastern Europe. There was a group in the Carpathian Mountains who were probably the most kicked-around group of people in Europe. The Poles didn’t like them and want them; the Ukrainians didn’t like them or want them; the Czechs thought they were superior to them and didn’t like them and want them; and the Hungarians in Austro-Hungary had inserted themselves in there, and weren’t even ethnically related to them, particularly. And so these people trying to dig a living out of the mountains were grubbing a living out of the rocks and trees and who knows what.
They were shepherds and lumberjacks and barely surviving. Who knows how many came to America? I know there are maybe a half a million. Many of them wound up in steel mills and coal mines and all that, and my dad was one of them. He came in 1909. They’re a Slavic-speaking group; their language is closely related to Ukrainian, but then frequently doesn’t use Ukrainian words or uses Russian words. They’re related to both Russians and Ukrainians. If I talk to either one of them, I can get in trouble with them in a hurry because— [Milton enters; speaks to Milton about lunch plans.] Milton and I are involved with Gabriel, who is a post-doc, and Milton’s students, and we’re involved in light emitters in transistors and lasers. Just quickly to show you what happens. This goes all the way back to Bardeen, what we’re doing here.
We do some fooling around here about that, then show something about a heterojunction bipolar transistor. Milton has a group upstairs that is leading the world in transistor speed. This is just metalization on top of a bunch of semiconductor layers. You etch down in various patterns and you can contact the layers. So when you’re done, you have this big contact, which comes over on a bridge, over to an emitter contact right here. And then surrounding that is a base contact over here, and then at a lower level, wider out, is this collector contact. So what you have is basically a transistor, but it’s a transistor layered in such a way with quantum wells and reflectors that it’s operating as a transistor electrically, but also operating as a light emitter. There are reasons for that, which is actually what the subject of this account is. And if you make it in a slightly different form, this shows the sketch down underneath, contacting that underneath layer is the collector contact; up higher on a higher layer is a base contact; and up on the top is an emitter contact, which is that little region right there.
You can run this as a regular transistor with transistor characteristics, which you get up to a certain base current, and the photons that you’re generating generate enough of an electromagnetic field to drive the recombination process that’s fundamental to the transistor in stimulated emission. So it makes a fundamental change and instead of these characteristics stepping, stepping, stepping in some uniform way, they suddenly compact into sMoller steps. The device has gone from emitting spontaneous light up to here, and now it’s emitting laser light. You can see it. We see it right away on the IV characteristics, and that before this was never seen. Over here you see a corresponding spectrum, and then we see the certain current, we suddenly have a narrow spectrum, a coherent spectrum. This figure is going to appear in Applied Physics Letters here next week. Then here we’re showing that there’s a three giga hertz input signal coming in this base. Coming out this collector is a three giga hertz electrical signal. This laser signal isn’t just a laser signal; it’s modulated at three giga hertz. This is just a representation. The actual device is right here. We’re obscuring some of it with a probe here; you don’t see it. But coming out of this end is a laser beam here. And at the other end, a laser beam. And this is a fiber optic that’s picking it up. That’s how we’re seeing this signal.
Ashrafi:How much power is it?
Well, we can get into the— In fact, I don’t know exactly how much because this pc radians work is in its infancy. Down in here, you’re in the… Because it’s spontaneous radiation, it’s all out in four steradians. It’s scattered around in the crystal and some of it’s leaking out. The amount that you subtend and pick up with your measurement is rather sMoll, so it’s in the microwatts region here. Soon as you go into stimulated emission, it’s a single mode that is pointing the beam out of the Fabre-Perst facets of the structure, and now you’re into the hundreds of microwatts up towards milliwatts. So this power scale is right. From zero dBm would be probably around a milliwatt. It’s way down much lower, but when you hit “stimulated emission,” it comes up much higher out of the ends. It’s useful power. It’s power you can use.
So there’s significance to this, but the significance, we’re not there yet. We won’t be able to get there right away. But I just want to identify it for you so you know… Actually, this is something that IEEE—this was taken [referring to photograph] on the other side of the wall. The device is sitting here. Gabriel put this first one together. But Milton and I had been playing around with this idea for maybe two years, and now we’re at the stage where we actually have a transistor that’s a transistor, but it’s a laser. See, if Bardeen were alive and he came through the door, he’d have a big smile, because in some respects, the field is considered mature, done and all that.
Not really. Because the three-five compounds offer many different kinds of layering possibilities with quantum wells, with doping, with stacking arrangements, not to mention arguments about quantum dots and whether those can be laced into the structures. And then there are many combinations. We may be building one, but someone else can come along and now knowing that you can do this, can build another combination of the three-fives because on the periodic table, there are different arrangements of materials. Milton, in building these HBT transistors on several platforms, gallium arsenide being one, indium phosphide being another one, then you use different alloys to build the different kinds of layers and different dopings. So there are many possibilities. These are far from totally examined and exhausted. Not to mention, transistors have never run before as transistors and as lasers. Somewhere, a matter of about a month from now, he’s reporting in Applied Physics Letters running one of those HBTs that runs at 600 gigahertz. See, 300 gigahertz is a millimeter wave. 600 is half a millimeter, just E and M length. And those are fantastic speeds. The chip in a fast computer is, I don’t know, running two, three, four gigahertz at most. They’re clock speeds.
Ashrafi:You’re going to make optical switches?
You were looking at something that can carry information and handle electrical information and was running at three gigahertz electrically, and putting out a three gigahertz optical signal. This is potentially a natural way to be interconnecting both electrically and optically around on chips. If I were at Intel watching this, I would be really watching very closely because this has implications for a lot of things. Now, we ran into problems with equipment. We didn’t have high enough speed equipment to do this at higher than three gigahertz. With some power equipment, we have been higher. Then we ran into an equipment problem again. So we know that the goal is to try to go into the hundreds of gigahertz. His transistors go to hundreds of gigahertz. Optical signals do not go to hundreds of gigahertz. Why not?
That’s part of what the issue is, is to get at that issue and figure out how to get there and what to do to get there and what kind of ideas will work and what kind of ideas won’t work. Which is sort of between both physics and materials and electronics. When Bardeen came, he knew that in the kind of work we were doing, there was a playground for physicists, for metallurgists, for chemists, for electronics people. In other words, there were many things in there that were possible. Actually, I’ve had 60 PhD students, and I’ve lost track of how many of them were physicists whose theses were done in physics. I don’t know, at least a dozen, maybe 15, something like that. Three of my physicists are members of the Academy of Engineering and they’ve got physics PhDs. Actually, that’s sort of a mixed up world, because even though I was in the EE Department as a student, I probably in the last two years of my work had more contact with physicists than with electrical engineers.
Ashrafi:Let’s see how we got here. So your father was a miner in Southern Illinois.
He was an immigrant. When he came to America, he landed in Baltimore. My mother would tease him sometimes, because she came under better circumstances than he did. He didn’t know her in Europe. She was an orphan and had an uncle in Southern Illinois, a coal miner who had gotten a broken back in a mine, but had some money. We sent a note back to Europe and said, “If there’s a relative of mine in such and such village that wants to come to America, I’ll bring that person to America.” So she was an orphan, and she doesn’t know what became of her brother. She somehow or another lost track of him during World War I.
She thought about it and decided that, “I don’t have anybody here, and I have an uncle in America. I’m going to America.” So she came because her uncle had made it possible for her to come. She came in 1921. When my dad came in 1909 he didn’t know her. They were from the same part in those mountains, in the Carpathian Mountains, and those were the people that Americans called Hunkies. That’s why I wrote to this guy because I thought, “What the hell is he doing talking about Hunkies? What does he know about any of us? I know what it is to be called a Hunky on the streets of America. Does this guy know what this is?” Well, it turns out his account was accurate. Anyhow, my dad came in 1909. He was born in 1888 and came in 1909. How the poor guy got here, I don’t know. He didn’t have much money.
He landed in Baltimore. Came on some German skow, some old boat. He never liked to talk about it much. My mother came in fairly decent class. Landed at Ellis Island in New York, and came in fairly good conditions. But he came just in the lowest class on that German boat, landed at Baltimore, and knew that there were ethnic coal miners from their group in Pennsylvania, and he started walking to Pennsylvania. It didn’t seem far to him. And he thought Americans were crazy because you could buy bread for a nickel, and there was corn in the fields. And he says, “You can help yourself to some corn and eat the corn and buy bread.” And somehow or another, he made his way to Pennsylvania where these coal miners were, these people that he came from. So over the years, he would either dig coal in Pennsylvania. If he didn’t like the conditions in Pennsylvania, he’d come and dig coal in Illinois. And if he didn’t like what was happening in Illinois, he’d go back to Pennsylvania. But then he also dug coal everywhere else in America—in West Virginia and Ford’s Mines in Kentucky, out in Montana during the Depression when I was in grade school. So our living came from the coalmines, and it was a tough life that he lived. It was not an easy life.
Ashrafi:Was he someone that pushed you and encouraged you in school?
My dad was 40 years old by the time I was born, so he was a no-nonsense father. He was not the kind of father that you read about now where you’ve got some father that plays with the kids and plays ball and all that. This guy lived a hard life. You know how American kids are. I come home from school and…I have that in the thing with Nebeker, that this guy has dug coal all day long and he’s there, spading the ground and gardening. In Europe, they were not coal miners. They were used to a hard life, and consequently they could tolerate the conditions underground, whereas a lot of people couldn’t. They just wouldn’t be mentally tough enough to withstand that kind of a life.
In Europe, he was a shepherd, a goat herder, a lumberjack, a farmer, and he knew how to till the soil very well. So we had a big lot, and at this time of year, it’s time to start spading the ground and turning it over and all that. So he’s come from the mine and he’s turning over the ground, and I’m big enough already to help him. And he says to me, “Okay, dig.” He was fussy about how he wanted that done. I’d dig up some worms and I’d put them in a can, and I said, “I’ve got enough worms to go fishing.” And he says to me, “Koppai.” [?] That means “Dig.” We’d dig some more and I’d put some more worms in the can and we’re further along, and I thought and said, “There’s a mine pond not too far away. I’ll go fishing.” And again, he repeats himself. And then, I test him again. See, I knew that you could test—you’re probing all the time.
Kids are like that. And I probe him again and he growls at me and he says, “le holoden,” “Are you hungry?” He says, “Shto treba tobee,” “What do you do need?” And he says, “Go ahead. Go fishing. And where are you going to sleep?” he tells me in his language. In other words, what he’s telling is, “You don’t run things here. I do. It’s time to get this garden made and we’re going to make this garden and get it done.” If my sister or I was crybabying about something—we wanted this or wanted that—his reaction was, “Read your book. You have a chance to read a book and all that.” If I bring home the report card… The guy’s got a big hand. He’s a coal miner and he’s worked hard all his life and it’s like a claw. He’s not used to writing with a pencil and all that and he says, “What’s there? Read it.” So I tell him. You see, he expects good marks automatically. And then he says, “You sign it.” But he knew what was happening and he expected that, “You have an opportunity to go to school. You’re going go to school.” So I knew not to come home from school bringing trouble home to him. I knew that. See, I didn’t do that. Now my mother was different; she was probably quicker. He was deeper; she was quicker. But she was quicker to recognize American ways and stuff like that. They both were insistent that, “You’re in school. You pay attention to school.” But you see, it’s a puzzle to me when I hear these people talking about, “Well, so-and-so doesn’t have anybody to do homework and help him with this and all that.” There wasn’t anybody to help us with homework. We did that for ourselves.
In fact, neither of my parents was in a classroom, ever. My father knew how to read because his mother had taught him his Russian Orthodox—they were Russian Orthodox in religion—and his mother had taught him how to read his catechism. My mother was anti-clerical because as an orphan, she was thrown into the priest household as a servant, so she didn’t like the clergy at all. And when issues of religion would come up, she’d tell me, “You ask your dad. He knows about all that.” He knew all about the ritual and all that. So he knew how to read Cyrillic, the Russian text which is used by Eastern Europeans—it’s used by Ukrainians, Serbs, Russians—and that’s their alphabet. So he knew how to read what his mother had taught him. So then learning Latin letters and the stuff that the English-speaking world used was not that hard, so he learned that. He would write to me phonetically using Latin letters, but in his language.
So he could struggle with the reading. And my mother would do everything by memory. My sister and I would just read to her and she would shove everything in her memory and do everything from memory. Everything. In fact, she would laugh if we’d scribble things down because she could put all that into her head and just keep it there. But in his case, he could struggle through as much reading as was needed. But until my sister and I went to school, there was nobody in our family that had ever been to school.
Ashrafi:Do you think that the expectation that you’re going to do well in school was common in the families of your community?
Holonyak:It was and it wasn’t. It depended on the family. It depended on the circumstance of the family. Some families, the answer is yes.
Ashrafi:Coal mining places tend to be rough. You were saying…
Holonyak:Until I came here at the end of World War II to school here in Champaign Urbana, I didn’t know people lived this well. I didn’t know that they had circumstances this good. Because the school that I went to in Franklin County, which is deep in Southern Illinois, which was a coal mining town, was a brick school, and I’m not sure exactly what the level of education was of the teachers in that school.
Ashrafi:What’s the significance of it being a brick school?
Well, some buildings were not brick; they were wood, and cheaper. But that was one of the better buildings in town. There were a few buildings that were fairly good buildings. I haven’t gone back there for many years, but a couple years ago, I went through there and I was stunned at how little a place it looked to me and how poor it looked, yet, relative to economically much better places. Coal mining places were generally…the houses were sMoll. In many places, they were company houses that were the same thing—repeated boxes and all that. Zeigler, where I was born, wasn’t. There were different kinds of houses all over the place. But nevertheless, when I look at it now, they don’t look on the economic scale what Champaign Urbana appears to me. And they’re not. They’re living a more difficult existence.
Then in 1936, during the peak of the Depression, we moved from that county to the St. Louis area on the Illinois side of the river to another coal mining place. The schoolhouse there was only to the 8th grade, and it had the first four grades in a brick building and 5th, 6th, 7th, and 8th grade in a wooden building—two buildings side by side. Then when I went to high school, we had to ride a bus to Edwardsville, which was a bigger place. It’s the county seat of that county, and the school is a much more lavish kind of a school, it appears, to what I’ve seen up to that point. But I don’t remember in Zeigler, where I started, where I went to 1st grade, 2nd grade, and started 3rd grade before we moved, I don’t remember what the level was of the education of the teachers. In Glen Carbon where we moved during the Depression where I finished grade school, some teachers had maybe bachelor’s degrees. Some had merely certificates and were still working in summers to try to get some more schooling. But almost every one of them was a good teacher, as I remember.
Every one of them was a fairly good teacher for the basics—reading, writing, arithmetic, the core material; they were excellent. And then went I got to Edwardsville High School, then there was another level of teacher that was more educated and better yet, and the sophistication of what they could do was higher. Of course, I remember those a little bit better because of their higher standard of learning. But in my second grade school, Glenn Carbon, I remember that there were several of those teachers that were extremely good. They were extremely good, and it was all essentially classical kind of stuff. It wasn’t any of the kind of things you hear about now. Nothing frivolous. It was all grammar. So I have absolutely no complaints about the kind of teachers I had through grade school and high school. As a matter of fact, when I came here at the end of World War II, I had just missed—I had a draft card, and war ended.
Had the war not ended in the Pacific when it did, I would have been in the next wave of people. So I have a very distinct memory of World War II because I was old enough and knew enough about it to know what was happening. I wasn’t here the first year in college. I was in an extension center because the universities were all filled with returning war veterans and so forth and you couldn’t get in. So after a year in an extension center, then we could come to Urbana. A lot of the people that were these World War II guys had been in radar schools and around electronics and all that, and in class in some respects, they had an advantage over me. But I knew that I had an advantage on them in that I probably had a much better mathematics background than any of them, and I knew I could keep up with them, and that from that base, I could probably learn the next material, which is in many cases strange to them. And I learned after a bit that yes, I could compete with them. It was based on what I had in my elementary school background and high school.
Ashrafi:So your elementary school and high school wasn’t training miners.
No, no, no. Because the world had shifted again. In deep Southern Illinois, the Depression has occurred, jobs have been lost, mines have been closed. In fact, there’s been major calamity, warfare, and strikes. You see, John L. Lewis and the United Mine Workers have been at odds with the coal operators. You have to understand that in times that my dad worked, they were working frequently piecework, ridiculous hours. He didn’t come to America to stay in America. These are young guys that are strong and are willing to work. They work a shift, they go home and eat and sleep a little bit, and then come back and frequently did double shifts.
I remember when Tennessee Ernie Ford was singing the song, “Sixteen ton, and what do you get? Another day older and another day of debt.” In other words, loading up 16 ton of coal is a big deal. I can remember my dad saying, “What kind of coal miner he?” You know, in his broken English, he says, “Sixteen ton? That no much coal.” He says, “Twenty-five, thirty ton.” And of course, he’d go into his own language and say, “Real hard work.” They’re trying to dig as much coal as they can and earn as much money as they can and go back to Europe to pay off debts and get back some land that they’re renting or paying to some landlord or something. Also somewhere in those papers that the secretary found, there’s a piece of paper that shows that he went to the Hungarian consulate in Pittsburgh to turn himself in to be sent back to Europe to serve in their military because his family is left there, and if he doesn’t come back and serve in the military, his family is going to suffer consequences there in Europe.
His hand has been ripped open by a pick accident in the mines—someone near him slung their pick in cutting the coal and it hit him and ripped his hand open. The doctor, the Hungarian consulate looked and he says, “No, you’re a cripple. We don’t want any cripples in our army.” So he stayed in America instead of going back. They didn’t come necessarily to stay. They came to try to earn enough to go home to pay the debts. But then life was primitive, travel was primitive, communications were primitive. One year goes into another year and then another and then another, and before long, all contact has been lost with the past in the other place.
In fact, the Soviets, years later, invited me to come as their guest to visit and all that. And I said to him, “Do you want to come with me? Maybe we can get a visa for you.” And I said, “Well, I’m supposed to go Kiev and maybe they’ll let us go down to the mountains where you come from and have a look.” And he says, “What for? I’m an old man. I was in a lot of places and all that. You’re a young guy; go ahead and look.” And he says, “If I go there, I’ll know the language, I’ll know the place and everything else, but all the people will be different people. They’ll see me as a stranger and I’ll see them as a stranger. What for?” And he says, “This is my place now. I’ve spent my whole life here working for this.” And I had to give him credit for it. But the life was so different that, you see, in those coalfields, they’re not…there are probably still some coal miners. Not many.
They’re doing something else. And in the St. Louis area where we went, people, particularly during the war years, went to work in the steel mills in St. Louis here and there and the manufacturing places and all that. So I don’t know if there are any coalmines operating there anymore. Coal mining in Illinois still exists, but it’s not like it was at that time. The railroads used coal; coal was being burned everywhere. Now there’s a problem with cleaning up coal to try to get it to go back to coal, but to under different circumstances. So the whole picture is different in how they operate and what they do. It was primitive then; it’s not so primitive now.
Ashrafi:So your schooling was really seen as a way out.
Holonyak:In the case of my parents, they felt constantly handicapped by the fact that they lacked education. That was a total handicap to them. What the hell is the education like on the edge of a jungle? Does every kid go to school from the longhouses? GABRIEL: (Gabriel, post-doc, is from Malaysia) Not according to age. You’re standardized. Like if you’re six or so, you go to grade one and grade two, but different varieties. It’s not necessary for them to go to school until you are six or seven, basically.
Holonyak:Babak, when I was a kid in Southern Illinois in the coalfields, anyone with a couple years of schooling was considered educated. Anyone that could read was considered education. And if you could do further things, you see, if you were educated up to the level of a schoolteacher or a lawyer or the clergy or someone like that—My dad wanted to send me, during the summers, to learn Russian from the priest because he knew that the clergy was educated and that there was literature and higher learning and all that, and they were the lucky ones that had been exposed to that, and they weren’t the ones that had to till the land and do the work and all of that. They had hardly gotten out of serfdom, hardly gotten out of essentially a certain kind of a slavery in some respects. And education, my God, that’s unheard of. So in their minds, school was everything. I knew that you didn’t come home and essentially start some kind of campaign to complain about the school and all that. That would never work. That’s not going to be acceptable. That’s not going to happen.
Ashrafi:How about your classmates? Did a lot of them go on to college?
Some did. But the problem with a place like that is the connection. I notice now that schools were crude and there are almost no pipelines that are working to move children in the direction of school and all that. That didn’t really exist, and I didn’t see anything like that. There was a young guy in high school with me, and we’re getting near the end of high school, and he’s going to join the Navy. And he’s trying to talk me into joining the Navy with him. I guess I’m too young to join without my parents’ signature.
If my dad signs, I can join, but if not… So I’m badgering him about joining the Navy with this other guy, and in his language he says, “Okay. If you insist, I’ll sign it.” But he says, “I want you to know that I left Europe because I did not want to serve in the other guy’s military, in his army.” These were Slavic people. They weren’t something else; they called themselves Zakarpatski rus. That means “Trans-Carpatho Russian.” They were some ethnic group that went— I would badger him about that. I’d say, “Who are you people?” And finally sometimes he’d bristle and that’s when he would say, “Don’t you have anything to eat? Don’t you have a roof over your head?” And then his answer would be, “Muk,” that’s “us”. He says, “Muk bile tarn davno z davno..” “We were there a long time and a long time,” meaning, “We go back to ancient times.”
In other words, we are not recent settlers in that part of Europe in those mountains. We have been there and been there and been there indefinitely. So I knew what he was saying was, “You’re not going to get an answer to that question. That’s a bigger question.” They were there. In other word, their origins go way back, and they were basically some Slavic group that’s closely related to the Eastern Slavs—the Russians and Ukrainians—but have evolved the language that’s still that kind of a language, but still modified with some contaminating words from the Poles, from the Germans, from the Hungarians.
So sometimes when I talk with a Russian, I have to be careful because I may throw a string of words at him that he’ll understand and then suddenly, the next word is Hungarian, and he doesn’t understand it. A word that has been brought in, put into their grammar, into their language, and now has been incorporated. But at any rate, they are a distinct group. He knew they were a distinct group, but not accepted as distinct; never able to put together their own country. It’s just like, I’m sure in the Middle East— I had a young woman from Lebanon, Nada El-Zein, and Nada drew distinctions between various groups in the Middle East, that yes, she speaks Arabic, but her Arabic might be somewhat different than some other group.
In other words, a Syrian and a Lebanese and so forth, all for one reason or another ethnically have their differences. And so too my folks, relative to all the others. In other words, they didn’t consider themselves Poles, and Poles wouldn’t give them their due. They would then perhaps consider themselves Czechs, and Czechs wouldn’t give them their due, either. Ukrainians, likewise. The Ukrainians now hold that area. And similarly the Russians. Everybody looks upon the sMoll and the weak as somebody that you could step on. And you see, when they came to America without education, the power is in the hands of the educated person and not in the hands of the uneducated, maybe illiterate person. So my folks were very aware of that. In Southern Illinois, there were many ethnic groups. They were all immigrants at one time or another, but there were some longer-range Americans and so forth. Germans probably were some of the quicker ones to recognize how fertile the land was in Illinois, and there are a lot of German immigrants that got into parts of Illinois ahead of other groups.
So there were plenty of ethnic Germans where I grew up. There were plenty of ethnic Czechs. There were some various Americans. And it didn’t seem to me that they were quicker to go to school. In other words, I think it was a matter of your family, how your family felt about school. If your family felt strongly about school, then probably they encouraged you. And whether they could help you or not didn’t matter. It was a matter of what they believed in. That worked better than anything else. So I can remember some classmates going on and a lot not going on, for one reason or another. When he was going to sign those papers for me and he said, “Remember, I left Europe because I didn’t want to serve any other guy’s army.” See, Austro-Hungary is really ruling his part of the world then, and he has to serve in their military. His reaction is, “Why should I serve in this other guy’s military? I’m just his slave. He’s not taking care of my part of the world; he’s using my part of the world. Forget it.” And so I thought about it and decided, “No, I think the old man is right. I better go to school.”
So I had lied about my age during World War II. All or some of the kids are older than I am, and they’re able to get jobs in some of the war industry plants. Not far away towards the Mississippi in the direction of St. Louis, there were some plants that worked where you could get a fairly good job. At least, they looked good to us. And I tried to get a job there, and they checked up too carefully on my age. I was only 15. You had to be 16; I’m only 15. But the local railroad didn’t check that carefully, so I lied and told them I was 16. And so I’m working out on a track game, on the Illinois Central Railroad at 15.
Ashrafi:In the summers?
Summers and school holidays. The war is intensive enough and so work is going on school holidays and weekends. And so it was almost continuous, in some respects. So 1944, ’45, ’46, those three summers, I’m working full time on the railroad, ten hours a day, six days a week. And that’s all dirty, heavy work. As a matter of fact, my earnings from railroad work was what brought me to school. What happened that last summer was, there was a creek that wrapped around our part of the railroad and it went under a trestle and then would loop around and then down the line somewhere again, it would loop around another trestle and all that.
And there were some heavy rains. And the heavy rain washed out one of those trestle. In the wee hours of the morning, it’s still raining; someone’s banging on the door; and it’s some of the railroad people: “Come and work. We’ve got a big washout and all the traffic has stopped. Nothing can move on the Illinois Central Railroad from St. Louis to Chicago. There’s no way to move anything because we’ve got track that’s hanging in the air. Everything washed out from underneath it. We’ve got to get that repaired and get those trains moving.” So I go out there, and I know it’s wet, and mosquitoes. If you didn’t have a family to bring you food, you’d just work and go hungry. I remember we worked 33 hours straight—no interruption, just continued to—They’d run in a railroad car of rock and dump it, and we have to move it and get it underneath there, at the track, and fill and fill and move all this stuff, and jack up the track, and try to fill it in to make sure that we’ve got enough underneath the track to be able to move a train. Thirty-three hours straight.
When I finally get home, I’m dead tired. I’m thinking about this and thinking, “That’s no damn way to live. That’s no way to survive.” And I had heard that the University of Illinois was going to open up an extension center in a local high school. In other words, the school here was so packed with returning people from World War II, they couldn’t accommodate everyone.
Ashrafi:This is in ’44 now?
This is in ’46. See, the war ended in ’45 and we’re swinging into ’46. By the time all the soldiers, sailors, and airmen are returning home, and we’re into ’46, and the schools are getting organized, a lot of the World War II people are—The GI Bill has been created. The person who’s been in war realizes that his life has been interrupted; the only way he can get back into anything proper and productive is to go to school. And so the colleges have been cranked up. But this place, for example, couldn’t hold everybody. The gymnasium was converted into a dormitory and underneath and stadium and all over the place and a cafeteria is put up, and all sorts of accommodations were made to try to field the bigger class of people, students and so forth.
So I heard that the University was opening up an extension center in an adjacent place called Granite City. I took my dad’s car, went down there to find out about that, and I decided school seems to me like the right answer. See, I’ve been in school all my life, from kindergarten through grade school, through high school, through now, and it just seemed to me, it’s unnatural at this stage to say, “School is over.” And it’s just going to be working. Ridiculous circumstances like this, where you’re just, “The work has to be done. You do it.” So I checked up on that and found that, “Gee, I can enroll there.” So I enrolled in that school with returning World War II people. So I’m one of the few people that’s coming out of high school when all the returning GIs are coming and flooding into universities. So I come in with a high school background, and they’re coming in from a World War II background. And so I’m part of that group of people that came through the University at that time. And after that first year, then you know what the circumstances to go from that place to the full campus.
By the time I’m a senior and getting a bachelor’s degree, I realize, “This is just the beginning of it. This isn’t the end of it all; this is the beginning of it all.” See, the first problem in coming onto a campus like this from where I started is, to get calibrated to realize, “Are you in this game or not?” I remember a guy by the name of Butler in the physics sections. The physics building is named after Wheeler Loomis. Well, Loomis was still lecturing. Some of the lecture demonstrations, I could tell you of some of the ones that Loomis did right in front of us. I got to know him later, but then I was student. In one of the quiz and lecture sections, there was this instructor, and I remember the place was packed. So the exams were given not during regular class periods, but they were given evenings frequently in special places. And I did the exam, came back to the quiz section, whenever it was, and they’re handing them back. And he’s handing them out, and he says to me, “What happened to you?” He’s holding this. And I’m thinking, “My God. Did I get wiped out?” And he handed me back my paper and I looked at it and I realized, “What does he mean what happened to me?” I had one of the highest scores, you see. And I think what he meant was a lot of people are going to get wiped out. He’s pointing at me as one of the people getting wiped out, but it’s the opposite.
Ashrafi:He thought you were going to get wiped out.
Holonyak:Yes. You see, you had to come up to certain standards. Because the classes are crowded and they’re not going to give any quarter—no one’s going to hold your hand to get you through. You’re either going to have to perform, and you if you don’t make it, you’re out. Because there’s still people waiting to get in, and so if you don’t maintain a certain score, certain record, up to a certain point… I remember a certain number of people in certain classes that just didn’t make it. I remember one fellow that was apparently very good through high school, and he winds up here and what we’re doing with electronics and physics and mathematics and finding it’s tough, and tough competing with these returning guys who are very serious students. They’re not your usual play-around students. These guys are really focused. And I remember this guy at some point said, “I can’t take this.” He says, “My score’s dropped. I’m not used to this,” and all that. And so he shifted into civil engineering, which he considered easier, and he found for him, was easier. Those of us who were getting the full dose of physics, electronics, and mathematics, and he says, “Couldn’t handle that.”
Ashrafi:What kind of math and science did you get in high school?
Holonyak:Oh, it was just the regular: college algebra, trigonometry. They hadn’t yet introduced calculus in the high school, but all the way up to that edge, we had the full thing. One thing that I never lacked any semester was mathematics. Mathematics was something that just seemed like it was natural.
Ashrafi:You mean in high school?
Holonyak:In high school. Our high school was very good with mathematics.
Ashrafi:How about science?
Holonyak:Science, too. For some reason, I didn’t like the physics teacher and I didn’t do physics. Or was it chemistry? One of them I didn’t do and one of them I did do. Oh, I know what it was. I didn’t do chemistry, but I did physics and biology. Our biology teacher was really good. There was something called general science, too, but I don’t remember much about that.
Ashrafi:You were telling me about the courses that you did take and didn’t take in high school.
Yeah. The high school was very good. I can really remember a lady, Ms. Helm, who taught us trigonometry. She was very good about insisting that we work hard in that class. And then another guy, a tall, thin guy, that taught us college algebra. I know those two teachers were extremely good on both the algebra and trigonometry. And somewhere along the way, it seems to me like there was someone that was very good with geometry. So we had very good background in all of that. Then in the science, there was a lady that was extremely good with biological sciences and so forth.
Since I had grown up in a sMoll town and was around fields and animals and stuff like that, and my mother, since she was a girl in the old country, raised chickens. And so we had animals around the place and it was just natural to pick up biological things. That biology class worked very well for me. Then this fellow that was teaching physics was a strict disciplinarian, and he knew a certain amount, but there was something that was puzzling me back then. He did the thing about the Magoleburg Menispheres fears. There’s a famous picture in physics and the physics texts, in the old ones.
You had two hemispheres, and you put them together on the great circle that separates them. You put them together and seal them and there’s a valve on there so you can pump a vacuum in there. And then it shows a hook on either end, and a horse on this side and a horse on that side; you can’t pull it apart. I couldn’t understand how he could get… All he was doing was using the formula that says, “Or2 the area times the pounds per square inch of the air pressure.” And what puzzled me was it wasn’t quite obvious to me that that’s the answer. And then later, when I’m in college and I see calculus and vectors and all that, I realize right away, what’s at issue. “The components of a vector this way and that way are opposing; it doesn’t mean anything and it’s just Or2 times…” It was clear to me later that he really didn’t know the right answer. In other words, you finally get to the stage where you begin to be able to see into the instructor and whether the instructor really knows something.
The schooling was good enough that I could tell already when the answer was good and when the answer was not so good, and that more learning would straighten that out. And so I was prepared when I came here to use what I had learned. What I found out in this guy’s class… See, there was some more of these night exams and all that, and I remember I left one answer, “Something O over something.” I don’t know why. Whoever graded that problem took off a little bit. And by then, the instructor knew me and boy, he was incensed. See, I was getting better scores than most of these people and he was incensed because he could use that to account for the fact that he was doing a good job, that he had some students that were at the top. So it was funny. From starting out where at first, he’s doubting that the person sitting there should have a good score; by the end of that class, he expects me to have a good score.
Ashrafi:Was that Loomis?
No, it wasn’t Loomis. The thing was broken up in a lab section, lecture section, and quiz section, and this was the guy in the quiz section. Loomis was lecturing to a big group of us. Oh yeah, I can remember his lectures very well. One day, he had a steel beam from here to there, and he’s showing that everything is elastic, and there’s some sort of arrangement of levers with light beams that would give you big magnification and he could show that…just put his thumb down on it, and you realize that you don’t think of it in those terms, but yes, even though it’s a sMoll displacement, there’s… And I remember him doing that. I remember him doing other things.
I remember him charging his assistant up and the guy had long hair and his hair was sticking straight out. [chuckles] I remember, also, one day he came in and he had an iron sphere. An iron sphere that was hollow on the inside, but a thick wall. And he filled it with water and put the plug in and put it in his tub, which was ice and some salt and so forth to get it to lower the temperature. He’s lecturing and lecturing and suddenly, there’s a thump. And he walks over to the tub and picks up the sphere and its split open by the force of freezing of that water inside that sphere—broke it. Yes, I can remember Loomis’s demonstrations.
Ashrafi:Now that’s dramatic.
Holonyak:Yeah. I’ll never forget that. It was so well done. Also, I’ll never forget him shooting the dart from way over here somewhere at the monkey that’s dropping on a guide, and of course, the dart is dropping the same as the monkey, and so, even though it was aimed up there, the dart hits the target over here. Sure, I remember some nice things that Loomis showed us that were very dramatic and it’d get the idea across right away. Physics really worked for me. Then I continued. Physicists wanted me to transfer, at that point, to physics, but I thought I would run out of money. See, I’m going to school on my own money, so I didn’t want to do that because I didn’t want to run out of money. Then by the time I became a grad student and I’m supporting myself with my assistance wages, then Bardeen comes soon after that, and I don’t have to switch; I just sort of take more courses that are in physics and less courses in other areas.
Ashrafi:Before we stop with college, did your sister go on to college? You had one sibling?
Yes, I had one sister. My sister went to nursing school in St. Louis. There she met a young guy graduating with his MD and they married, and she went to Boston. She didn’t go beyond being a nurse. But my sister was capable in school, too. Both of us had, essentially, straight-A records through the school, through elementary school and high school. And essentially the same thing in the university, except occasionally, you run into something that for one reason or another— Well, for example, when Bardeen taught the first semiconductor course that I’m aware of that might have been taught anywhere was here in January 1952.
That was a hard course for me because it was listed in both electrical engineering and in physics, and I’m the only one taking it from electrical engineering, and I hadn’t yet studied quantum mechanics. And here I am, trying to get into the middle of semiconductors before I’ve even filled up my background with the right material. So that course was a difficult course for me. But obviously it changed everything to struggle with that. There were no books then. The only book was Shockley’s book—Bardeen’s notes and Shockley’s book. So I know that I must’ve focused very hard on that because there’s some parts of Shockley’s book that I remember to this day because of how many times I must have read and re-read parts of it.
Ashrafi:Why physics and electrical engineering as opposed to medicine or biology or accounting? So why science as opposed to accounting, and then why physics?
Holonyak:There are some reasons for that. For example, I ran into a guy early in college, an instructor, very good guy, and it was in a chemistry course. I’m doing well in his course and all that, and he wants me to switch to chemistry. I considered chemistry too cookbookish and not sufficiently connected to mathematics and so forth, whereas electronics and physics were more heavily connected to mathematics, which was more penetrating in terms of being able to look into things and understand things. It just seemed to me like chemistry wasn’t quite the right thing. Medicine is a different issue. I think you probably have to have some contact or insight or something to lead you, to make it apparent to you, to see it, before you can appreciate it as something that involves you.
Ashrafi:What’s “it” now?
Medicine. I didn’t have any contact or association with something that would be medical, whereas with the kind of things that are in the physical world and electronics world, I did. For example, the cars back, when I was a kid were really old and not many. In a part of Europe where my parents came from, life was so difficult. They would have a godparent from one family, a godfather and a godmother from another family because that gave you two godfathers and two godmothers. And if your family died, one of those families would take you.
It was that kind of a world that they came out of. In Zeigler, where I was born, was my godmother and her husband, who I considered my real godfather. In another town not too far away was my godfather and his wife, who were a little bit more remote, as far as I was concerned. But my godfather in Zeigler was a very handy guy. He could make anything, and he was always working on his Model T Ford. I’m there with him frequently. This is the Depression, the jobs have vanished, and my dad is looking for a coal mining job in West Virginia, and later found something in Montana, but I’m there with my godfather.
My mother would send me with our clippers and say, “Have you godfather cut your hair, but don’t let him use our clippers on those Montenegros.” Montenegros in Yugoslavia are tall and they have dark, wiry hair, and my mother would say in her language, “Your godfather shouldn’t used these clippers on these other guys.” And I would tell him, “I’m so big and these are coal miners.” And I’d tell my godfather, partly in English and partly in their language, “Don’t use our clippers on the chorme hortse [?]. Those are Montenegros. And the guy’s swearing at me because I’m little and I’m already being derogatory to this coal miner. But my godfather was a handy guy and he could do anything, and he’s always working on this Model T Ford, and there’s spark coils in there and they’re humming and buzzing and making sparks and he’s trying to get the engine to fire right. And I’m helping here.
I can remember he’s taking something apart underneath and I’m there with him, and something dropped in his eye and he says, “Noh (Slavic noh), you’ve got nice little fingers. You take that out for me.” So I reach in there and do this and help him with everything, and he’s showing me how to sharpen my knife and how to do everything. See, this is all physical world—making things out of wood, out of rubber, out of iron, out of whatever. He’s very good at this, and I’m picking this up. And then also, the other things that happened. These were poor people; they didn’t hire a carpenter or hire anybody. They did that all for themselves. So hammer and nails and all the tools for repairing things, you did that yourself. So I see all of this as a kid. I see repairing shoes and building things and hammering things and sawing things and how you sharpen a saw that’s dull. So I’m not seeing medical things; I’m seeing physical things. I’m seeing things put together and taken apart and repaired and all that because that’s the kind of world we were in.
That’s a physical world, and physical things made more sense to me. Then in my reading, you see, when I’d be bothering my dad about this or this or this, he’d— Our school was very good about sending us home with books to read. So the books I’m finding to read are books that have something about science and that kind of thing in it, so that’s leading me, too, because I’m seeing and getting an account of that world—the world of science and essentially physical science.
Ashrafi:These were assigned, or you selected them?
Both. I don’t think they were assigned. I think they were Tom Swift books about magic cars or electric cars or stuff like that. Incidentally, we were good about making all of our own toys and things. If you’re doing all of that, the physical world is much more sensible world. Also, I never remember buying a marble in my life, but we shot marbles. We’d play with marbles and we’d play for keeps. The winner takes them all. Then one day, my mom sent me from one side of town to the other side of town where there was a farmer where we bought milk. And I’m not coming home and I’m not coming, and she’s wondering what’s happened to me.
We had a bucket that I carried milk in. When I got to that other end of town, I found these kids shooting marbles, and realized that they’re not very good. I always had some marbles in my pocket, so I get in the game with them, and before long, I’ve got my pockets full of their marbles. By the time I come home, I got marbles you wouldn’t believe. Look, shooting marbles, for example, someone told me later that Bardeen was quite a billiards player up in Wisconsin when he was student and all that. Well, these are spheres that you’re knocking around. You’re lofting them into the air and dropping them, and so you’re making parabolas and shooting—this is a game of mechanics, when you think about it. Inevitably the things you’re doing as a kid are making you comfortable with the next thing you might be getting into. So put mathematics on top of stuff like that, and it just seems like it automatically gravitates toward something that’s in a world of physical science.
Ashrafi:So it wasn’t a real question for you when you came to…
Holonyak:No, no, no.
Ashrafi:Why not physics or math?
Holonyak:The real question for me was the business of shifting from… I’m far enough along and I’m doing well in all the physics things, which were in many ways much more logical than sometimes the things you run into in engineering. Because you can run into an engineer, for example, where his mind is already, I don’t want to say “corrupted”, but actually corrupted to see economic consequences of something, and to be concerned more with that than the technical aspect of his own subject. In other words, if an electrical engineer is teaching me about some aspect about amortizing some equipment in a power plant, what the hell do I care about that? When I’m in a physics class, we’re not talking about anything like that; we’re talking about some stuff that’s germane to what goes on in the world of physics. So physics, in some respects, was more logical to me than what was going on in some of the electronics courses. But I was so far along, and by the time Bardeen came, I could essentially splice in more physics, then it was not necessary for me to change anymore. I’m just doing the physics that I need, just automatically.
Ashrafi:Where did you learn that attitude you just expressed about the purity of physics and the corruption of engineering?
Holonyak:I felt, even when I was a student, that I’m paying for the schooling, and I don’t want the professor telling me, “You need this because this is what General Electric expects of you. When you’re in General Electric. You will have to be able to handle their business.” See, in the world we’re in today, they’ll tell a student, “Well you really ought to go and expose yourself to certain kinds of business things,” and all that. My reaction when I was student was, leave me alone with that. I’m trying to learn the core material, and you’re telling me about what this employer wants from me because of his needs. But my needs are to learn this material and the foundation material and the logic of what holds it together. I don’t care to know about amortizing General Electric’s generator at somebody’s power plant.
Ashrafi:How did you come to this view? Was it your father, your godfather? Is it something you developed yourself?
Holonyak:I probably picked it up myself just in the process of the learning, that the thing that was most consistent that I was learning was the mathematics. In fact, I took enough math as a grad student that I could have, if I had declared myself as a math student, gotten a Master’s degree with just the amount of math that I had taken. Do you know a Geoffrey Chew on the West Coast?
I had two courses from Chew, and in fact, he was on my doctoral committee. I don’t know if he remembers me, but I remember him. In fact, it was sort of interesting because—See, he came here because of problems in Calif.; I had one course from him in quantum mechanics, and I had another course from him in mathematical methods. It was evident to me that Bardeen had a good opinion of him, and at the time, I didn’t know why. I couldn’t tell why. I found out later, when I did more studying and reading and all that, what Jeff Chu was all about. He didn’t sign a loyalty oath at Berkeley when California had instituted that, and then later, that was rescinded. He came here.
What was funny was he’d be lecturing to us—he was tall and wiry—and he’d flip the chalk and snatch it out of the air, or he’d pitch the eraser and grab it with his other hand. It was obvious that he had good reflexes. I found out from Bardeen, eventually, that he was a good enough ballplayer that he could’ve played semi-pro or pro ball, he was that good. His back would sometimes be turned because he was looking at the board, and I’d see his hair, and I’d think that his wife had been cutting his hair, or something, and it was not done well.
And I’m thinking to myself, “Jeff, you ought to let me cut it,” because you didn’t go get a haircut from a barber in the coal fields. That’s wasting money because you’d give 25 cents to the barber; 25 cents will buy a bucket of beer, and coal miners are going to buy beer before they get a haircut. My dad once cut my hair. He just puts that big hand up there and just moves my head where he wants it. If the clipper cut off the hair, fine; if it didn’t, I’d holler. My mother was saying, “Hey, you know what you’re doing?” And the (in size) old man would say, “Mõtch-ka.” That means “shut up.” And so when I get to a certain point, he hands me the clippers and says, “Cut hair.” And so I learned how to cut his hair just because that was done. I’m looking at Jeff Chew and thinking, “Hell, I could cut your hair better than your wife is cutting it,” but I didn’t tell him, I didn’t tell him that at all. But at any rate, I can remember sitting there, and he’s doing something in complex variables. And I’m sitting there thinking, “Hell, that was a sloppy way to do that.” I didn’t tell him, but it was a sloppy way of doing it.
I already had enough rigorous math that I could do that in a fancy way with some mathematical rigor. Now, what was funny as I think about it, and I think, “Boy, could he do some clever things to make a Greens function solution for a problem that I didn’t know anything about,” and I realized, “Oh no, he knows something that I don’t know,” and that he really was a clever, ingenious guy. But this was complex variables, and you know the Reggie stuff is going around in a plane in all that analysis, and I’m sitting there thinking, “I can do that complex variable problem better than he’s doing it because I know a rigorous way to do that.” But over the years, it’s funny. It must have been the exposure to a lot of hands-on stuff in my life; that was drawing me. My suspicion is that if I had pushed in an analytical direction, I would have wound up doing something more theoretical. But because of my background with hands-on stuff, putting things together and devising things and seeing ways to do that came very easy for me. So I just gravitated towards making and building and knew that from childhood, you make and build what you need and want. You conceive it. You see it; you understand it. You see a way to do it, and you do it. You find a way to carry out the experiment, the idea, to render it.
And Bardeen was a wonderful guy to be with for this reason; that even though he was theoretician, John was a guy that looked at the facts and the data. When he and Walter were doing the transistor, Bardeen is copying down the data while Walter is twiddling the knobs and moving the wires and all this kind of stuff. But John is actually looking at the hard facts and all that. In other words, he relied on the fact that the world is a real world and not just a world of symbols and ideas. That fit very well with my ability to do things in a lab. I had already had lab experience by the time he showed up, and so I knew how to do a lot of things in a lab that were going to be required to get going, to be able to work on the semiconductor, this peculiar new thing that nobody had seen and people don’t know much about and all that kind of stuff. It was totally in its infancy when Bardeen came. My God, since then, it’s been a major revolution.
Ashrafi:So when you arrived here, you took calculus. This is ’46 now, ’47.
Holonyak:I came here as a sophomore in ’47. The first year I was a freshman in the M of I Granite City extension center. I remember we did the chemistry in the extension center. As a sophomore I was in physics right away. And I don’t remember what we were doing with mathematics in the extension.
Ashrafi:You took math all the way through?
Holonyak:Yeah, all the way through.
Ashrafi:No quantum mechanics as an undergraduate?
Holonyak:Not as an undergraduate. What did I take in physics? I was mainly taking extra math instead of extra physics as an undergraduate. What did I take in physics? I took some courses in physics. When Bardeen came in the fall of ’51, a course I took then was an atomic physics course that he taught, and that was my first contact with him. I remember sitting in on a number of courses in physics. John Blatt was lecturing on E&M and I remember being in there, and I listened all through some lectures of Jim Snyder on electromagnetism, but I really liked Blatt’s E&M better. I took these two courses from Geoffrey Chew, one of them being quantum mechanics and the other one…But those were already grad courses. It wasn’t the two-semester sequence; it was the one-semester sequence that Jeff taught, and mathematical methods. And then there was a guy here from whom we could take special courses that were sort of self-taught reading courses that he guided. Some of that was in thermodynamics and statistical mechanics.
Ashrafi:How about EE courses?
Holonyak:EE courses, I took the usual ones that were needed and required—an E&M course, a certain more advanced circuits-system kind of course. In those years, the business of digital stuff didn’t exist yet. That didn’t exist.
Ashrafi:Did you take any electives?
Holonyak:I took a lot of math electives.
Ashrafi:Right. How about physics?
Holonyak:I took a real variables course; a complex variables course in analysis; a Laplace transforms course; a vector and a tensor analysis course from a well-known guy from the West Coast, Harry Lass. And one other one that I can think of. I had enough advanced math that I could have gotten a Master’s degree in math. I was really relying on the math, to a large extent, for anything and everything from high school on through undergrad work. Plus, even as an undergrad, the usual bachelor’s person in electronics had—In those years, calculus didn’t occur to you till college, and they still had a requirement for trigonometry, and a lot of these people came here taking trig. I didn’t. At some point, they called me in and said I have a deficiency; I don’t have trig. “That’s okay, I’ll take a proficiency on it.” “No, no, you can’t do that. You have too many courses in advanced mathematics.” And I said, “Then you’re admitting I know trigonometry.” So they swapped some substitution. I took a whole bunch of courses. The typical bachelor’s person took only calculus, differential equations, and that was the end of that. But at that point, I took advanced calculus, linear transformations and in matrices. What else? Something else I had.
Yes, I had complex variables. I had several doses of complex variables, both at lower level and at higher level. In fact, some complex variables, I’ll probably never forget till the day I die simply because some of those ideas are very potent. The thing that was probably most consistent in my background all the way through was mathematics. Then probably at some point, I began to work on physical electronics. In other words, all the things in and about what the electron does either in the free electron in things like vacuum tubes versus…not receiving tubes, triodes, and such, but in the microwave sense. Setting an electron down a slow-wave structure and getting the interaction of an electron in motion with an electromagnetic field and getting bunched electrons and things like that. Now, once you get into the semiconductor, you’re not back dealing with just the electron again, but now you’ve another entity, the hole, which is totally equal to the electron.
As a matter of fact, I’m talking about a time when…I remember being in this professor’s class and he’s talking about some advanced stuff in the circuit world. He’s laughing and saying, “The physicists have got this crazy concept of something like a positive in electron.” And he dismissed it as though it didn’t make any sense. I’m laughing at some later point, thinking, “Yeah, that is as real, that hole in the substance, in the semiconductor is absolutely as real as the electron and does all these magical things.” I laugh and I think about it. People now, you don’t have any problems getting them to believe there is such a thing as a hole. But then, the usual person in and around electronics has heard about electrons coming off of the cathode and being collected at an anode and cathode ray tubes and on and on and on. An electron is real to that person, but not something that’s going on in the semiconductor called a hole. That doesn’t make any sense. So there’s a period of time when the semiconductor world was showing the other people how real the hole was and doing experiments showing it. You could launch in an n-type material, inject some holes, drift them down a bar, collect them over here and see how all this happens.
Ashrafi:You said at some point, you started working on physical electronics. At what point?
Holonyak:No, I started getting into physical electronics as a graduate student. I had an assistantship right away when I got a bachelor’s degree. A professor I knew wanted me to get involved in antenna work, in E&M, and in and around what antenna people were doing.
Ashrafi:Do you remember who?
Holonyak:Yeah, a guy by the name of Cleve Nash. Nash wanted me to join their project, and I thought, “No, I’m not so sure I want to do that. I’m more interested in electron devices.” For one semester or two semesters, I was a teaching assistant, and then the project that dealt with electron devices in vacuum tubes, microwaves tubes, had a place for me and I went into that project. I was already into that and working in that lab when Bardeen came.
Ashrafi:Now whose lab was that?
That was Heinz von Forester’s laboratory. There’s a book that just came out on Norbert Wiener. If you look in the index of this book, you find in there von Forester. He was one of the people that came from an operation at the end of World War II where the American military and others, the Russians, were grabbing up scientists and hauling them in the Soviet Union, and we were bringing German scientists to America; I think Heinz was one of them. That operation was called “Paperclip.” I think Heinz was one of the Paperclip guys. Heinz nominally was in charge of that activity. Actually, he was loosely in charge. I wound up working with more senior students on microwave tube problems. If we accelerate an electron in an accelerator, we’re probably going to use some high-powered klystrons or whatever and run it through some dynodes and some high-field regions to accelerate that particle now.
Suppose what happens is, on the edge of this dynode, one of my electrons bangs against the plate and releases some secondaries. In the meantime, the field has reversed, and the electron is accelerated back towards the other dynode. Now, if the electrons and secondaries begin to bounce back and forth in phase with how the field is changing, you can develop quite an electron cloud of…it’s called the multipactor effect. I was working on a multipactor to see if we could get a bunched beam of electrons. Actually, there was something that we did there that I could probably still publish showing that yes, indeed, from this cold cathode, there’s a bunch of electrons bouncing in this microwave cavity between two electrodes at some proper phase point. You could get a bunched beam out and then analyze it and show it and all that kind of stuff. But anyhow, that’s what I was doing in that project in Heinz’s place.
That’s when Bardeen showed up, and laboratory in the same building. I’m taking a course from Bardeen. Heinz wasn’t very good in teaching courses. He didn’t teach very much. Bardeen comes and he’s teaching, so I immediately take a course from Bardeen and I like what I’m learning from John. The legend hadn’t started yet that he’s Silent John and Whispering John and difficult to learn from. I found him fascinating to listen to because I’d like to see why he does this instead of that. What leads him to go in this direction instead of some other direction? Why is he deciding to do that? So to me it becomes fascinating to see how John is thinking and where he’s going. The next course that he’s going to teach is the one on semiconductors, and even though I don’t have the background I should have, I’m right into that. And so John, I hear, is going to open up a laboratory in the same building I’m in on semiconductors, and I’m thinking to myself, “That makes more sense to me in many respects than what’s going on in this Heinz’sgroup and what’s it’s doing.”
So I went to see Bardeen, and Bardeen’s looking for a guy that’s got some lab background to help. Heinz von Forester has got so many people that he doesn’t know what the hell to do with them, and so he’s totally agreeable. “Sure! That looks interesting,” and so forth. So there was no problem for me to move from Heinz von Forester’s group to a new group that Bardeen’s starting. I’m the first student in Bardeen’s semiconductor laboratory. He had two post-docs and one student, and another professor had a student. Later that professor was sort of a dropout, and that student became Bardeen’s student, too. So there were two of us. In fact, there was a metamorphosis of how we went… Did you see Lillian’s book here?
Holonyak:Okay. Somewhere in here, she has a table of Bardeen’s students. Here. I don’t think this is right because John came in ’51. Now, this might have been someone who was doing something and needed someone to sign the thesis and so forth. But I’m the first student; Tom Morgan wound up with IBM later and he probably was the second one; and I don’t know who this guy was. This was the other student in the lab.
Ashrafi:So you’re looking on pg.185 of Hoddeson’s book and you’re saying the first entry of the table might not be right.
Holonyak:It may not be right because John came in the fall of 1951, and it’s very unlikely that… This might be a visitor; it might have been someone that John was guiding when he was still at Bell Labs and merely had to sign the papers or something, but I don’t remember this person at all.
Ashrafi:Who, it says here, got a PhD in ’52.
Ashrafi:But the other student was R.C. …
Holonyak:Sirrine. He was an older guy who had been through World War II, and, well, I could tell you a lot of stories about Richard. Of course, Bardeen had brought into the lab Schrieffer and some other physics students a year later (1953). It’s years later, and the Japanese wrote something up about how I wound up doing one kind of problem with Bardeen and Schrieffer wound up doing another kind of problem and why, and how John did that. A Japanese guy I know at Texas Instruments translated it for me, and then he said to me, “Is that story correct about how you wound up as Bardeen’s student, and how Schrieffer, and how John assigned problems?” I said, “It doesn’t seem quite right because it seems to me that when Bardeen brought…” We started the lab in ’52, and Bardeen brought Schrieffer in the next year, ’53,with some other young physicists. This guy, this, guy.
Milton, Valenta, Willie Portnoy, and Couvert. I seem to remember, too. And I said, “Bardeen brought Schrieffer in…” [looking through pages] Oh here, Steve Arnold. A year later, he brought Schrieffer in, Mike Miller, Milton Valenta, Steve Arnold, Willie Portnoy. Those people came in about a year later. So when I see Schrieffer come in, he’s working on a project that Bardeen’s got going there with the post-docs. He did such a nice job on it, I assume that Bardeen thought that this guy is doing such a nice job on this, that I (Bardeen) can take him in the direction of superconductivity. Well, I write back to my friend at Texas Instruments and I guess I must’ve sent a copy to Bardeen. See, I’m over in this old building we were in, and he’s down in Loomis Lab. Generally, John, if he wanted something quick, he’d call me. If he wanted to talk, he’d walk in to talk. But sometimes he would write me a letter, and this time, he wrote me a letter explaining, no, the Japanese had it right.
That actually, he had recruited Schrieffer from MIT because he had the vision of working on the superconductivity problem from the very beginning when he was leaving Bell Labs and coming here. In getting that problem going, he saw Schrieffer had a good background and was a person that could help him on that problem. He merely brought him in to our laboratory because he felt that even theoreticians should have background with what goes on in a laboratory and what the problem is in getting data and getting results of that form; he really didn’t intend for Schrieffer to stay in our lab and continue with the semiconductor problem. He really did intend him to work on superconductivity, as it happened. You see, I had assumed that since he had worked in our lab and was doing such a nice job, John moved him from the semiconductor to the superconductor project. But that’s not what actually happened.
He had brought him here to do superconductivity in the first place and exposed him to the laboratory in our lab. So actually, I had quite a bit of contact with Schrieffer there for a year. I know Schrieffer quite well, but I don’t know some of the other people as well. Now some of the people he had in our lab, I know very well, like Tom Morgan. But this one (book list) I don’t know, but that’s a superconductivity guy. And Mattis, I didn’t have contact with, but Schrieffer, I did.
Ashrafi:The physics post-docs and students, you didn’t have much contact with?
Holonyak:No, I had contact with all of these that he brought into our semiconductor lab, and were all physicists: Schrieffer, Mike Miller, Milton Valenta, Steve Arnold, Willie Portnoy. I had a lot of contact with Willie over the years. I don’t know some of the others, then. They occurred (appeared) well after my time, so I didn’t know them very well. But these first people I knew very well.
Ashrafi:When did you take quantum mechanics?
Holonyak:I don’t remember exactly when, but it must’ve been simultaneous with the semiconductor course or just after that. But it was out of sequence.
Ashrafi:And the course was in ’52?
Holonyak:Probably ’52. It had to be ’52.
Ashrafi:Is there much coursework after quantum mechanics?
No. After that point, I’m so heavily engaged in the laboratory and what I’m doing and reading in and around the project at work that I probably had next to nothing. I don’t have the transcript handy, but I could check that and put hard numbers on that. What I’m saying is that my pattern of study was essentially eclectic. It was sort of what an individual gets into, who didn’t start off necessarily aiming at a specific direction, but was in an area in and around physical electronics.
As things were happening to me, I realized that I didn’t necessarily want—When I got my Master’s degree in ’51, they had started Illiac, the first digital computer that was built here. They were building, two of them, side by side—one to stay here and one to go to Aberdeen Proving Grounds to the people who had commissioned the project. They tried to hire me to work on Illiac, and I thought about it, and I said, “No, I like to work better on the physics of electron devices.” When I started working on electron devices, which was microwave tubes and free electron devices, little did I know how much the materials component of this thing would be. And then as you get into the semiconductor, it becomes even bigger.
It’s many, many, many years later, and Bardeen and I are in the old building, and he made a comment to me. If there was a mistake made in the old lab—he means in his laboratory, this one where everything started, the semiconductor laboratory that he had, an electrical engineering research lab—that if there was a mistake made, it was to underestimate and not figure out how big the materials component would be. Now see, that’s a very major statement to come from a guy like him, because you have to understand, who understands materials as well as John does? He’s worked on diffusion problems and been into the solids and all of that, and understands what the materials issue is all about.
For him to say that, they still underestimated how big that part of it would become, is a very revealing kind of statement. This doesn’t speak to the fact that Bardeen didn’t know what he was doing and didn’t understand materials and all that; this says something else. It says that in the beginning, way up front, very early, nobody knew that the materials aspect of the situation would be so huge. If you’re up there at Intel now making a chip, and you’re stacking layers and processing and all that, that’s all you’re doing. I mean, there’s some component in there of the organization of what these elements are and all that, but the biggest part of what’s causing that factory that fortune, is how you do, how you make fine lithography and stack these layers and what materials and what dielectric constant’s and what levels of contaminations and all that. See, John was absolutely right. It wasn’t so much a mistake as nobody knew how vast the materials thing would become, and how the need for the new kind of device, built into the solid, would let you render geometry into electronics that otherwise couldn’t be done.
You cannot sit out here in the world we’re in and assemble piece parts. I think this is a mistake that’s being made right now with everybody, dreaming that nanoparticles are going to lead to some new electronics. That’s basically like saying, “Here’s a bag of dust. Now assemble it.” That’s like vacuum tubes. Here are a bunch of things—now put these things together and render a geometry with them. If we go into a platform, like a solid, where the issues of how crystals grow has already assembled your atoms and stacked them and connected them. Now you have the freedom to go in there and put other kinds of stacking on there, where you have a platform, and where positions are already defined for you and set up for you on a microscopic scale, on an atomic scale, and where you can play with your doping and your super structure of geometry on there and render a geometry into a certain form that will allow you to do electronics and move currents and move photons and do all this.
Because there’s a magic in having a substance which will give you the electron, give you the hole, and give you a photon because all three are vital to doing particularly optoelectronics. I don’t see the semiconductor even getting close to being superseded. Not even close. There’s a lot of things that fit, fit, fit into all of this, that you learn bit by bit by bit. So John I don’t think John is making a statement about the weakness of his thinking. I think he’s making a statement about the world did not know—nobody knew the importance of that materials and working on materials, all essentially being allowed to you because of the physics of materials and the atomic physics of materials. Not the inner physics of the atom, but the outer physics of the atom and how things can be hooked together in what kind of things and what form. It’s a fantastic thing when you think about it.
Ashrafi:How was that first lab organized? Did everybody work with Bardeen? Was everybody on their own? Was there a hierarchy?
There were just a few people at first. He’s got some surface physics going that are experiments still with Walter Brattain at Bell Labs. Walter would come and visit, because without John, he’s partly lost. Walter was a good experimentalist, but with John around, he’s a great experimentalist. That was Bardeen’s real partner at Bell Labs. Walter missed John and he would come and visit. In fact, it was too bad then that we didn’t have all the boxes that we’ve got now because I could’ve had some fantastic pictures of some of the things that went on in there. They frequently were in the lab. There was a blackboard just outside my little cubicle, and they would be working things out on there. Walter had these half glasses. He’s doing something on the board, and I had come out of my little cubicle and went back by a bench back further in the lab. I’m sitting there watching, he and Bardeen are doing something, and Walter got something wrong. John then would move up, take the chalk and do something.
I can remember then, he’d move back, and Walter looked at the board, looked at him, looked over the glasses and said, “God damn it, John. How the hell did you do that?” There was a very interesting kind of interaction. So Bardeen had the surface experiments going on, and then this other guy, Richard Sirrine, was working for this other professor trying to make lead salts—lead sulfide, lead telluride, stuff like that, and that was sort of a messy job and he wasn’t getting too far on that. I was working on a related surface thing, but what Bardeen wanted me to work on was to work on… It’s fundamental to a transistor that you’ve got either injecting contacts or collecting contacts. He wanted me to try instead of doing some of these experiments that they did in the beginning with point contacts on the semiconductor, to actually make sMoll PN junctions, and then see what happened when I varied the surface conditions around those junctions—what would happen in terms of injection, collection, and other things.
So a lot of his ideas were related to some of the stuff that had actually been part of their thinking and work at Bell Labs. Then as he brought in more people, the project would expand. At some point, he came in with a post-doc from France, and that guy had measured germanium in France as his thesis, the PME effect, photo magneto electric effect, which was a takeoff on some Russian work. This guy wanted to repeat this in silicon, and somewhere, Bardeen got some silicon so he could do it. It was sort of a horrendous experiment, but the visitor got some results. His name was Bulliard. He got something and Bardeen wouldn’t let him publish it. He says, “No, you can’t publish it until you make measurements on the silicon and tell some of the fundamental things about it.” In other words, it’s doping concentration and stuff like this.
So the Frenchman is stuck. In those days, we didn’t have all the tools you have today. We made everything (apparatus) in there, and I had made a four-point probe, a homemade one. When you drop this four-point probe, which is point contacts, on top of a piece of silicon, you’ve got four little rectifiers, four little detectors. And somewhere here in another building, there was a radar swinging a signal around. Every time that signal would come past our lab, those four little detectors would pick up some of that signal, and the meters are bouncing around and Bulliard is going crazy trying to measure this. Bardeen looks at it, and he says, “Well, I guess we’re going to have to move all this stuff down the hall.” They’ve got a screen room down there.
Then you’d move all the equipment down there and you can get the screen room and maybe shield out all that. And this poor Frenchman doesn’t want to do that and he’s stewing and stewing and stewing. I let him stew for a while and then I said, “I know how to do that experiment. I know how to do that measurement.” He swore at me and he says, “No, you don’t.” And he says, “Bardeen doesn’t know how to do it.” I said, “Okay, fine. Go ahead and move all the equipment down to the screen room.” That doesn’t please him. So finally, it’s a typical American’s typical opinion of a typical Frenchman. The little guy had a cigarette dangling from his lip. He must’ve had polio because he had a problem with his leg. He swore at me again and he says, “No, you don’t. Bardeen doesn’t know how to do that. How could you know it?” And I say, “Okay.” So then he finally came back and he says, “Okay, Olonyak.” He couldn’t say the “h.” He says, “Hey, Olonyak, okay.” I had made a homemade sandblaster, and I knew that I had to essentially sandblast that surface and make its surface recombination go way up. You see, if I roughen that surface, I don’t make it possible for an electron hole to exist.
In other words, if a signal comes along and generates an electron hole pair, I’m going to do some collecting, I’m going to have false signals, and I’m going to have meters bouncing around. But if I make it impossible for there to be an electron hole pair to be alive at that surface, and I do that by roughening the surface, I know that it’s still going to have the same ohmic behavior; it’s still going to behave the same way in respect to the four-point probe measurement and all that. Immediately, I sandblasted it for him, put it back there, and he got it. He said, “Olonyak, you’re a genius. Not even Bardeen knew how to do that.” Well, that’s an experimental thing. John was not a hands-on experimentalist. He was a smart guy; he was very ingenious. But John, like everybody, knows something and there’s certain things he doesn’t know, and we’re all like that. He was a fantastic guy. I never ran into anyone in my life like Bardeen, in many ways. What Bulliard was saying was ridiculous. It’s just that I know something about these things that John would have no way of particularly knowing.
Ashrafi:So up to now, you’ve talked about yourself as the math guy. Now all of sudden, you’re the experimental methods guy.
Holonyak:Well sure, because as a kid, I’m in and around people who are rendering everything, doing everything. Once I see certain things, I’m able then to make my own transition to the next aspect of that. So once you go into someone’s laboratory, once you’ve been into any laboratory, you pick up certain tools. Tools are the things that we’re always using. In fact, math is a little bit different because it also is a conceptual thing. It’s a tool thing in many respects, and particularly in physical science since we’re always using it for reasons of the tools that it gives us. But it also is a conceptual thing that lets you look into things, whereas in many things in the experimental world, the tool aspect of it is before you and first in your face, and you see some tool that you see right away is useful for something else that you have in mind or can go do. I guess I was alert to that. Mathematics always fascinated me and I found it useful.
Holonyak:The transistor laser (current work)…is of major consequences. It means more fooling around with crystals and building and cutting and trying, and we’ve been within 30 degrees of running it room temperature continuously, so we’re very close to having something that is a major new component that nobody’s ever had before. I find myself thinking more about that, and that some of the past, is important because it’s partly how we got here. New thought on top of old work make new again.
Ashrafi:So were you all working directly with Bardeen?
Yes. His office was in the old physics building, which is now metallurgy, and that was only essentially spaced from one building to another building apart. So every day he would come over into the lab to see what we’re thinking and see what we’re doing. He concerned himself with this: what idea we’re working on, whether it makes any sense, whether it has any value and all that, what kind of problems we’re running into working on what we’re working on, and also about enough funding to make sure we got assistantships, everyone’s getting paid, and we got the supplies that we need.
Of course, it was a different kind of a world then in terms of the machinery, but we were, in those years, building a lot of our own equipment. We were making our own electronics. You didn’t go buy all this stuff—there wasn’t any of this stuff around then. Imagine if when Bardeen and Brattain were talking there, if I could have been filming or recording. That would have been very valuable at this point. Those tools now have come into existence because of the semiconductor, but didn’t exist then. Everything that we were using was essentially a vacuum tube “box” of some kind making measurements on semiconductor items that we’re trying to uncover and study.
So it was a different thing. And he’d come in every day to see where we were, what kind of problems we were running into, what we have to do next. Then there would also be little in-house seminars. We had a room down the hall where we could assemble and have a little seminar about something that’s either on his mind or… There was another thing he was working on with a post-doc, it was an electrolytic analog of a transistor. What it looked like was a beaker, and it had a metal plate down in the bottom, and it had two metal plates vertically on either side, so you have then three electrodes, one of these vertical electrodes, another one, one’s the collector, one’s the emitter, and one down at the bottom is the base.
Then there’s some kind of electrolyte in there. What Bardeen was interested in was this was something like an electrolytic tank, but an electrolytic tank for a transistor. If they could launch a current on one electrode and then collect it on the other one and vary the potential on that base, they could see the equivalent of transistor action, but with this fictitious kind of a transistor. It’s not really a transistor, but a little bit like it. Bardeen, once, many, many, many, many years later, brought me his notebook, and it was not unusual for him to do something like this—he just dropped it off. He didn’t say he wanted me to look at it or not look at it or what; he just left the notebook. Now, this notebook he had started keeping from the time he came, and it had entries, I can remember, in January ’52. We didn’t start his laboratory until September of ’52. So he came in ’51 in the fall, I took atomic physics from him, and then I’m talking about his notebook entries in January, coming up from ’51 into ’52, and he had some interesting things in there.
He had some comments about what he was interested in, what kind of transistor things he was interested in; he had some comments about superconductivity in there. Then we go further in the notebook, it’s a later point, and we’re already into when our laboratory exists. He comments about what happened, that project of the electrolytic transistor and this post-doc. I don’t know if the Center for the History of Physics has this or not. That post-doc, Harry Letaw, is still a guy that’s alive, because he contacted me a year or so ago. He was a physical chemist by origin. Here, this guy right here.
Ashrafi:Ah, mentioned in Hoddeson’s book.
Yes, on page 177. He was a loquacious guy and a personable guy, and, see, it says here, “In the Electrical Engineering Department, Bardeen continued to work on semiconductor physics. One goal was to establish an experimental semiconductor group in the EE research lab. Holonyak and Richard Sirrine were the first two grad students in the semiconductor lab, the first two post-doctoral researches were Harry Letaw (and that’s who I’m talking about), a physical chemist, and Roy Morrison, a solid-state physicist. Bardeen directed them in an informal, paternal fashion.” That’s more or less right. Anyhow, in that notebook, Bardeen comments about this electrolytic cell, and how the project went off in the wrong direction.
And John Hoddeson references in there about where he’s getting some of his right ideas and thoughts, and how the post-doc didn’t follow his suggestions properly and did something else that turned out to be an embarrassment, and John talks about that a little bit in there. Very unusual statement, because generally he was not given to saying anything negative or left-handed about anybody, but in this particular case, it embarrassed him that something that they had promised or were close to getting done they got delayed because the post-doc went off in the wrong direction. Actually, John Bardeen had the right direction, and if the post-doc had followed what he said, they would have made the progress that, of course, didn’t occur.
This thing was interesting to him also practically, because there was a possibility that he could make a constant current source out of it, in other words, as a battery, instead of having a constant voltage, you could have a constant current coming out. And they published something in the Journal of Applied Physics, and that finally essentially was it and it died. See, some things don’t go anywhere, for one reason or another, and that finally just essentially died. The field went off in a different direction, and that didn’t become an issue anymore. It was forgotten.
Ashrafi:So what does that “informal paternal manner” mean?
He didn’t come in and say, “Here, I’d like for you to measure this or to look at this,” or anything like that. He would come in and talk with you and say, “Well, that looks interesting. Yeah, that’s worth looking at and pursuing. Incidentally, I learned right away, in the beginning I asked him about putting some things together and whether I do this in a vacuum or in a hydrogen atmosphere, or whether I do this in a nitrogen atmosphere, or whether I do this in forming gas atmosphere or something, and sometimes if you asked John a question about something that he doesn’t know about or doesn’t care about or whatever, there’s no response—there’s none, there isn’t any.
I learned after a bit that there are some domains that are not his, and it’s fruitless to bother him about stuff that he doesn’t know about. Why would he know about them? And why bother him with things like that? After a bit, I began to learn what I could ask him about that made any sense. The reason I think I got along with him very well was I never felt bothered by the fact that he may be thinking poorly of something I’m asking him. In other words, I don’t think John was bothered at people asking him questions. I think if he found something of interest, he responded; if he found something of no interest or he didn’t know about it, he didn’t say anything.
That can unnerve some people. Talking to a guy like that on the phone is difficult sometimes, because in Japan, if you’re talking to somebody and they don’t see you, you hear “A so, des ca. Ah. Uh. Oh.“ There’s feedback. And talking with Bardeen, sometimes, you don’t get the feedback, and that can be disturbing. But it never bothered me very much. I never pursued him on something, but I would not be bothered by asking him about something, thinking that maybe I’m asking him something foolish. If I thought about it and thought it was something worthwhile, I’d ask him about it and see what his response was.
Ashrafi:Did you work on one project in Bardeen’s lab, or more than one?
Holonyak:Actually, for myself, I worked on this business of the sMoll junction and surface conditions in and around it. But I also, in doing that, inevitably wound up making some of the equipment that was needed in the place, I wound up learning some of the methods that we would need. In other words, I wound up making some of the devices for the other people—something for Tom Morgan, I don’t remember what it was anymore, and for Richard Sirrine, a device that was more like a transistor—because I had learned some of the technology and methods to do that, and I left some of that for the others. And as I said, I helped the Frenchman get his measurement, and in some general way, maybe there was some contact with the other. But specifically on an assigned project, no. I worked on this one thing and got to a certain stage and finished and went to Bell Labs.
Ashrafi:Now did everybody in the lab make things for each other, or was it mostly you that made these—?
Well in the beginning, there’s always Richard Sirrine and I, and Richard had a lot of lab and electronics background, so Richard and I probably made all the first stuff. I don’t remember the two post-docs doing—Letaw was a chemist, and he didn’t do any of the electronics. Richard and I did all of that, and the physicist, Roy Morrison, all he did was worked on that specific experiment of Bardeen and that Bardeen had reproduced here, the same one that he was doing with at Bell Labs. Roy Morrison didn’t make any of the stuff in the lab. That was Richard and I who did all of that in the beginning. Letaw, the physical chemist, helped out with some of the stuff which was the wet chemistry part of our little laboratory then, but he didn’t do anything relative to electronics.
All electronics was done by me and Richard Sirrine, and in fact, once, Bardeen had something sick in some audio equipment, and since Sirrine had once run a repair shop, he was handy at repairing some of that stuff, and I remember Bardeen brought it in to Richard to repair, a home unit of some kind. And I don’t remember what Richard did, maybe it was just check the tubes and replace some tubes. I don’t remember now, exactly. But all this stuff related to the electronics of what we were doing. Richard and I did all of that in the beginning. See, the others were not there the first year. We started in ’52, it isn’t until ’53 that Bardeen has brought in this next wave of young grad students. And I overlap with them for a year, and then Richard is still there longer with them before he’s done. But I’m the first one to finish.
Ashrafi:Is there more to say about your thesis than the description you gave a few minutes ago?
It’s like a lot of theses: at this stage it’s more or less uninteresting. Because what I did was when I left, I went to Bell Labs, and I wind up working on silicon. Now, I go back to germanium for other reasons at General Electric for some other stuff, but basically, the silicon things erupts and it becomes a major whole new world, so I didn’t even—Bardeen had sent me some letters about my thesis and getting published, because there were some results in there about how surface channels work along the surface.
I did publish some little pieces, but not the main part of looking at how the surface conduction behaved around a sMoll junction. Basically, later on I got myself busy with all the silicon stuff at Bell Labs and then wind up in the Army, and I never get back to it, for one reason or another, and by the time I’m not even looking at it again, it’s already something that’s the past and who cares. So even though Bardeen wanted me to write it up—He made a reference to it once in something he wrote up; I don’t remember what it was, in some encyclopedia or something, because of some further insight on leakage and surface behavior on germanium. But germanium became more or less uninteresting once you get to a certain point with silicon.
Silicon supercedes it in so many ways, and germanium doesn’t have that nice tough oxide that is inherent on silicon, SiO2, you know. The natural oxide that you can put on silicon is a fantastic thing that allows you to do so much. So once you get to a certain point with silicon, germanium is essentially forgotten.—It’s an interesting semiconductor, but it’s not an interesting device material to compete with the silicon device. And so silicon runs away—I’m sure at this stage, the silicon literature is huge and germanium literature, even though it’s big, is miniscule compared to silicon.
Ashrafi:So you got a PhD, and was it clear to you where you wanted to go with this while you were learning it?
Holonyak:Oh, well, that was interesting. One day Bardeen came into our lab, he came in the door, and there was a partition, and you went up around a little bit, and you came around a partition and we had our little chem. Lab. I’m in that little chem. lab and the wall doesn’t go up to the ceiling, and so when he came in, I hear his footsteps. After a bit, I could tell his footsteps, and I hear him say to someone, “Where’s Nick?” And I’m thinking to myself, “Wow, I wonder this is it,” in other words. “Is he dropping the bomb on me? Am I doing a rotten job, and is this the end?”
Ashrafi:Were you walking on eggshells at the time? Why did you think that?
I don’t know, because it’s not typical for him to walk in and say, “Where’s Nick?” I heard that much, so I come around this way, and he’s coming over this way and around there, and just a little bit further, there was a little office built into the place that was up on essentially a higher platform or something, and it was an office there that he could use if he wanted it, and he says, “How about coming in here with me?” I go in with him, and we go through the door and close the door, and he sits at the desk like where you are, and I’m sitting here near the door in a chair, and I’m thinking, “Well, this is it. He’s probably going to boot me out,” because why did he come in and say, “Where’s Nick?” And he says, “How would you like to go on to a fellowship?” I almost fell out of the chair, because, you see, I’m expecting some criticism and he’s asking me would I like to go on to this fellowship. He’s just gotten a fellowship from Texas Instruments from Gordon Teal. He was good friends with Gordon Teal. Gordon Teal was the germanium expert at Bell Labs, and he’s the guy that would in the Shockeley, Sparks and Teal grow the first germanium transistor, the NPN junction transistor.
At any rate, Gordon Teal is a big wheel down at Texas Instruments, and he’s giving Bardeen a Texas Instruments fellowship and Bardeen is determining who in the place is going to get the fellowship. So of course I went onto a Texas Instruments fellowship, and so that throws me into contact with Texas Instruments and a whole bunch of people. I’m already in contact with some General Electric people and Bell Labs people and other people, you know, and so forth. When you were a grad student then, and you were ready to go out into the world, there were many opportunities. Frequently a recruiter would leave you alone and you’d say, “Well, I’ve got six more months to go,” and that didn’t matter. They were just finding who they wanted. Bell Labs in particular found what they thought were the people that they wanted. So I get to the point where I got offers at Texas Instruments and General Electric and Bell Labs, and I’m down to Bell Labs and General Electric. One day, Bardeen would ask me, “Well, where are you going?” and I’d say, “Well, Bell Labs.”
Ashrafi:Now let me just get the chronology straight. You were a student here, you get your PhD in ’54, but a fellowship is for 1954?
Holonyak:’53 and ’54. We’d been in his lab for a year, and it’s ’53. I’m on this fellowship that school year, the ’53-’54 school year.
Ashrafi:Was it ever an option to go on, for you to receive a professorship at this point, or a full scholarship?
Holonyak:Well, yeah, but that was not the pattern of things in those years. There was post-doc-ing. The Frenchman was a post-doc from France. You see, a lot of places felt the U.S. was deep in a lot of these things and a lot of other places were not, and it was not unusual—He had some post-docs at a later point, the German post-doc and others, that I knew about but I didn’t overlap with. The only one I overlapped with at that time was the Frenchman. Rupprecht No, he’s not in here. She missed it. She missed John’s post-doc. There were post-docs coming in this direction, but I don't think many of us were post-docing the other way. The people I was around or the people that I had contact with were frequently going somewhere in Industry. The real big players—Texas Instruments was a player, but it wasn’t the biggest one, then. The biggest ones were Bell Labs, GE, maybe IBM, I don’t know. I’m not so sure. RCA—
No, they didn’t exist. They didn’t exist. They didn’t come into existence until much later. Silicon brought them into existence. Anyhow, no, those guys didn’t exist. There’s a different pattern of people that existed. So I get my offers down to, well, then he asked me which one am I taking, and I say, “Bell Labs.” “Well, that’s a good place,” he says. In a matter of a few days later, he says, “Are you still going to go to Bell Labs?” “No, I think I’m going to GE.” So I seesaw back and forth for awhile, and then in the end I thought, “Bell Labs is giving me $600 a year more, giving me $7,800. It seems funny. Like, if you ask me certain things about salary, I wouldn’t even remember. But I remember that. $7,200 at GE, $7,800 at Bell Labs. Six hundred dollars a year is $50 a month, and there was no way that I could say that Bell Labs was inferior to GE, not alone, $50 worth.
So the decision to go to Bell Labs was probably one of the best decisions in my life because I wind up working on silicon with John Moll. And it’s Moll that was bringing, that was selling Jack Morton on silicon, and to go “silicon”, and why go “silicon” and all that, and it’s that that Shockley is going to be handed. And his contact point with Bell Labs, to take this technology to the west coast, is with Moll’s group. It’s the calling back and forth, and handing the technology. When the silicon technology thing started, there were maybe, there’s Goldie [?], John Moll and myself, in John Moll’s group. And then Morrey Tanenbaum, up in a different group, and Carl Frosh and his technician doing the oxide of silicon, and Carl Thurmond. There were some other people playing around with other parts of it, but we were the only ones who were doing the part which was diffused device, metallized, the technology that’s invariant that still part of all of today. When I think about it, that’s one of the best decisions of my life. Because what do I get my hands on? I get my hands on the thing which is going to be the final resting place for the business of the IC and all that. That’s the beginning of all that kind of stuff.
Ashrafi:Why was Moll pushing silicon?
Holonyak:The hierarchy was up here, with various things, but the part Moll was in was switching devices. If you look at a computer, you’re making one’s and zeros. You’re switching on and off. If you look at the Bell system, there was a pie diagram like this, and I remember John Moll taking us down to the switching systems people and looking at the pie diagram. If you’re out in the world of Electronics where I was, you hear about amplifiers and all that kind of stuff. But the truth of the matter is, in a huge telephone system, you’re switching and routing and setting up paths to go here and there. Then the thing has to tell you that, “I called you from here.” You guys are in Maryland, right?
And that you didn’t call me this way. In other words, I should be billed a certain amount for a talking path for a certain length of time to Maryland, and so we’ve got to do some billing and switching and all that, then we’ve got to do the switching to set up the thing. So the switching part is a huge part. And the Bell system, at that point, electromechanical, fancy relays—they were really good at all that kind of stuff. They were going towards electronic devices, but it’s silicon that has a high enough gap so that when it’s off, when it’s in a reverse direction at the junction, it’s low enough leakage that it’s off. And when it’s turned on, of course it’s on. So the concept of an ideal switch is just a concept. How do you build something that in the off state has no leakage, and then as soon as you go into slight forward bias is infinitely conductive, in other words, on? It’s not trivial.
That’s impossible. If you look at the business of an energy gap, and injection, you’re always going to have to have some kind of an activation energy. You’re going to have to have some kind of an E to the Minus E gap over kT, and so you’re never going to have—Some electronics people say, “Well why can’t you give me instant “on” a…” You begin to laugh and think that shows poor understanding of physical science, that you’re not seeing the activation-energy facts of life. So if we look at silicon, we’re going to have to pay a slightly higher price in voltage in the forward direction to make things conductive. But that’s trivial. But the off thing in the reverse direction, is not trivial. You want this thing to really be off.
If you’re making a power device and you want to hold off a thousand volts, you don’t want that thing to be leaking at a thousand volts because that’s a lot of dissipation. And if it’s something like germanium, it’s inherently a lot leakier that silicon. So it has to be silicon. In fact, I proposed to John Moll that—Well, he wanted to make a PNPN switch. Shockley laid claim to it based on something in the early Bardeen and Brattain transistor, but he had never seen one, he’d never made one. Then Efers of the famous Efers-Moll equations, Efers hooked two transistors back to back in such a way that you could see that it was a possibility. We knew that much. But no one had ever seen a working PNPN switch. Tripped on, it’s really conductive; off, it’s really off. They thought, they had dreams of having these things all over by the billions, in a switching system to route phone calls and all that.
That’s what Shockley thought he was going to build on the West Coast, and he made some mistakes. But our goal was to make a silicon PNPN switch, and I remember arguing with Moll, a friendly argument, you know, “John, I can make one of these things out of germanium.” “Yeah, you can make it, but it will be leaky.” “No, we’ve got to do silicon. If we have to find the technology, we’ll find it, but we have to do silicon.” So it’s a switch that’s off that’s the motivator. You’ve got to do it, and it’s got to be silicon.
Ashrafi:Why did you want germanium?
Holonyak:Germanium, I had already worked with. I had worked with germanium for two years in Bardeen’s lab and I knew how to make germanium devices. Suddenly we’re going to make a silicon device, and do we know anything about making them? We know something.
Ashrafi:What was hard about making silicon devices?
Holonyak:There was no technology. There are some things, but with germanium it’s easy to make alloy junctions and to make PNPs and NPNs and all that. The whole Industry is making PNPs and NPNs by the alloy process, and I can quickly devise a way to make myself some kind of a switch with this. Silicon, well how am I going to do this? Which alloys, which processes? Nothing etches the same, nothing works the same—you’ve got to find a whole new technology. Now what is known is that you can diffuse a slab of silicon and make it into a solar cell.
Ashrafi:You can make a rectifier.
Mort Prince, and a guy, Harold Veloric are making rectifiers at Bell Labs using diffusion into silicon and then some sort of either deposited metal or chemically deposited electro-less metal on that to make a rectifier. And solar cell people are making solar cells involving diffusion. You see, the business of growing a crystal out of the melt meant cutting it up and all that, that exists in silicon, but that we consider a primitive technology and limited in what we could do. Incidentally, I’ve written a manuscript on this that exists somewhere in the electrochemical society but never published, and Howard Huff down at Sematech is trying to see now if they can get the electrochemical society to publish this. I thought at the 50th anniversary it would be published, because I wrote up what happened in Moll’s group.
I comment about this in the Nebeker interview. In that paper, I go through in some detail what we did in Moll’s group and how we arrived at some of the first diffused silicon transistors, the metalization and the first PNPN on switches. All of that leads to the thyristor. That paid off to General Electric, and the thyristor did not pay off to Bell Labs, in a switch for the switching system, nor did it pay off for Shockley, in what he thought he would get on the West Coast. But it did provide a technology that then Noyce improved, could take and go into and form Fairchild, and then started the Silicon Valley thing. You see, I maintain that the start of Silicon Valley is not Shockley, in Silicon Valley. The start of Silicon Valley is what John Moll was selling Bell Labs management on why to do in silicon for switching devices, and why to do diffusion and metalization and all that.
Then we go from ’54 into ’55, at about this time of the year, and Frost discovers the oxide. So when I go in the Army that Fall of ’55, the Bell Labs lawyers say that I can talk about anything that we’ve been doing, but what I can’t talk about is the oxide. And they hide that for two years from the world, because they know that it’s very valuable and very important. It’s the oxide that is the key to being able to build multiple devices on top of silicon and then metalize it across and interconnect and make a chip. Take away the oxide, and it all goes away.
Ashrafi:Just so I don’t forget my questions, do you know who funded the research in Bardeen’s lab before you got the fellowship?
Holonyak:Yes. Bardeen was at the Naval Ordinance Lab during World War II. In fact, Charlie Kittel has told me something about that because he was only a couple of desks away from Bardeen. So Bardeen knew some of the Washington people, and he had some grants or contracts from I’m not sure if it’s Navy or Air Force people, or the Air Force Office of Scientific Research. It’s some of the agencies, but I don’t know which ones, exactly. We could probably find that out.
Ashrafi:Okay. The funding I meant, was funding semiconductor research.
Holonyak:Oh, yes. Oh, yes. Sure.
Ashrafi:In Bardeen’s lab.
Sure. Sure. Then he had like this Texas Instruments fellowship, and then we had gifts. We didn’t have any capability to grow crystals, so we were getting gifts of crystals from Bell Labs and GE and various places, you see. In fact, in those years, there was no industry that could supply you with wafers the way there is now when you go buy certain kinds of crystals and all that, no. In his notebook he (Bardeen) makes a comment about working on certain kinds of, on silicon. In fact, in that January of ’52 entry in his notebook he comments about leaving germanium and working on silicon.
See, but I don’t see silicon work, however, until ’54 when I go to Bell Labs; but Bardeen is already making comments about working on silicon because he’s already worked on silicon at Bell Labs and knows about silicon, and has some preferences for silicon for various reasons. And he had some of that in his notebook in January ’52 or in February. And he makes a comment that we’d have to get some silicon, so maybe we can get it from Bell Labs or maybe from the Naval Research Lab. John, from his Naval Ordinance Lab experience, he must have known some Navy and other people and places to get either materials like crystals or funding or what have you.
So I know that we were getting funding essentially from some of the agencies, because we would wind up even writing reports. I don’t remember too much about them, but there would be some reports. It was interesting about him. The post-doc Letaw would be raising holy hell about, he says, “You know, this guy is going to be so famous.” See, Bardeen had no Nobel prizes yet. With them (the remodeling of the M of I building) disrupting me, I don’t happen to have a lot of stuff to give you which I would have had nor Molly. But I make a comment in here. This is not something you’re apt to see. It’s one of the electronics honorary societies. But I made a comment here about Bardeen, commenting it was all lucky for me, because by the time John Bardeen came to Illinois in ’51… and so forth.
But down here, I commented about some of the things that we did in his lab and who visited and how we wound-up doing what we did. “I should mention that in spite of the fact that John Bardeen was not yet a Nobelist, all of us in John’s lab, were keenly aware that he was indeed much, much more of a talent than any of the other people around us. We could sense or feel that John was special. That would merely become more evident with time and the great fame about to descend upon John. He was the kind of person who would even send his former graduate student stationed in the Army in Yokohama, a greeting card from the 1956 Nobel Prize ceremonies. I am sure our high regard for Bardeen had a large effect on us. It defined for us a whole different perspective on learning and research and what mattered and didn’t matter.”
Ashrafi:So what was that perspective on research?
Holonyak:Well, that how you choose a problem—why you go this way instead of that way. Why work on this instead of something else? In other words, why should he even be concerned with the semiconductor? He and Waller had invented the transistor. That’s going to make him famous. Why not go work on something else? Why not just concentrate on super conductivity?
Ashrafi:Can you sketch some of the answers to those questions? What did you learn from Bardeen about why he decided that?
Because he, as Seitz and Lillian point out, Bardeen was the kind of guy that when he knew an area—I found this to be true working on some other things working with him. If John had ever worked on something, you didn’t have to go back and re-teach him what the field was. We ran into a diffusion problem with some interesting intermixing of layers where we don’t lose the crystal entity, but where we scramble the atoms. Where we have, let’s say, a layer of aluminum arsenide and a layer of gallium arsenide.
If you thermally anneal this, to try to scramble aluminums and galliums (the atoms)—aluminum being a funny gallium and gallium being a funny aluminum—maybe you can interchange them. You can. You can do this thermally. But you have to heat the crystal a long time at a high temperature. But if you diffuse in some of the favorite impurities that will diffuse in there, like zinc, you can drop the temperature from the 900-1000 range down to the below 600 degrees, and in a very short time, intermix the layers, say, the ALAs and GaAs layers of a superlattice. There is very interesting physics of diffusion involved in there. When we ran into this and I wanted to deal with him and talk with him, I knew he had already worked on some diffusion effects, the Kirkendal effect in various kinds of brass and other stuff.
And when I did get to him, when we looked at this and all that, I didn’t have to re-teach him anything about diffusion. He needed some of the references on these materials that we were dealing with, but he didn’t need to know anything about all the fundamentals of how all of this worked. He immediately had some insights that we were missing. He tended to stay around what he knew about. He didn’t pretend to know what he didn’t know. For example, you didn’t see him running off to take the pairing ideas and the BVS theory to go do something in astrophysics, neutron stars or something, I don’t know what. He tended to stay in and around the quantum theory of the conductivity of solids. No human has ever lived, that did more on the quantum theory of the conductivity of solids than John Bardeen.
But you see, he never leaves that. He doesn’t go off to do the theory of a photon working on chlorophyll and green stuff, and going off to suddenly be a biophysicist. He’s not one of the Nobelists that suddenly sees a whole different world out there that he’s going to enter and do biophysics, for example, or something like that. No, he stayed relatively close to the stuff where his mind has been closing the trap tighter and tighter and tighter on problems he understands. He doesn’t try to be everything to everybody. That’s why, for example, if he were alive now and he walked in and I told him about what we’d done in transistors, he’d have a big smile. Because he would see immediately why and how we went after that. Because that’s something that Bardeen was excellent at. He may be stymied about something. That doesn’t mean he’s abandoned it and gone off into something else. That just means he’s doing something else for a while that he’s productive at, but as his thought processes and the knowledge base is getting better, he’s going to close in on the original proplem again. The superconductivity wasn’t a thought he had when he came here. He gave a seminar here, that I listened to once, where his thoughts on superconductivity he said, were already occurring when he went from his thesis work to Harvard.
This guy, Bob Laughlin has written a book called A Different Universe. I’m reading it now. And he mentions the Schrieffer story in there about what Schrieffer did and how that unearths the notion of an energy gap. I’ve got news for Laughlin. Bardeen was thinking about an energy gap long before BCS. He mentioned in that seminar thinking about the gap has being part...You see, look, if he knows about gap and stuff like this, and he’s been around notions of semiconductor, that doesn’t mean he’s lost his ability to use those notions again in a superconductor context. You can’t really leave what it is that you know about because in something strange, you’re not going to be very good. But in the thing that you are prepared in and know about, you can attack and attack again, and if circumstances show up, you may be the person that’s best able to attack the problem. So I don’t think there’s any mystery about that. I don't think any of us have the freedom to leave very much of what we know about. We won’t be very productive at something else. Of course as you age, that’s another factor. But even in the beginning, you can’t flit into everything.
Ashrafi:So do you know if there was Federal funding in Moll’s group as well, or was it all…?
In our group, there wasn’t. But there was a thing that was a different group that was working on silicon. I think it was a group run by Mason Clark. It was a transistor for an Army contract or someone like that that was called a “five-and-dime” transistor. Which was a jargon designation meaning this: five Watts at ten megacycles (it was megacycles then and not Hertz), or five megacycles at ten Watts. In other words, take your pick. Give me ten Watts at five megacycles, I’ll take it. Or give me five Watts at ten megacycles, I’ll take it. I think they said they needed this because, for example, if you had a pilot down somewhere and you want to have him outfitted with a little transmitter or something so that you can send up a signal to find the guy, you wanted a transistor that could put out an appreciable amount of signal, and preferably up at a fairly high frequency.
I guess where 1.4 megacycles is down to 550 or something like that, kilocycles are the broadcast band. You’d like to go up out of that into this other area. So Mason Clark had working with him a guy, I think his name was Peterson (I think his first name was maybe John, but we always knew him as “Pete”). Those two guys were working on this thing called the ‘five-and-dime” transistor. And I remember we’re making all kinds of progress in John Moll’s group in our attack on silicon. And these guys are coming to us to get some of the knowledge, some of the data because they have an obligation to report something to their contract source. I think it was the Army, but I’m not sure.
And they’re bootlegging on us to get the information they need. Mason Clark wrote something not long ago that someone sent me about processing silicon, and silicon burning up, and all that, and they mentioned some comment in there. And I started laughing because I thought, hell, Mason didn’t know anything about what he’s talking about there, because he wasn’t part of that first work. He was in this other project, the “five and dime” transistor, which did have military support. See, if you’re in a place like Bell or GE, and they’ve got a lot of in-house support, they can sort of parcel out this little bit here and there for which they need other support, and sort of control what they’re going to give away and what they won’t give away. You can get into some funny predicaments with that. If you’re very heavily supported outside, the agency knows they paid for the work. But if you’re in a place like GE or Bell Labs and have got a lot of inside support, you can juggle that a little bit, and who knows how much they’d juggled and cheated? I don’t know.
Ashrafi:So how was life in industry? Well, how was Moll’s lab organized?
Moll was a little bit like Bardeen. I mean, he was sort of slow spoken and quiet and a lot of outsiders, I’m sure, thought other people were the real important figures in Bell Labs. But to those of us inside of Bell Labs, working in there, I know that all of us had much more regard for John Moll than for almost anyone else. And I don’t know where they had Shockley, then. I told Bardeen years later they told us that Shockley was at the War Department. That was what DOD was called in those years. And he looked at me and shook his head and he said, “No he wasn’t. He was in the hospital.” At some point, Shockley had some kind of breakdown. And I think Shockley was a troubled person.
He was troubled because you see, the semiconductor is beginning to pay off for various people and he regards himself as the father of it all, and why isn’t it making him fat and rich? So who knows what wild things he was capable of. But at any rate, I didn’t see Shockley. I saw Shockley in that period. I saw him at a dinner in Philadelphia when he was, when it was already fairly heavily and substantially rumored that he was going to leave and set up a company. But we didn’t see him in the activity where we were building our devices. We saw nothing of him, and he had no influence at all in how we made those things. It was John Moll that was the captain of it, when he says, “It’s got to be silicon,” and “We can do this,” and “We’ll work it out.” And even other groups would be coming to him and seeking his advice. I know that he was the guy advising Jack Morton who said that, “No, no, we can just compete with Philco’s jet etched transistor,” the thing Bob Noyce was working on with some of the people at Philco. That’s a germanium device, and we don’t want to stop and use the technology we’re developing on silicon to make some equivalent of that.
We’re into something bigger and better. Moll, was different than John, not nearly as deep as John in terms of knowing quantum physics. Moll, I understand, was getting a PhD in Mathematics at Ohio State and something didn’t work out, and he said, “To hell with this,” and went over and finished a thesis in electronics and then wound up at Bell Labs. But Moll was a good thinker and he understood what we were doing sufficiently well. That he was reminiscent of John, but he was not a quantum physicist. John knew what quantum mechanics was all about, and he knew how quantum mechanics lets you look into the solid. Bardeen’s case was unlike anyone else’s. John Moll was an excellent person, outstanding person; deep in his own way, but not nearly like a Bardeen.
Ashrafi:How was it organized? Did you each work independently?
Holonyak:We more or less worked independently, yes. It was almost like back doing a thesis project. It wasn’t unusual for me though, to get some silicon wafers ready and take then down the hall to Carl Frosch to diffuse, and then bring them back to do some more work at our place. In other words, it was a collective project in the sense that it was a group, but then there would be individual things that we were doing. I was doing something totally different than what any of the other people were doing in the group, and Goldey was trying to make—I was working out aluminum metalization of silicon, and he was working out, gold-antimony metalization of silicon. And in that respect, we were in some parallel thing. And George Benski was working on a deep level, unknown kind of trap state in silicon, trying to figure out why it was that silicon junctions were doing peculiar things, and that was in Moll’s group. And then he had some young guy working on some other project. But Goldey was doing his own thing, I was doing my own thing, and John Moll was sort of the glue between me and Goldey and some of the others. John Moll was the glue with the systems and other people about the fact that we were going to make a switching device. But this was essentially a small project, but the goal was to make a switching device. That was the common factor in there.
Ashrafi:Was anybody at Bell pushing germanium?
Holonyak:There were people still making germanium, yes. Jim Early was still making germanium devices. Almost everybody was making germanium devices at that point.
Ashrafi:So was it a fight to work on silicon? Was anybody…?
Holonyak:Not in the beginning, because nobody knows when the big payoff is coming. It’s Moll that sees the fact that silicon has to be the thing that we work on because we need a switch that’s “off” when it’s “off”.
Ashrafi:I meant to ask, did it require a fight to work on silicon…?
Holonyak:No. No, no, no. I think that no one was in anyone’s way to do that. Nope. No one was in anyone’s way to work on silicon. I think the goal there was, “We have to do it.” I don’t think that there was any argument that it be silicon, uh-uh [no].
Ashrafi:Was there some aspects of the work that made you think, “Huh. I’m not at a university anymore. I’m at a company now.” Or was it really just like doing another thesis?
It wasn’t like doing another thesis because when you’re doing a thesis, you’re thinking of something that you’re going to give as a requirement to the system and it’s going to indicate a completion of a certain level of attainment in an educational system. And you obviously don’t think about that anymore, because—And there was a big mistake we made in John Moll’s group. We had a lot of stuff that we should have been publishing. We left a trail of unpublished stuff. We were so focused on what we were doing and so much was happening, that we just didn’t, we weren’t as focused as we should have been, in writing this stuff up.
So there were things that could have been, and nowadays, I’m sure people would publish a lot more of the stuff that we didn’t publish at the time. And which then, leaves you like when you see something that someone else does and you say, “Wait a minute. We’ve been down that road already,” and you find out, “Well, but yeah. Where is it so that anyone has access to it?” For example, John Moll made me stop with another guy and write a design theory for the performance of these new diffused devices. And some of what we get in the way of numbers and what we’re going to get and all that, I know was fresh information at the time. It could have been put out there (journal or meeting) as a one of the device things of what silicon will give you. That was just a memo, and there were other things like that that were lost along the way.
At one point, we were going to put out a transistor paper to a meeting in Washington, and for some reason that was cancelled. I don’t know why. In other words, I think whatever the hierarchy was doing, and their concerns were of greater interest than the importance of the work as such and it’s reportability down at the bottom. Obviously, the top frequently did not have it right, because you can’t sit at the top and know what the importance is of something that’s being generated out of the bottom. It’s very different.
Ashrafi:Who’s the top, now?
Holonyak:Well, it’s the hierarchy who are the bosses on up above Moll, on up to whoever existed up there. Well, one of the guys above him was a guy, Anderson. I don’t know his first name. And then the next level up was Jack Morton. Jack was a difficult guy. He could be rougher than hell, and hard to argue with, and he was right simply because he was Jack Morton, and frequently that’s not right. I see this all the time, people in high station (wrong as hell!) who think they are right simply because of their exalted job position.
Ashrafi:Is it because he’s Jack Morton or because he’s your boss?
Ashrafi:How much did they direct what people, like you say, at the bottom were doing?
Holonyak:Well, at one point, he was going to stop us to duplicate this Philco device, and finally John Moll won the argument with him, that “no”, that was not going to be a smart thing to do. We’d be smarter to keep doing what we’re doing because it’s too important and it’s yielding too much. And you’re talking about something when you’re talking about that other thing, which is not the answer and is not the right thing to do. And so I think that for a couple of weeks he had us screwed up on what we were doing, but we finally won over that case. But he was a difficult guy and he could thwart you just by saying, “No, that’s not what I want,” and stop the project.
Ashrafi:So there was more hierarchy than there was in Urbana?
Holonyak:Oh yes, sure, sure, sure, sure. No, I never saw that kind of thing in a discussion with Bardeen. Bardeen’s reaction was, “Let’s see what we can get done,” and he never came in and said, “Stop what you’re doing.” Jack Morton could come in and say, “Stop what you’re doing. I want this.”
Ashrafi:I’m trying to understand, was there either personality differences, or differences in the way the institution was organized?
Holonyak:It’s both. It’s both, because that, Bell Labs is very hierarchal, and it’s a top down structure, and it tends to have people in the hierarchy who are technical people so they feel like they never have lost their technical ability, though in fact you have to sometimes wonder about them. For example, like this whole—[A male enters; tape stopped] One thing I was starting to tell you about in the car, was this issue of tunneling and the business of accident. You know, some things pop up that you’re not necessarily looking for. When Leo Esaki reported tunnel diodes, a lot of us had worked with various native resistance effects in semiconductors, so this is an interesting negative resistance. [Telephone call; tape stopped] When the tunnel diode occurred, a lot of us had worked with the materials. We’d worked with negative resistance and we realized that we could do that. So when I’m in GE and I have some colleagues who were quick to take a look at germanium. But at that stage of my life, I know as much about silicon as anybody doing any silicon device work. I know that I’ve been into silicon in many ways and forms (begined)? others, and I’m going to see what a silicon tunnel diode looks now.
Ashrafi:What year are we talking, now?
Holonyak:We’re talking 1958, 1959. We’re talking way back when, and still in the ‘50s. So quickly, I had quickly put together a silicon tunnel diode, and I had silicon tunnel diodes right away. Now silicon is higher gapped than germanium and it goes up to higher voltage. But there are—
Ashrafi:When you added a fourth?
Holonyak:Well, yeah. When the applications engineer saw some things that could be done. That turned out to be useful, so there was an engineer—
Ashrafi:Oh, no! [tape interruption]
Holonyak:Germanium has a nice negative-resistance valley. Silicon goes out to higher voltage, but doesn’t have such a nice deep valley, and there are still problems there with that, to this day. So at some arsenide point, we look at gallium arsenide, because it’ll go out to higher voltage yet, and has a real deep valley At any rate, there’s a partner here in Syracuse who’s looking at germanium, and I’m looking at silicon. And at Schenectady Bob Hall and Jerry Tiemann are looking at germanium.
Ashrafi:Schenectady is Bell?
No, Schenectady is GE’s research lab, and Syracuse was their electronics laboratories. I’m in Syracuse, and my colleague Bob Hall is in Schenectady. He’s 120 miles away, 130 miles away. Sometimes we visit one another and we can talk to one another on the phone and everything, but we’re two separate labs. Hall, I consider another figure like one of the big figures. If I had been giving Nobel Prizes for semiconductors, after Bardeen, Brattain, and Shockley, the next guy who would have gotten the Nobel Prize would have been Bob Hall, my colleague, later, in Schenectady. Bob did some tremendous things. But at any rate, I’m with a guy by the name of Lesk in Syracuse and he’s doing germanium and I’m doing silicon.
And down the hall from us, there’s a guy I know, a Chinese American, Hsu Hsiang, and Hsu’s got a helium tank. They’re transferring helium for maser reasons and all that, and he’s got this tank and there’s always some helium in there. So I can put some of my tunnel diodes on a ceramic rod with some leads leading in there and slowly lower it (the tunnel diode) into the tank and take a look at my diodes at helium temperatures. And I know that they’re going to go down to helium, because the degeneracy of the doping makes the behavior very metallic rather than just like a semiconductor. If you still have a semiconductor with semiconductor properties but you don’t have the freezing out of the conductivity, you can get down into very low temperatures.
I know this and I want to see how my diodes look at helium temperatures, way down there. So we go down in there and we look at the I-V characteristics and wow. The characteristics, instead of looking like a normal tunnel diode, the characteristic is beginning to flatten out and become less conductive, then it hits something. It hits, something that makes the slope go up and there’s a notch. And you go up a little bit further in voltage, and there’s another notch and another notch and another notch. Well, germanium isn’t doing this, but my silicon is. So we wrote a paper, send it to Phys Rev Letters, and we didn’t have it sorted out whether there were some split off bands involved or whether there were some phonon effects.
What we did at the front end of this was a wee bit sloppy in not checking that all out carefully. But it was a stupid referee, because if he had been more knowledgeable about phonons, he could just written a paper right behind us and said, “Look at what these guys have found. They have found inelastic tunneling.” You see, in a tunnel diode, the electron is sitting in the conduction band, off in K space, over here. You’re raising it up in energy to tunnel it through the barrier over here, to a place in K space where there’s a hole, but sitting at k=0. The hole is over here in K space at k=0, and the electron is over here, not at k=0. So in the tunnel diode, when you’re tunneling across from where the electron is to where the hole is, you can’t conserve momentum. K value. We’re seeing directly, on an I-V characteristic, the phonon spectrum, directly ( in voltage). Later, a guy coming through Schenectady mentioned, one of them, had the phonon not been known, it would have been a Nobel Prize for discovering the phonon, because you’re looking at the phonon right on the screen.
Now, Lesk and I didn’t start that out very well, and Phys Rev Letters sent this back rejected. In the meantime, I’ve talked to Bob Hall, my partner in Schenectady, and Bob’s a wonderful experimentalist, a smart man, a good physicist, and he’s got available to him also Henry Ehrenreich who’s a solid state theorist, now at Harvard. And Henry knows the phonon and other literature well. So the puzzle is, why do I get this in silicon, and my colleague in Syracuse, Lesk, doesn’t get it in germanium? So Hall figured out, “Oh-oh. The doping that Lesk was using was too heavy, smeared the band edges and screwed up the structure enough that it smeared things out enough that he didn’t see it. Hall changed the doping, fixed it up a little bit, and there it is. So we wrote a Phys Rev Letters in 1959, which is the first demonstration of inelastic tunneling, plus it is the initial demonstration of I-V spectroscopy—of using I-V characteristics for spectroscopy of whatever fundamental thing you’re looking for in tunnel processes. This is our work. Tiemann, who was there with Bob Hall, who was one of Schiff’s students but liked to do electronics, Tiemann made a derivative machine, and of course on a derivative machine you see much fancier steps. Derivative sharpens things; integration smoothes it. And Bell Labs one-upped us, and does a second derivative. So some people will be quoting Bell Lab stuff because of the sharpness of second derivative data, but the spectroscopy starts with me and Hall.
The Physical Society still owes me and Hall for that, because tunneling spectroscopy starts with me and Hall. Oh, I have gotten so many awards, who cares, and it doesn’t really bother me. But from the standpoint of a fundamental thing that you didn’t set off to get but got, we’re the ones that instigated tunneling spectroscopy. Now, that later, was used by Giaver in the superconducting tunneling. That came and went from us to Giaver. Giaver’s in Schenectady and the group he was with sort of, he didn’t have a PhD yet, and they sort of point this Master’s level kind of guy, “Go ahead and take a look at these things,” and all that. Little did they know that this is going to be the spectroscopy that shows the gap and so forth in BCS. I think it was Bardeen that was responsible for that nomination. Now you throw the Josephson Effect on top of that, and that pulls it into the Nobel Prize category, and Giaver, who made superconducting tunneling measurements, gets pulled along with that, and Esaki gets pulled along with that because semiconductor tunneling, the stuff I’m talking about, preceded the spectroscopy that comes later with tunneling.
Now it wasn’t Leak that saw the inelastic tunneling. It was my silicon tunneling diodes that I still have in a filing cabinet on the other side of the wall, some of those first traces. There’s the origin of tunneling spectroscopy—not planned, we’re not looking for it. Then later, Keldysh, and Russia and the others, France and Germany, realize that the—they had talked about tunneling effects, but not in PN junctions. See, no one had said, “Take a look at a PN junction and you will see this.” It’s funny how one thing can lead to another thing, can lead to another thing, can lead to another thing. There’s no way that anyone can take that away, that Hall and I were the two guys that were the first to see inelastic tunneling and now it’s everywhere. It’s a standard technology of low temperature physics, tunneling spectroscopy and all that. That comes from essentially the business of making devices and realizing that we could go down to helium temperatures and then, lo and behold, look at what the heck you’re looking at, something you didn’t plan to be looking at, but there it is. Tunneling spectroscopy came from my silicon tunnel diodes.
Ashrafi:Interesting. Did we cover everything you wanted to say about how Moll’s group was organized?
Holonyak:Well, I don’t think there was anything special in the organization. We were embedded in a good environment, because obviously—
Ashrafi:Not particularly special. Just how was it organized? In what way was it different from how Bardeen organized? Or now you’re in Industry.
Holonyak:Well, I wasn’t even aware that there was any organization as such. That is, it wasn’t a structured kind of an organization. It wasn’t something where, “Now, if you do this, and feed this to the other guy, then we can make this next step,” and all. It wasn’t like that. It was more or less the kind of thing you do as a graduate student, where you’re working on a problem and you find something and it’s an opportunity to go find something else. Or maybe some other person can also start looking at something. But it wasn’t structured in the sense that, “We’ll do this, and this will lead to this, and then it will lead to this.” It wasn’t like that.
Ashrafi:Were you aware at the time of the politics involved in picking the research projects and pushing silicon?
I was aware of some of it, but I wasn’t aware of all of it. Look, if we go back in this thing, when I was a student, and I told you here I am trying to learn about semiconductors and I’m out of sequence with the quantum mechanics (courses). I should have really had all of that before I’m trying to do this, but Bardeen’s teaching a semiconductor course January of ’52. I can’t say to him, “Hey, John, let’s wait another semester and I’ll do Quantum Mechanics and then you teach the course.” I mean I do it as it happens.
So we don’t know. None of this is laid out according to some structure and plan. And so when I’m in that part of my life, I’m just trying to learn as best I can what this stuff is, what the ideas are. I’m not far enough along, my knowledge base isn’t big enough so that I can stand back with any perspective and see what it is that we’re really looking at. And I don't think semiconductor devices were far enough along that the world of electronics could look at it and say, “Well, do you have any idea what this is going to be down the line?” Nah. If Bardeen were alive today, and you said to him, “John, could you see coming the “chip” and everything else, and everything else?” Or that, “Your first student and the spin-offs from that would be making a transistor into a laser?” John would slowly shake his head. He’d smile. But he would say to you, “What we knew with our first work was that it was important.” In other words, he realized and knew there would be more that could be obtained from it, and it’s the opening to further studies, and that more and more of this will finally guide you, guide you, guide you. But could you see where it’s going to go? No. Shockley, if he were alive, oh, yeah. He could see it all. But you know he’s wrong because you can specifically point to mistakes that are serious mistakes that indicated he didn’t know. He couldn’t know all of that. None of us are that good.
It’s just not humanly possible. It’s one thing you’ve got to love about Leggett. When you’re talking with Leggett, Leggett is obviously a very bright, creative, very worthy Nobelist. But at the same time, Tony will be the first guy that if you push him too far and say, “What about this, this, and this?” he won’t be so sure about what the next answer is and whether it’s available or not. And that’s very much like Bardeen in that sense. Shocked that he was totally altogether different—totally different. But if I go back and I look at this, and I say, “What’s happening?” I’m just trying to learn this stuff, get to the stage where I can do something with it and all that, and it’s only little by little, as time is elapsing—as it would go further along and further along, I’m beginning to see that there’s something more in all of this; there is a dynamic.
There is a funny thing going on with people. Maybe it’s friendly in some of the arguments and guesses; maybe it’s unfriendly; maybe there are some nice things; maybe there are some not so nice things. For example, it’s later, and I’m at Bell labs, and I hear a few stories about business with Shockley. Then I hear all of Brattain’s stories, Brattain who knows that I’m John Bardeen’s student, and he knows me, and they go, “Come here; I want to talk to you.” And he’d tell me something, and little by little, I’m beginning to get more and more and more of the picture. Then at some much later point in the spring of I think of 1980, Bardeen comes into the lab, and he asks me if I’d seen Electronics magazine. I said, “No. I haven’t seen it.” MALE: Is this the story you told in the Nebeker interview?
Holonyak:Right. And see, this picture there (on the wall), I have other versions of this, and it is in the magazine.
Ashrafi:There’s a little one here.
Holonyak:Yeah, that’s right. But in the Electronics magazine, a student went and retrieved it from across the street of our teaching building, and I said, “Don’t ask any questions, just bring it.” Bardeen are going through it and going through it and going through it, and he’s as far away as you are, and the magazine is here. We get to that picture, and he says to me, “Boy, Walter hates this picture.” And I looked at him, and I said, “John, why? Isn’t it flattering?” And I can tell he’s provoked and irritated because he knew that I knew Walter as Walter. And that Brattain I ought to know about Walter that it has nothing to do with flattering picture or anything. See, normally, if he didn’t agree with something, his head would go slowly, like this “No.” But in this case, it went like this rapidly, and he made a face at me.
Holonyak:Yeah, and he made a face. And he says, “No. That’s Walter’s equipment and our experiment, and Bill didn’t have anything to do with it.”
Ashrafi:Shockley is sitting…
Sitting there like it’s his stuff. So I knew that at that point I couldn’t say to Bardeen, “John, stop. Let me write this down.” I knew that would be stupid because he’d say, “Wait a minute.” We’re merely talking to one another, and I’m not a lawyer creating a record or something. So what I said to him was, “John, how about signing your picture here in the magazine there?” I knew exactly that—I didn’t need to know because the reason for that signature was the fact of what he had told me, and that would immediately remind me of what happened. And so, he did. It was years later, and we’ve got some of these (pictures, good ones), and I asked him to sign it and he wouldn’t sign it because Walter was dead. He didn’t treat Shockley like he wasn’t a partner. He just was separating the fact that Shockley didn’t have anything to do with that initial work that discovered the transistor effect. Now you see, as these little things are happening to me and a piece of the story here and there and all that, I’m beginning to see, “No, the story isn’t the way Bell Labs tells it.” It isn’t three guys who are in harmonious relationship everyday, passing information back and forth, and bit by bit by bit look at what happened and the three guys did it like that. It didn’t happen that way. And it wasn’t necessarily all friendly. In fact, it wasn’t all friendly.
Even when he died, we found a letter that Bill Bardeen had found or someone found, and Lillian had it. And I got a copy of it in the filing cabinet over there, a copy of the letter he sent back to Bell explaining how Shockley had pushed him and Brattain into the corner away from the next part of the work that would be the junction transistor, and all he (Bardeen) could do would be to work on the point contact and not be able to touch junction stuff and all that. They knew that the junction was worth going after, but if Shockley ran the effort, as he did, he could push them, (Bardeen and Brattain) aside and do the next part, then obviously he will make the next contributions, and he did. He was a good enough physicist to get it right and to do that. See, as a beginner, you do not see the real dynamic. You don’t really know what the arguments are that are going on. I wasn’t sitting in a room with John Moll when he was working on Jack Morton about stopping us from working on the diffused transistor to go ahead and try to duplicate in silicon what the Philco people were doing (on germanium via jet-etching), but I can remember being…
Holonyak:Yeah. Well, they were doing it on germanium, but Morton was worried about us working with silicon. We’re getting behind, and we’ve got to use some of our fancy methods to make devices—to make a competitive device—and not be concerned with what we were doing. And we knew he was wrong. I’m arguing with John Moll that “No, Jack’s got it wrong. And we’ve got to keep going.” You reach a stage where you’re almost ready to rebel, but I don’t see the actual argument that he has with Morton about this.
Ashrafi:So you were converted. When you first came in, you wanted to pursue germanium?
Holonyak:Well, no, I didn’t want to pursue germanium. I can build this new device, PNPN Switch, I can build it quickly the first version—it might be ugly, but we could see how it works. And John Moll’s reaction is, “No. You’re going to waste time on that, and we want the device that doesn’t leak, which is the silicon device, which is a switch that’s off when it’s off, and let’s concentrate all of our time and energy and effort on that one.”
Ashrafi:And you were pressed for time because they wanted it for a telephone switch?
Holonyak:Yes, but it’s not pressed for time so much as the world marches on, and why waste your time on the thing that you know right at the beginning is a lost effort because it’s too leaky and it’s not off when it’s off, and when you think it’s off, it’s leaking. And it’s not a good switch.
Ashrafi:So you had to develop new techniques and new technologies for working on silicon.
Holonyak:Yes, because the kind of technology that was working fine on germanium are not technologies that are going to work very well on silicon, and the technology that you want to use, which is to diffuse impurities into the material to make the junctions to metallize all of this, not to mention foresee or anticipate anything like the oxide, a lot of good things are in the cards by going into silicon, but you have to go there because you’ll never make a switch on germanium. So let’s not even try to make a switch in germanium; let’s get on with making the thing that we really need, which is silicon.
Ashrafi:So developing these new techniques for silicon was your major work in those one years at Bell? Is that right?
Working on the PNPN Switch. When we were working on the PNPN Switch, we were inherently building transistors. So we were inherently building diffuse transistors and PNPN switches and developing the technology of building diffused junctions in silicon and the metallization, which is all part of how things are even to this day. You see, up to that point, a technology comes into existence; it yields something and goes obsolete. And another and another and another. And once you arrive at silicon—once you arrive at diffusion of impurities, metallization of contacts and that structure—once you get to that stage, once you hit the oxide, you are never going to leave that. That becomes invariant and becomes part of the base of what silicon technology is.
Now at this stage, it’s pushed sophistication way beyond the beginning, but the beginning has never been superceded. It hasn’t been replaced, rejected, and made obsolete in any sense. It has been made more sophisticated, more developed, more involved, fancier. Like when Milton (II of I) was talking about Leo Intel making a dry oxide process. Yes, in the beginning we used a wet oxide process to lay it on the crystal and do the pattern defining and all that. In the beginning, we were not even keeping the oxide on to use it, but we know that that’s a possibility. In the beginning, you don’t have vision—people can pretend to be visionaries; they can pretend they see all this coming. They’re lying. You can’t. There’s no way to do that. You can’t for many reasons, one of them being there are a lot of smart people that contributed to the science and technology. The science and technology becomes a lot bigger than any one individual. And as these things go further and further, the whole thing becomes more than any one person’s contribution. And if one person claims it all, he’s a liar.
Ashrafi:So how did it go in those one year at Bell Labs?
Holonyak:Well, I was there only one year. It was one of the most fantastic years of my life because who knew that—See, that’s why I said when I decided, “No, not GE. I think I’ll go to Bell Labs,” little did I know that John Moll was as good as he was. I didn’t know him that well in the beginning. I knew who he was, but I didn’t know he was that good. And I didn’t realize we would be right on top of a major new technology that would become fundamental to the field.
Ashrafi:So it was successful, your developing [inaudible]. You got a lot done that year.
Holonyak:Oh, yes. Yes. And you see, it’s a good time in my life because you’re healthy, you’ve got a background to work with, you’ve got some speed and you can do things that you can’t do as you get older. You just can’t attack problems that well later. In fact, Bell’s very clever, very smart. When they were in the era of AT&T and a big monopoly and built-in budgets and all that, they could scour and go to the best places. Goldey, whom I mentioned, was from MIT. Where do you think Noyce was from? He got his PhD at MIT. These guys were getting picked up from the best places. You got them at 25 and 26 and 27, at the pinnacle of their speed and ability. If you’re Bell Labs, and you can get the next one to ten years out of that person, wow! You’ve gotten the best part of the productive life of a physical scientist. And you can understand why it was very difficult for anyone to compete with them. If you look at when the laser happened, at the various groups here and there, there were five people, four people, six people, seven people…
Ashrafi:You mean laser diode?
Holonyak:No, just laser, period, when it happened. If you look at what Bell’s got, they’ve got 50, 60, 70, 80—they can come in with a lot of talent that they in essence have in place already. And if the people aren’t on a really particularly hot project and something hot pops up, wham! They come in, and there’s going to be a very fertile period there at the beginning of discovery and work, and they’re going to walk off with the biggest part of it because they’re going to have some of the best people and more of them and right in the middle of something new, so even if it’s the first discovery somewhere else, by the time a couple of years go by, people will think it was a Bell Labs thing because of all they could do with that pool of talent (“stored” talent ready to go) that they’ve got there.
Ashrafi:So now it sounds like a big difference between research in industry that you’re describing and research in the university is that whereas research in the university is dependent on the vision of the leader of the group, the corporation could organize a lot more resources much more quickly to react to the ??? ???.
Holonyak:They may not be doing something real red hot and fundamental. You may have some maverick in the university that’s a bigger far-out thinker on something. But if wherever something happens that looks very good, this massive effort can come right at you, and they can essentially take it, like a herd of locusts. For example, let’s look at what’s happening right now in electronics in China. Wow! We’re talking about a monster there with a lot of talent floating around in there—a lot of able people, a lot of talent. All they have to have is access to something that works, and now that they know what works, they could very well put more of it out in the world than anyone. And then what? You see.
Ashrafi:But you guys didn’t publish?
Holonyak:Yes, we published one important paper that’s considered a classic, which is the basis for the whole thyristor field—this whole business of the PNPN Switch. We’re the first ones to make working PNPN switches in silicon and know that we’ve got a thyratron and set the basis for doing that. And that’s fundamental to the whole business of the thyristor as it exists today. We had another paper we were ready to give in Washington, and I don’t know why it was dropped. For some reason, something internally that they didn’t want to talk about yet or something.
Ashrafi:So were there restrictions because of research secrets or trade secrets on what you could publish?
Holonyak:There were in the following sense. For example, I said that when I went in the Army (Oct 1955), the Bell Labs lawyers said to me, “I could talk about anything and everything that I was around, but I could not talk about the oxide and the oxide masking.” They sat on that for two years while they got their patents and other things together. They new it was important. The only thing they missed on was how important. In other words, had they realized how critical this would be to something off in the future, the chip, they would have never licensed it at small rates; they would have probably exerted more control in how they did all of that. They knew it was valuable, and they sat on it for two years and did not report it.
Ashrafi:What about the memos that you didn’t publish? Would those have been published if you had been in a university setting?
Holonyak:Yes, sure. And this one on the design of this class of transistors and silicon, that would have been published. There were some other probably transistor things that we would have put out and published.
Ashrafi:Did you have a sense that you weren’t in academia anymore? That it was a different kind of ballgame? Or not?
Holonyak:Well, it was sort of a joke. I can remember John Moll saying, not sarcastically, but sort of cynically, sort of laugh, and say, “Well, this little outfit in Texas…”
Ashrafi:…and Western Electric?
Well, Western Electric was the manufacturing arm of the Bell system. And presumably, something that had been discovered, put together, worked out at Bell Labs, would wind up in Western Electric for further development and manufacturing, then go off into the operating systems and be part of the vast Bell system. But Moll knew that what we’re doing is valuable, but he started chuckling, and he said—Well, it’s years later, and he was at Hewlett Packard; he was on the West Coast. Actually he was at Stanford, and he was an advisor to Dave Packard at Hewlett Packard. And Packard was starting up a start-up called HPA, Hewlett Packard Associates.
They had contacted me to come to be one of the principals. John Moll is sort of laughing and saying, “You know, you could always beat these big outfits because they don’t have the flexibility, they don’t have the speed, they don’t have the drive to go be different and do something different, and all that.” I’m still at that point a captive of places like GE and Bell Labs in the mentality of thinking that you need a nice lab to be trying ideas, and I’m not thinking the way these entrepreneurs think, about, “Boy, look, this is real hot. I’m going to go make it and sell it.” You see, the idea of the making and selling isn’t a big motivation for me. Milton is very prone, if we get a few little steps further in what we’re doing, to make it into something that becomes an enterprise. That’s a new way of thinking, and I think around the engineers at MIT with these spin-offs, they think in terms of a company coming out and all that.
In a lot of our minds, what we really do is enjoy getting into the middle of the work that leads to something, but not necessarily being the ones who convert it into something that is out in the world of competitive electronics hardware of some kind. At this stage I don’t even worry about that, but that’s not where I’m coming from, anyhow. You see, if I look back, I see some nice things that paid off. And I did learn in my contact with what I saw at Bell and at GE to understand that some things are not just nice, publishable information. For example, the phonon-assisted tunneling, the inelastic tunneling, that’s a Phys Rev Letters. And it’s a real nice piece of science and it starts something very important in the world of science. That one doesn’t produce any money. Now the thyristor, which is a different kind of thing, produces money.
It’s controlling motors and all kinds of stuff like that, and running the power line and all sorts of stuff like that. In other words, some stuff can be a product. And I could see from my background in industry, what was the kind of property that you didn’t just put into the world, but you got a patent. So while I’ve been here there are a couple of things that we have patented that I’ve been responsible for that have paid money. In other words, the MIT people would be happy with me doing that because, in fact, yeah, it would be one of the kind of things that is part of the business of bringing in some money. See, we all understand that—even people around here frequently couldn’t understand Bardeen and wonder, “How could this man be such a deep thinker and physicist, and still have concerns sometimes about jobs?” See, John knew that we couldn’t have the benefit of a place to work in a world of ideas, and that be handed to us because that costs money—someone’s going to pay for that. Now where does the other guy get a livelihood out of the fact that we’re doing what we do? See, it’s one of the great problems with something like the accelerator in Texas.
You finally get to the point where even a big country that should be doing some experiments like that finds that it has to pause and wonder whether it can do it or not based on its problems paying bills. And in the world we’re in, the tunneling spectroscopy is a wonderful thing, but it doesn’t pay any bills. On the other hand, some of this other stuff pays a lot of bills, makes a lot of things possible. So there weren’t strict boundaries between these.
Ashrafi:So after one year at Bell, you go to the Army.
Holonyak:I can’t talk about everything that I saw there. There’s something yet that I want to write up that involves a Cold War thing that was real interesting. I don’t know if Uncle Sam would appreciate me talking about one item that I wound up doing that I keep chomping at the bit to put down on paper because it did something very interesting in the world of ideas and it does touch the world of clandestine stuff. But it involves the world of physics too.
Ashrafi:Can you tell me what in general you were doing? Where were you stationed?
Holonyak:After basic training I was at Fort Monmouth.
Holonyak:New Jersey. It’s the Signal Corp’s primary home at Fort Monmouth, where they do their electronics and all that. And so I was a GI there. Then I was shipped to Japan to a so-called signal supply center, which was a part of Yokohama on one edge that was not bombed out because there was a sea-plane hangar that was next to a piece of artificial land where they held American prisoners of war, and if we bombed the hell out of their sea-plane hangar, we maybe kill our own prisoners. So at the end of the war it’s left standing and it becomes a major facility as a big warehouse to handle all the Signal Corp stuff: batteries, equipment, radars, whatever electronics, for the whole far east, for Taiwan, Formosa, South Korea, Okinawa, all of the military in the far east. But a lieutenant who was on that base who had access to a lot of the supplies had a little clandestine activity there that was in operation, little fiefdoms in the intelligence community—a fiefdom here, a fiefdom there, this group, that group, and they all compete and they all hate one another and compete with one another. He had a little group that was an electronics group, and he heard there was a PhD on the base sent there, and so he got me transferred from that unit to his unit. So I was with this group of people for I don’t know how many months and got involved in one job that was sensitive enough that at one point they figured I had given them what they needed and all that, and so then they sent me back to Fort Monmouth. So I finished up at Fort Monmouth, and then out (discharged), and then to GE.
Ashrafi:How come you didn’t go back to Bell?
Holonyak:It got complicated with various things. I consider, first of all, the New York area too crowded.
Ashrafi:Oh, I forgot, you’d gone from Southern Illinois where you’ve been, out to New York.
Holonyak:Yes. And also, when we had done some of our arguing about devices and all of that, there was arguments where Jack Morton got pretty hot and heavy. I thought, “Oh, I got some good colleagues and friends up in General Electric, and I had a good offer from them, and now’s the time to look around again.” I decided I liked what GE had going for them and I went to GE. When I got there in Syracuse, I realized, “My God, I’m in a group, real nice man that’s head of the group, but he’s not a person in the field. He doesn’t know anything. Nice person, but he doesn’t know anything.” For a couple of weeks I was panicked and thought, “What have I done? I’ve done something stupid!” Then I get busy working on the stuff that we’re working on, and one thing leads to another thing leads to another thing leads to another thing, and I couldn’t have wished for six better years.
Ashrafi:So you didn’t have a supervisor anymore.
Holonyak:Not really. See, he doesn’t have anything to do with what we’re doing. As a matter of fact, when that would come up, his reaction would be, “Well that’s why we’ve got you guys hired. You’re the thinkers and understand these things and all that kind of stuff.”
Ashrafi:Was he a scientist in a different field, or was he a businessman? Do you remember his name?
Holonyak:Yes, Harris Sullivan. He had some kind of a nondescript technical background, but God only knows in what. I don’t know. I had never encountered the man before. Nice person. Nice guy.
Ashrafi:So who were your colleagues in Syracuse?
A guy by the name of Arnold Lesk, whom I had known here as a graduate student who was ahead of me, and then he left GE and wound up at Motorola and then I don’t know what’s become of him and Motorola over the years. Then there were some others. There was a really brilliant circuits engineer, Pete Sylvan. Every time we could make a new device, Pete could make ten new circuits that would be ten more patents. Wonderful guy. And some people in the rectifier department, one of whom is the cousin of this Senator Gore.
You see right there, that black book there, Semiconductor Control led Rectifiers. I wrote that book with the guy I’m talking about, Gentry Gutzwiller was an applications guy (circuits resin) and Gentry was a device guy. Gentry is the cousin of the Gore that was in politics recently. This Gore that you see, his father was a Senator, and Gentry’s mother is the sister of that Gore, Gore #1. The Gore you see is Gore #2. Gentry is his cousin. Gentry was a good device engineer, and he and my other two partners were responsible for this book, which is the first book ever on thyristors, and was quickly translated by the Russians, the Poles, and printed in various places (including India). This is really a spin-off of the business of the PNPN switch. The guy who was the manager of this rectifier department that these other guys came out of was a good electrical engineer, good manager, and he was supporting a lot of my work because I was helping these guys with this. When I got there, they’re trying to make the thyristor, and they don’t understand it. The semiconductor control led rectifier, which is a silicon PNPN switch.
They’re in the middle of building high-powered versions of this, high currents, high voltages. See, in Bell Labs, we’re trying to build a low-voltage, low-current device for telephone switching. These guys (GE) want to make a thyristor that goes into the tens and hundreds and thousands of amperes and up into the thousand-volt range. They want to build a big power device. And they’re fooling around with something that we’ve already conceived at Bell Labs and demonstrated, and they don’t know what the hell they’re really doing. They (GE) don’t understand it. So my job in the beginning, was for the engineer who was the manager of that whole group that was supporting this.
You see, what a lot of people don’t know is there would probably be reasons to check some of this stuff (my Urbana books and papers) and see what pieces are paper are in here for what reason, because they’re worth—Like this guy Ray York, he was the angel that was paying the bill. He was a first rate manager-engineer, the man who wanted a solid state (Si) thyristor. [Conversation with Milton about a picture.] Yeah, see, I like it. You’ve got to have them make some more. That’s a nice picture, Milton. You see, you get to room temperature, you’ll be telling stories about all this stuff (the “transistor laser”) the way I’m telling stories about John.
MILTON:[Inaudible side conversation]
Milton’s been looking at this because of some issues of stuff that we’ve gotten into. See, that’s the Russian translation right here, (our book). I didn’t start off thinking, “Let me get my hands on all these goodies.” That’s not where you’re coming from. You’re just trying to learn all this stuff, learn something, add something to it and all that. You see, I really loved my contact with someone like Bardeen because I never saw in him greed. I don’t think that’s what this is all about, “This owes me something. I’ve got to be a big deal,” or “I’ve got to be rich” or something like that. You just want to try to know what the stuff is and see if you can learn it and add something to it, and just have some fun with it. A lot of it has turned out to be like that.
You see, when you drop a device in a helium tank, obviously the first thing I do when I see that I got something real interesting is I go change everything I know about that device, its contacts and everything else. At first I’m thinking, “Did something go superconducting in there, and is there some peculiar thing going on?” I want to get rid of all of that kind of stuff right away. So what you’re doing is having a fun thing going on about, “What is this, and what is it worth?” and when we’re running the first gallium arsenide phosphide lasers, we know that maybe someone else has got this somewhere else, but not very likely. And so the reason Bardeen wants to stop in (in Syracuse, at GE) and see it, “I got the only gallium arsenide phosphide visible red lasers on Earth.” Now I’m going to be caught—
Ashrafi:This is at GE?
Holonyak:GE in 1962. I’m going to be caught by others at some point, but for a certain length of time, I’ve got myself a little monopoly having fun with something that nobody else has got. And actually, some of my first LEDs are gone, but some of those first lasers are still sitting in here in this box that I call a “coffin”, and there’s little crystals and little lasers on some of these things (on transistor headers, TO-18’s) sitting right here, and actually the first one itself—( (the first visible-spectrum diode laser) and the first III-VI alloy device!)
Ashrafi:What are those dates on there?
Those are 1962, maybe summer ’63, and everything’s fading, but these are beginning to corrode, and I’ve given some away. There’s one here that I know that you can see very nicely the polished face on there. Not that one, that’s not it. I’d have to do some work in looking. But that’s partly irrelevant. The first gallium arsenide phosphide laser is right there, and the guys down in Schenectady for various measurements, and one of the guys wrote on there the “magic one”, “October 1962, first gallium arsenide phosphide semiconductor laser.” But anyhow, it’s in there and Gabriel’s re-photographed some of it. I really ought to have tweezers to get at it. You’d have to go under microscope, but if I tilt that just right, there’s a polished crystal face there, a mirror face. I can see it from where I am. Do you see it? There’s one there and then there’s one on the other side, and if you were to look through that with infrared, you can look right through it because it’s like a window pane.
A window pane, visible light will go through and go from the one plane surface to the other plane surface or propagate an image from this side through the plane surface without scattering, and then do what the index of refraction is going to do and come out here and see it. Well, in this case, you can’t see through this because this is 7200 angstroms, which is red, but it’s way off on one end of the spectrum. But if we run infrared through there, you can see through this from that one polished face through the other polished face, and I actually have a photograph somewhere with a distributor’s infrared microscope, and we looked right through it (like looking through a window pane).
Ashrafi:Is this the very first LED?
Holonyak:This turns out to be also the first LED, because when I’m making this, also, there are different compositions, and this was part of the first package of stuff that we converted into LEDs. The LED version is not stimulated recombination radiation; it’s spontaneous. This one went stimulated, and in fact, I probably ought to do a better job about caring for some of this. “This” is the first thing (another device in another pill box) that became the device which is the wall dimmer, which is a symmetrical switch based on PNPN switches. Then there are some other artifacts floating around in here that are in and of that time.
Ashrafi:That’s fantastic. Thanks for showing me that.
Holonyak:But for me, the greatest fun is to now do what we’re doing with the new devices. Let me get you up here for a second and let me show you something.
Ashrafi:Should I bring the tape recorder?
You can if you want. That diode that you saw there is here at about 71 or 7200 angstroms, and it has at low current, a broad spontaneous spectrum. As you run the current from low to high value past threshold, it narrows and has laser output here. And this is the published figure in Applied Physics Letters which was, 1 December of 1962, Applied Physics Letters had just started and was coming out once a month. If I’d written my stuff up quicker I would have been published a month earlier. But this, that diode that I showed you (in the pill box in the desk drawer) is the diode that produced this figure. Now it’s years later and this guy, Jon Weirer, that’s on that patent, Jon is out there where they’re making the high, brightness diode.
And this one high brightness diode is lighting up this Plexiglas, which is a wave guide. It’s high index compared to the world, and the photons are propagating in here, and anywhere where you put scattering, you can put an image. What the guys did, is they taped this reflecting tape around here so they’re not throwing away the light, but that’s why you see this thing on the edge here. That’s just reflecting tape. But all the light is being made by this one LED. In other words, we’re now in the world of high brightness and various improvements and on and on and on. So what they’re doing is saying, “Here’s one of our high brightness diodes lighting up a picture of the old man’s work, and there’s where it starts. And that diode I showed you generated that spectrum (The red image on the plexiglass display, lit by one red high brightness LED). This is its spectrum.
Ashrafi:That very diode?
Holonyak:That very diode. That very one is this spectrum.
Ashrafi:Wow [chuckles]. That’s fantastic.
Ashrafi:Too bad you can’t mount the diode in here [chuckles].
Or anywhere else. No. And see, and at a birthday, I started way down here and over the years things have happened, happened, and happened, happened, happened, happened, happened. Now you go over here. Now if you were to put this one—See this was a year ago or so that they sent me this, but this stops at 1998 (the wall display) and there was no diode to put in here yet. Now there’s a diode to put in there, and it’s up here. It’s beyond all of the standard sources. It’s up here and it’s still going up. So you don’t get everything all at once. It takes a lot of work from a lot of people and a lot of things with it. You go in there and you’re lucky if you can get one piece of it somewhere. I feel pretty privileged because I’ve seen a lot of nice things and more than one guy has got a right to. Like Let me show you this, too.
This is another thing they sent me somewhere along the way. These are high brightness diodes that they’ve got in some kind of frame, and so I then put it in this box and set it up so that I can pick out what I want [demonstrating switches]. Or I can go over here and turn these on and do the same thing. These are indium gallium aluminum phosphide. And there’s a picture of me and Bardeen where what John and I are looking at is Indium gallium phosphide, which is the precursor for all of this. Anyhow, they send me some of this stuff now, just trinkets, I guess, of the field. There are practical consequences, but for me the fun has always been being somewhere near the front of the idea rather than worrying about getting everything done. It takes a lot of guys to get all of this stuff done. That becomes products).
Ashrafi:So tell me about the differences between Bell Labs and GE, and how you worked and how you worked with people.
Holonyak:They were both fantastic places, but at GE, I realized, after a bit—At first I was scared because I had never worked in a circumstance where my boss knew less than I knew about what I was doing. When I worked for Bardeen, obviously John knew things I didn’t know. When I worked for John Moll, John Moll’s ahead of me sufficiently, there are some things he knows about that I don’t know. I really valued the fact that I was with these two wonderful people, with this wonderful ability. When I’m at GE working for this guy and I realize, “God, this guy doesn’t really know much about this stuff,” for awhile, that disturbed me. But then after some point, when I got “rolling” again, (after 2 years in the Army) my reaction was, “Gee, this was made to order. There are just so many things that I can go up and do.” What I couldn’t do was this. I’m in a semiconductor thing. I can’t decide, “Well, I’m going to do a plasma experiment. But as long as it’s in and around semiconductors, wow. What a playground—what a fun thing to do. (The semiconductor realm is a “playground” for me).
Ashrafi:So were you part of a lab or did you have your own lab?
Holonyak:I was part of a lab, but we were essentially all doing stuff that we individually had interest in.
Ashrafi:Did you supervise anybody?
Holonyak:In the beginning I was just taking care of myself then I picked up a technician. Then at some further point, that lab that was connected with the Semiconductor Products Department, and was getting into financial trouble, so various people were either let go, shipped off here or there and so forth. In the end, everybody was gone except me. And so the whole lab was mine then, and I was left with a chemist. He had a Bachelor’s degree and was almost ready to get his Master’s, and four technicians and a secretary. So when I made the laser diode, that’s what’s left: it’s me and those several technicians working for me, four or five of them, and they were really left over from the old lab.
Ashrafi:When did this happen? When did you become the sole scientist?
Holonyak:Oh, let’s see. I came here in ’63, and we had already started some of this when we were in one building, Building Three (in Syracuse), and we went to Building Seven. In ’61 we were still in Building Three. It must have happened somewhere in early ’62. Actually, when we still had this boss (Sullivan), the whole lab was cut up into units. I had one group, Lesk had a group, Richard Sirrine had another group, and there was another group. There were about four groups. I had one, and I was a supervisor then already. Then Richard left, Lesk left, and that was the end of those two groups. In the crystal-growing group, I forgot what they did with it. Some of the people went to another unit, and all that was left was my particular group and a few support people. That must have been the summer of ’61 or ’62, somewhere along in there.
Ashrafi:Why was the lab getting into financial trouble?
Holonyak:It was probably a lab that was freewheeling, doing anything and everything but not very well focused on what was happening in the product realm. Even though I frequently was doing an experiment that I wanted to do, I frequently was doing something that was related to something that people had product problems with. For example, since I knew about PNPN and switches and all that, there was a certain kind of an NPN transistor which had a rod geometry. And I knew about switches, and I knew the way they were mounting it, that you could put another junction contact on it to a free contact electrodes that they had there and turn it into a four-terminal device that could be triggered like a thyristor, and that there would be use for it. One of the supervisory engineers said, “Oh, always four-terminals are problems,” blah, blah, blah. But it turned out that the line that was making this particular NPN had been a three-shift line (day shift, evening shift, midnight shift). Then it went down to two shifts because it was losing market. Then it was done to one shift and in other words, it’s in the process of dying. When I added the fourth terminal…
So what I started telling you is that this applications engineer (Richard Stasior) caught onto some applications where they could sell the device. People needed it, and they started buying it. So in the production engineering line where they were doing this, there was an engineer in charge of this product, a guy by the name of Tom Mapother. Now there’s a Dillon Mapother here, a physics professor, a solid state physicist. And I remember finding out from this guy, Tom Mapother, I asked him, I said, “You’re tall, like the Dillon Mapother I know in Urbana. Are you related?” It turned out they were related. There was a family relationship. Well, to show you what kind of a small world it is, it’s many years later and I’m back here and had a lot of contact with Dillon. I find out that his cousin that I’m working with on this device (at GE) that increased its longevity. It’s gone from being almost something obsolete.
It’s kept alive longer and produced some jobs. Dillon’s cousin back there in Syracuse, Tom Mapother, is the father of Tom Cruise, the actor. It turns out Cruise is a family name, a grandmother’s name or something and little did I know—it’s a small world. I never met Tom Cruise and I don’t care. I’m not connected to the movie industry, but that just shows you that these peculiar connections. Just like Gentry, who I’m a co-author of a book with and work with and all that, is the cousin of this, of Gore that was running in politics.
Even though the connection is father and mother, brother and sister, Gentry and Gore have the same jaw line, and you can see it. You can see the Gore family in there. And I have to laugh when I think about it. But in other words, we’re frequently not that far away from somebody for one reason or another. The world is big, but it’s small. There’s not really much separating us, and that’s why people had better stop and think about their problems and not be too quick to be denying one another or their existence, because you know, we’re very connected, and watch out. You can’t just dismiss the next person like they’re remote from us. They’re not remote. They’re close. That’s how life is. And that’s an interesting thing about the world of science and technology and all that.
Someone thinks that we’re off in something cold and heartless and non-feeling and all that. Nonsense. Even when it’s done, there are these currents and arguments in things like the Bardeen -Brattain- Shockley kind of stuff, and other stuff. And almost at all levels, there’s something going on that’s humanistic and not just formulas and just cold science. And yet, sure, for those of us who are scientists and engineers and so forth, there is something that we’re involved in that’s constructive and creative, and important to get work done and to get things accomplished and learn, and do what we’ve got to do.
Ashrafi:So can you describe to me either in generalities or more detail, the relationship between the research labs and the products groups at GE?
Well, presumably, in an ordered world, Mecca was Schenectady, where the research is formulated and done and then goes out into the world. Frequently that gets inverted, turned upside down. For example, they would hold conferences there and there were two occasions where I remember speaking at those conferences. I remember, it must have been Roland Schmidt who got prominent in Washington and in science and politics and all that. Roland apologized to me about something that I was reporting, where he says, “You know really, that should have started here, but started with you guys in Syracuse. Thanks for doing that work.” It had something to do with thyristors and switches and all that. GE’s silicon controlled rectifors, later the “thyristor” came form Syracase, not from Schenectady.
In fact, I made a p-n junction in GaAsP already in 1960. My visible spectrum direct-gap, 3-5 alloy (GaAsP) diode laser is the beginning of the “alloy road” to all of today’s lasers and high brightness LEDs. It was the first demonstration of high efficiency photon extraction in the visible spectrum from a diode. In the case of the laser work, I left a device research meeting in New Hampshire in 1962, thinking, “I’m going to go home and work on gallium arsenide phosphide because I see that (visible light). I’m going to make a laser before anyone.” And my colleague, Bob Hall was in the process of making one out of gallium arsenide in Schenectady, and we’re running almost neck and neck. Bob had one idea. I’m thinking of putting my junctions in an external cavity and Bob one-upped me by knowing that, “Uh-oh. The cavity should be the crystal itself.” And that was a very clever idea. But you know, as I said, Bob is a very clever guy, and a wonderful guy, and if I’d been giving Nobel prizes, he would have gotten one before a lot of people you’ve heard about that are not his equal that are Nobelists.
At any rate, presumably, it all emanates— For example, Hall’s gallium arsenide laser is in the infra-red, and the spontaneous version, which is an infra-red light emitter, a recombination radiation emitter, I don’t know if it is an LED because it’s not light that we see, it’s light that you can see with an instrument. That doesn’t really start the LED. The gallium arsenide phosphide I’m doing in Syracuse starts the LED and is actually the first time anybody made anything out of a 3-5 alloy. The first 3-5 alloy device came from me. See, presumably, it should have started in Schenectady, then gone off into the development groups, and into the production groups and all that. But that’s not always the order of things. But sometimes it is the order of things. In an orderly world, that’s how it would be. Just like in Bell Labs, it would start in Bell Labs, go to Western Electric and wind up in use and all that. But how do you know that something that got halfway done in Bell Labs, a better idea emerged in Western Electric, and then it became better. You see, it isn’t totally orderly.
Ashrafi:How was information shared inside GE?
Oh, there were memos that were written, and then there is the mailing list of people who would be in and around related stuff. And then there was also a direct pipeline, because after you’d been there a little while, you’re beginning to meet and have contact with the counterpart in your area of work who’s in Schenectady, or related to you, or the guy from the Lamp Department who used to be in Schenectady or Syracuse but now is working with the lamp people and all that.
So there was a sloshing back and forth of information. Yeah, there is a connection. Sometimes it’s not quite as tight as it ought to be. Maybe it’s not focused directly from, “Let’s take this thing and drive it, drive it, drive it,” and here it is tomorrow, a product. See, under this guy Welch, Jack Welch, it was probably, “If you don’t promise me it’s a product right away, you’re gone.” Actually I saw a different kind of person who was the creative person inside of GE who was actually doing the stuff. In other words, GE is more than just that management structure, even though the kind of management structure they’ve got at the top is considered very effective and very focused on doing GE’s business. But it is a business organization. But at the level of the ideas and the work, it also has this other kind of person who’s still working in the world of ideas. If you insult that person too much, he’s going to leave. At one point in Schenectady, an interesting thing happened.
New York Headquarters let them know that they were over budget and they were going to have to make an adjustment. They decided in New York and upstairs, they decided that instead of working two weeks, ten days, work 90%—in other words, nine days out of ten. A lot of the people felt insulted and their reaction was, “We would have taken the 90% pay, but don’t tell us how many days to work, because that’s the same thing as telling us that our work isn’t valuable.” So Henry Ehrenreich picked up and left and went to Harvard. Walt Harrison picked up and went to Stanford. Who knows how many people, for one reason or another then drifted off. Because when we get down into the world of ideas, we’re going to have to get back to the kind of people who are to some extent, independent thinkers and don’t necessarily follow the party line 100%. I think that business, certain kinds of governments, they’re all totalitarian in a sense. The top wants to think that it controls everything, and it’s an insult to a human being, because it’s making thinking individuals that you’ve got at the bottom useless as thinking people. They’re just ants. They’re not just ants, they’re people—they think. And you can’t run anything that way, governments or anything, totally that way.
Ashrafi:So when this information about what was being developed, either in the labs or in the products division was being passed around, how were decisions made? How were priorities set for what you should do in research? Was that up to you or did you have to argue for it?
Holonyak:Well, one thing that was happening to me in Syracuse, when this guy Harris Sullivan was there, Harris didn’t bother us very much about what we were working on. Then as the top higher up above him began to put more heat on him, then that got transmitted down to us, “Well, they really want more. You’re working with three fives and they really want more silicon work.” And see, that could be an irritation because they knew that I knew about silicon and here I’m working with these other materials. Finally what happened was, at MIT there was a famous professor, J.C. Slater. And that’s where Shockley did his thesis. That was one of the solid-state theory groups in the U.S. Another one was Wigner and the people who generated Bardeen and Seitz and Conyers Herring. Those were the first ones. Slater after the end of the war, had a guy who did an E&M thesis and then came here; Paul Coleman, and he’s retired here. And he had another one, Ben Lax, who was famous at Lincoln Laboratory at MIT. And he had a third one, Len Meier, and Len Lax wound up as the manager of engineering of this Syracuse thing and was my direct boss in the last couple of years that I was there, two or three years.
Ashrafi:After Harris Sullivan was gone?
Well, Harris Sullivan, I don’t know what they did with him. I guess they managed to get rid of him and he wound up with a job in Milwaukee somewhere. What was left of the remnants of the lab were mine, and I report to this guy Meier. And Meier was a rough and tough, and pretty mean guy in a sense. At some point, he says to me, “If I can’t get Ray York (the manager of the rectifier department) to fund my work, and the Air Force (whom I had a contract with, with Hall at the Air Force), there’s the road.
You can leave. You’re not working in what I’m interested in. I’m interested in these silicon things that we’re competing with, and you’re doing this other stuff. And if you can’t get funding for that, goodbye. You’re gone.” Just in those, in more or less, those terms [chuckles]. And he could be pretty tough about, “This is what I want, and you’re doing something else. You’re dangling on a thread. You’re not going to last.” And I pretty much ignored him. I had a lab. We’re getting something done, and it’s funded and all that. I considered General Electric bigger than he was, and that what I was doing I knew was valuable. Now when we made this gallium arsenide phosphide laser, boy, he was beyond himself. He was so puffed up, he can’t believe how the—A pretty big wheel, Charlie Ryan from Air Force, Cambridge, comes through and he stops in Schenectady to talk with them about what’s happening on gallium arsenide.
And Hall’s boss tried to hide a little bit of that as GE-sponsored and not Air Force sponsored, and got Charlie Ryan sort of irritated because Leroy Apker said, “Well, you don’t really own Hall, we do, and he really did that work under our sponsorship.” Charlie Ryan knew that that couldn’t quite be correct. But anyhow, his next stop is Syracuse and he and Len Meier and I are having dinner. Everybody’s had a couple of drinks, and Len Meier’s is expansive about, “Well, how do you like what we’ve.” Well, first of all, “How was your stop in Schenectady?” “Fine,” blah, blah, blah. “I’d be happier if those guys would say that we sponsored that work there. And, well, look at what we’ve done here in Syracuse.” And I can remember Charlie Ryan saying to Len Meier, my boss, “Yeah, I like what Nick’s done.” In other words, what did any of the rest of you have to do with this? And we were the ones putting in support on this. But at that point, Len Meier has changed his tune and I’m doing something very valuable, as far as he’s concerned, then. I think it’s partly a matter of how good they feel or how good they look, or whatever.
Ashrafi:Did you have connections in the rest of GE that you could have relied on?
Yes. In fact, normally it was considered that if something happened, some good thing was made in Schenectady and then wound up in Lamp Department, it would go from in Schenectady, out to the operating divisions and be carried there and become the product. You squeeze on carbon and you make diamond in Schenectady, but at some point, when it’s gotten to a certain point, you ship it to Detroit to the Metallurgy Department and it becomes a Diamond product. In other words, that’s how the flow goes. The flow doesn’t go the other way around—something happens, and you do something fantastic with a diamond and that group goes back to Schenectady, back to Research.
But I was invited by Schenectady to bring my lab, what I had and what I was doing, from Syracuse to Schenectady, which was very unusual. In other words, come back to home base, the research headquarters. Because you see, I’m already out there in a complement with the lab, out in the competing world of products. But my reaction was, for a long time, Bardeen and I had been writing letters back and forth and he had invited me long before, to come back to the University. There was stuff for us to do here. And at some point, his comment was, “You’re used to doing exploratory work in a small group. You could do that with graduate students.” So when that all came up there (at GE), my reaction was, “No. Instead of going back the other way toward Schenectady, if I’m going to pick up and start over anyhow, I’m going to go and see what the University is like.” But not being sure if Universities had caught on yet to what semiconductor work was really like in a lab that was really focused on that, I took a leave of absence from GE. But then I dropped that and I never looked back.
I never, you know, I never regretted it, one moment, the fact that I left. I had six extremely good years in GE in spite of the differences in thought about where devices, which stuff and what to pursue and what not to pursue. In fact, it’s 1962 and at this device meeting, I’m in the cafeteria line with Art D’Asaro of Bell Labs, and Bob Noyce of Intel fame. And Noyce and I know one another. We’ve competed in the same area. In fact, we had a patent interference going, and in fact, that wasn’t settled until some time later when I’m already here. And I’m happy to say, he lost it. It was something at GE about what we had done that overlapped with something that he had done. He and I knew one another well, and he’s arguing with me and Art D’Asaro about how much time we’re wasting, fooling around with 3-5 semiconductors. And Art and I are trying to explain to him, “Yeah, but there’s stuff to look at there and there’s things to do there,” and all that. There is no semiconductor laser yet. But we are already doing some stuff with these materials and know some things about it, and can see reasons to work on that. Noyce, his comment is, he knows that I was in silicon before him, and also Art D’Asaro, and that there are things that we know about silicon and can do with silicon.
And he says, “Yeah, but I’m going to push silicon further ahead of you guys.” Now, he was right. Silicon still had many further things to do. See, he’s still in Fairchild, and at some point, in that line he says, “Nick, I’m going to beat you on this interference.” And I remember telling him, “Yeah, and if you do, you’re back working for Shockley. You’re not in Fairchild anymore. You’re back working for Shockley.” That, “Our work is more extensive and goes back earlier than you think and it’s going to go drive you back into Shockley.” And that stopped him, when I told him that. But at any rate, see he was right. He’s in Fairchild. There is no Intel yet. And look at what’s happened. He was right. Silicon had a long way to go. But we were also right. There was a lot happening that was coming, that no one could foresee totally that’s going to come out of the 3-5’s. It just says that there are fertile places to work, and you can’t be sure that you know exactly where the payoff is and the other guy is wasting his time.
Ashrafi:Was staying in Syracuse an option?
Holonyak:No, because sooner or later that was being essentially shut down. And you know, over the years, not only was it shut down, that whole GE operation that existed in Syracuse vanished. They sold off a piece here and they had some stuff that was doing military contracts, and so forth and so on, and that all wound up getting sold. And by the time we get to Jack Welch, GE is not in TVs anymore, not in a whole bunch of electronics anymore, and it’s gone.
Ashrafi:So tell me about funding while you were at GE. You said you had to go out and get your own money from the Air Force?
Ashrafi:And you had to get your money from another division at GE?
Holonyak:From the Rectifier Department, which made thyristors and silicon rectifiers and all that. Because when I came there, I was the only guy that had worked on some of those original devices at Bell Labs and at General Electric, and had had any background in understanding what it was in the beginning and how the PNPN on switch could make a thing like a thyristor. So I was involved in getting those guys pretty well schooled and into the understanding of how the piece of silicon becomes that kind of device and switches and does the switching thing. And I was back making those and studying them also. Because it’s a negative resistance device, that gets me more broadly interested in negative resistance, which comes up then with the silicon tunnel diodes. And then that gets me involved with 3-5s because there’s some things that the 3-5s will do in that realm. And then that gets me involved in the thing that involves light emitters and all that. And so one thing is leading to another thing, and even though I’m getting some support from Ray York in the Rectifier Department, it’s taking me in the direction of something that will be a light emitter, which won’t be as useful to him as it will be to the Lamps Department. That Lamps Department is now working out arrangements with other people to get back in this, because the LED has become really a lamp, and is beginning to do major things. And Lamp Department at GE can no longer ignore that.
Ashrafi:But is that the normal way that you had to work at GE, that you had to go the other divisions at GE and get funding for your research?
Holonyak:I don't think there was anything normal about it, because one of the buildings, we were in Building Three, did a lot of projects for contract reasons, with the military and others. So we were sort of used to the idea that you got contract support from elsewhere, anyhow. Just like that five and dime transistor at Bell Labs, they were getting support from Army or someone. And Schenectady, even though it was funded from Company money from here and there, you could farm out a piece of something like what Bob Hall was doing because he’s so prominent, and Air Force is so interested they’ll support a piece of it and be glad to do it. And I didn’t really have much trouble getting Air Force money almost all the time, continuously, when I was there. I soon got involved with some stuff involving switches that they were interested in, and when I got into the other stuff, they were very involved in that and interested in it. So I had no trouble keeping a contract going with those guys.
Ashrafi:So what fraction, roughly, what fraction of your money was coming from GE and what fraction was coming from outside contracts?
Holonyak:About 50-50, or maybe even more from the outside. Maybe 40% inside and 60% from the outside.
Ashrafi:And the “outside” was all Air Force?
Holonyak:It was all for me, at that time, was Air Force, Air Force Cambridge.
Ashrafi:And for the internal, GE money, did you have to go to make proposals to other GE managers?
No, no. Ray York was so happy with what, some of what I had done with those guys, on the thyristor, that he just automatically put some of his engineering research money into what I was doing. He just automatically did because I think it was essentially he considered it like pay-back for what I had done for them in getting them into getting a good, basic understanding of what they were doing. Plus, in fact, making the symmetrical switch and a thing called the Shard short emitter, which is incorporated into all thyristors and all that. It’s a thing that allows stability to exist against false switching and makes it possible to get to higher temperature operation.
He was so happy about all that that he knew that I was always onto something that would have some bearing on those guys. So I didn’t have to write any formal reports for him. But I also was very careful. When we hit something good, I would make sure that he got—Like for example, when we made those first visible spectrum lasers, I had the only ones on Earth that you could photograph with a camera. And I had a beautiful picture, and I know when they framed that and all that, I made sure one was delivered to him. In fact, I was there with Len Meier, my direct boss when I gave this thing to Ray York and when we get in the car to come back to Syracuse, Len Meier is all over me.
And he says, “Talk about job security. How come Ray York gets one of these right away and where’s mine?” And when we get back, I said, “Well I thought all of the publicity and other people who handle this, they’re all part of your domain, and they would automatically feed one to you.” We get back to the plant and park, and there was a guy that worked in the building, and he comes running to me later and he says, “Nick, Nick. Look. The big boss. I’ve got to get one of these already put together. He really is bent out of shape. He’s got to have one of these right now,” you know? My connection with that engineering group (Ray York, rectifor dept.) was very good, and I really cared for that manager. He was a very astute, low-keyed guy that was helping you and he knew you would do everything to help him. And it worked very nicely. He’s my model of one of the best managers I ever ran into. It’s easy to do business with that kind of manager, easy. Because he’s not pretending that he has all the answers, and if you can help him he really appreciates it. And he also shows it because he turns around and makes it possible for you to keep doing your work. It’s just really easy to work with people like that.
Ashrafi:Is there anything else about GE, you can ??? ?
Holonyak:No, except that GE was a big payoff because of the things that we got into and what happened. Like this inelastic tunneling and this tunneling spectroscopy. I’ve got a picture sitting next to that Helium tank, and it’s in another box of stuff. And that’s a big deal. The business of what we did with the symmetrical switches and the stuff that became the wall dimmer and all that big time stuff. And then all the stuff that happened with the 3-5 compounds and the light emitters and lasers and all that. Because everything today that you hear and see as a high brightness, a high performance, I don’t care where it is in the laser world, DVD or whatever. I don’t care what it is in high brightness lamps and all that. Every one of them is a 3-5 alloy, every one of them.
Ashrafi:And what exactly does “3-5” mean?
Look up here. Look up here. You see. Look at column three. We go down there, aluminum, gallium, indium, and then column four is the elemental things; carbon diamond, okay? That’s a semiconductor. Silicon, germanium, and form of tin-gray tin. Now we go over column five, nitrogen, phosphorus, arsenic, antimony. Okay. So go look at germanium. We pull the germanium atom out of the crystal. Where is my model? Oh, there it is up there. See? Up there. Okay, you see that’s a diamond crystal lattice. And silicon and all, germanium, those items would all be the same, not black and white as in the model. If I start now, if I go in there and that’s germanium, and going back up here, I pull out two germaniums. I put a gallium in one slot and an arsenic in the other. There’s three electrons around gallium, five around arsenic. That’s eight. Two germaniums is two times four, is eight.
The crystal didn’t know that I’m now beginning to rebuild it, taking away the germaniums and putting in galliums and the arsenics, galliums and arsenics. And that makes it into a 3-5. Now, how do you know that I don’t want it to go ahead on that arsenic side and take out a few arsenics and put in some phosporuses? And I put in a lighter atom for the arsenic and start doing that and substitute, get rid of half the arsenics and put in half phosphoruses. Now I’ve got 50-50 gallium aresenide, gallium phosphide. But in the form I’ve done it, it will just be gallium, arsenic, and phosphorus, gallium, arsenide and phosphide. That’s what you were looking at, in that elevator (the red LED at the place we had lunch). That’s what’s in this LED here. How do you know that when you make an alloy, it’s not disturbed, disordered, full of defects and all of that? When I’m running gallium arsenide phosphide—Hall beat me with gallium arsenide. He says, “Boy, you’d better get that written up right away,” and I said, “Why, Bob? You already did gallium arsenide and got it sent in to the journal. You beat me.” And he says, “But who knew that an alloy would work?” Where’s the data? Where’s the proof? I got the first proof that a 3-5 alloy can be made into a device. Jerry Woodall pretends that, “Oh, well, the whole world is using aluminum gallium arsenide,” and all that. Aluminum gallium arsenide doesn’t come into existence until five years later. It’s virtue is that you can substitute an aluminum for a gallium without changing the lattice size. When you’re doing this, when you change some atoms for other atoms, your lattice…is sensitive, perhaps, to change in size.
Ashrafi:Okay, strains of…?
Holonyak:See, that’s a whole new ball of wax, further and further details. But the point I’m making is that nobody knows that a 3-5 alloy will build anything without being nothing but a mess, disturbed, full of defects, not functioning. In principle you could use the so-called virtual crystal model and say that, well, here’s gallium arsenide on a scale over here, and here’s gallium phosphide, here, and I make the alloy and I grade from one to the other and band edges will scale more or less linearly from here to there. More or less. That’s called Vegard’s Law. Maybe it will happen, but until you do it, you don’t know how bad this material is for some reason, what kind of defects you’ve got in it, how disturbed it is, and all that. Seeing is believing. And as Hall is pointing out, “Yeah, you’re the first guy that’s got data that shows that this stuff in fact works.” So at this stage, every, every semiconductor diode laser, unless it’s a 2-6 or lead salt, but everything that you know about that’s being used in communications and DVDs and on and on and on, is the 3-5. And all the high brightness LEDs and all that, all of those are 3-5 alloys.
Ashrafi:And that transistor you showed me, that is?
Holonyak:That laser diode.
Ashrafi:That laser diode, it wasn’t the first laser diode, or was it?
Holonyak:It was the first—Hall had the first gallium arsenide laser diode. This (the GaAsP diode laser in may desk drawer) is the first 3-5 alloy diode laser ever, in any material. And it’s the first one in the visible spectrum. Hall’s is in infrared; mine is visible.
Ashrafi:And the significance is that you can build materials that don’t upset the crystal structure by using it?
Holonyak:Right. And also the further significance is that you can shift the composition from one material to another material. You can shift it in such a way that you’re also shifting band edges, energy gaps, and shifting properties. And you’d have to go into the 3-5 alloys to be able to do that, and you need the 3-5 alloys to be able to ultimately make hetero-junctions, to make a higher gap material that’s part of a device that’s involving lower gap materials, and to make quantum wells and all that. Until you know that 3-5 alloys work, you can’t do any of that.
Ashrafi:So you couldn’t build devices with Hall’s laser diode?
You could build a gallium arsenide device. But you couldn’t build hetero-junctions. You couldn’t build different wavelengths. You couldn’t do anything like that. You’re stuck with whatever that binary material is because that has fixed properties. Whereas, (composition) the alloy can be varied in composition. I don’t have to have 50-50 arsenic-phosphorus content. I can have 20-80, 10-90. I can have any composition I want and scale my properties. And I’m the first guy to know that you can do that with a 3-5 and make an alloy and get away with it and not have disturbed material, disordered material, essentially a mess of some kind. Or lumpy material where it’s just lumps of, on the average of something that looks like an alloy, but really not very smooth in its atomic structure.
I’m the first one to know that and actually happen to have a working device. And it’s just a matter of a week or two separating me and Hall, relative to that. Actually, there’s some reasons why he beat me. See, gallium arsenide was single crystalline material. My material was frequently poly-crystalline. I wanted to cleave and make a cavity, and I was having a hard time finding a cleavage plane, but I still tried to do it. And then one day, Hall’s boss called me up and he says, “Nick, stop trying to cleave that. You’ve got good diodes. Why don’t you polish the mirrors on there?” \So I devised a scheme different from Hall’s to do that immediately, and immediately I had laser diodes. If I had polished my facets on there right away, I probably would have beat him. Because I figured to beat everybody because I knew I could build visible spectrum diodes and I could see whether I had (laser) diffraction patterns and all that. I overlooked the fact that he could see that with a Snooper Scope. And man, I told you, my partner, my colleague, Hall, was a real first-rate guy.
Ashrafi:Why was Hall’s boss involved?
Because, simply because he’s the boss and he’s following things, and he’d come to Syracuse occasionally to visit, the company visit and all that, and he’d stop in. And I’d showed him my diodes and how bright they were. We could be looking at a diode like that, and I could shove it into liquid nitrogen, and man, you’d pull it out and it’s cold and it’s even better, and more efficient. It’s blazing—it looks like something that’s on fire. And he’d seen that and knew that I had good diodes. And then when Hall’s running a laser, he knows right away, and he called me up from Schenectady and he says, “We’re running a laser,” and I said, “How do you know?” and he says, “We can see the diffraction pattern.” And I go, “Oh, my God.”\ You see it with a Snooper Scope.” And so he says, “Stop trying to cleave your Fabry-Perot facets on this, your mirrors. Polish them.”
So the next day, I proceeded to that. I devised a scheme right away, a simple scheme, and did it, and I got a crystal there and it’s ready to go. That’s the other thing that’s different. All the people doing gallium arsenide, those crystals were grown by someone else. Those were commercial or “on the shelf,” as they say. It was a shelf item. Mine were ones I made—I made my own crystals. And my methods of doing this were very quickly superceded by better methods by chemists and others, then, who once they knew it worked, could go to open tube methods, more controlled systems to grow things and they could make, in fact, crystals much easier and much more controlled than I could. And of course the industry went that way. But they’re not the guys that showed that this stuff works.
So how could I ask for anything better than that? I couldn’t have had six years figured out ahead of time that would be any better than that. In fact, I was a little bit cautious when I had come here and I said to John, I said, “John, I don’t know. Maybe I’ve seen the best stuff I’m ever going to see.” And I’m happy to say, you know, there were still more things to do. I mean, there were just more things to look at and do and get done. And it’s continued all the time. There’s been something further to take a look at. And in fact, at this stage, right now, I’m the only guy left who made some of the first stuff. You see, Milton and I, and Gabriel and Richard, who’s walked in, we have a front cover thing coming out in Applied Physics Letters either the 28th of this month or the first issue in April. It’s on the front cover.
It’s that picture right up there. It’s on the front cover. That is a legitimate transistor that is putting out a laser signal modulated at the same speed as the electrical signal in the transistor. This is not a two-terminal device that you’re driving and modulating with your tricks of circuitry and all that; this is legitimately a transistor running as a three-terminal transistor that is taking the base three combination that is part of the operation of the transistor, going into stimulated emission, and “squirting” a laser signal out of a facet on it. My first paper on lasers is in late 1962, and here we are in 2005, and I’m no further along! I’m still trying to make semiconductor lasers.
Ashrafi:And so could you follow the same line of research when you came to Urbana?
Yes, because I guess in von Neumann’s era, having seen the Los Alamos thing, he realized that materials were controlling and limiting. The many things you had to do to get all that kind of stuff done, you had to work in Materials. And I understand it was von Neuman who had sold the government on the fact that there were going to have to be materials research labs. And Urbana saw a couple of these things go up in a couple of places, and I think maybe got nervous and decided that they needed to pull together some of their stuff here into a materials research lab.
So Bardeen and Seitz, and who all, did that, and there was a materials research lab being put together here. Bardeen was cognizant of the connection with those of us who had a base in electronics, and it would be those of us who are semiconductor people who would do that kind of stuff, and then there were other people doing other kinds of things in superconductivity research and various other kinds of things. So when Bardeen invited me back, he also had indicated that he had some support money from somewhere else. He said, “Don’t go looking for money. I have some money I’m not spending, and spend that. Then, this materials research lab is going to pay for some of your tooling and stuff like that,” and so I immediately was part of the materials research lab. For many years I was, until it started getting into other problems with the—back to this issue of support. The materials research lab here had two sources of support.
One was Atomic Energy Commission, which is now the Department of Energy. Part of it was supported out of that base. That’s where Slichter gets his support. The other part was DARPA support. Then some number of years later, DARPA wanted to get out from under that and gave it to NSF. Then that lasted a certain length of time and then they wanted out, so I think all that’s left now is that original Atomic Energy, Department of Energy part. But the Materials Research Lab still exists, though some of us are no longer part of that and have just gotten other support. So when Bardeen had this other money through the Air Force Office of Scientific Research or something, he wanted me to spend money on that, and I did, but then Sah came, and Bardeen’s reaction was, “Well, Sah doesn’t know the agencies as well as you, and so probably we’ll have to look for some other money.” So I immediately found other money at other places, too. And in the era we’re in now, the guy who probably finds the money the quickest is Milton.
Nobody is quicker to find out that they’ll support this, or this or this, or this. Milton is very astute. He doesn’t go very far away from something that he knows they need. For example, if he’s got the world’s highest speed transistors, he can process information, via his transistor, faster than anyone else. So let’s say it’s a military system, and your military system is stuck in a certain frequency domain, and the one that Milton is building can move around and go past you. You can’t see him because he’s in signal frequency, way beyond where your equipment works and you don’t know he’s there, even. So the military is interested. DARPA’s very interested most of the time in anything he’s talking about that’s high speed.
Now what we’ve unearthed is the fact that there’s another wrinkle in what transistors can do, and can the thing that becomes high speed, super high speed, can you take laser operation that far also? The optical electronic systems you hear about, fiber optics and all that, you can easily get them into gigahertz range. You have to work a lot harder to get ten gigahertz. Now you start to go into the 20’s and 30’s and all that. You start to make a lot of compromises where you can modulate directly. Suppose we have a transistor running at a hundred gigahertz that is also a laser. Automatically it’s being modulated at a hundred—in other words, there’s a lot more to this problem. So there’s reason for these people to be interested in this. The big problem, always I felt is very simple: Don’t promise us if you’re not able to do it. In other words, don’t take the other guy’s money and then fool him, because then you’re going to have to go look again for the money of, someone else and keep lying.
See, you’ve got to be careful in science, if you turn into a liar and you’re always promising something and delivering nothing. I think some folks tend to do that. They get enamored with something else and playing games and being important and all that and thinking they’re doing something and making outlandish claims and promises and there’s no basis for that. But the real basis is do the work and let’s see what happens. And if anything happens, the support will happen. It will occur. What you can’t justify very much is this business of big promises, big promises, big promises. Because someone’s going to have to pay bills, and who is going to deliver on the promises?
Ashrafi:So you come here and you start out in the Materials Research Lab?
Holonyak:Yes. Actually we’re under the aegis, under the umbrella, under the support, but we’re not necessarily in one place in one building. See, we’re in an old building, and I didn’t find that had much to do with anything. What really had more to do with everything was the fact of what your project is and what you’re getting done. Actually many of us were scattered out in different places.
Ashrafi:And this is in ’63?
Ashrafi:And you are still working on 3-5 lasers?
Holonyak:Yes, and related kinds of things that come up relevant to that.
Ashrafi:Well, tell me how you set up. Tell me who you’re working with, how you set up your lab. How does that all go? Now you have to start from scratch, again? Is that right?
Holonyak:When I came, I had an empty room. I got some pieces of equipment that I thought I needed to do what I was doing with these crystals, and—
Ashrafi:From Bardeen’s budget?
From the MRL budget. I’d already made a list of a couple of pieces of equipment that they could order, and I started putting those things in place and started making, started back making some crystals, making gallium arsenide phosphide and making Indium gallium phosphide and going after some more problems that I knew about in that realm. And then just going further, and that leads you frequently to another experiment and another piece of equipment, and little by little you’ve got a laboratory full of junk, some of which should be thrown out.
But by then you’ve got microscopes and electronics and other stuff that will be useful for the next and the next and the next piece of work and experiment. And so one thing leads to another. I started off with, when I came back, with three students. Bardeen had already picked them up. He says, “I’m no longer in semiconductors, but Holonyak’s coming back,” and a guy, Charlie Wolfe, and another guy, George Clark, another one, Chuck Nuese, and then very quickly picked up another one, Stillman, who was the thesis director that Milton worked for. Stillman unfortunately died of cancer about five or six years ago, prematurely. But you get started. Charlie Wolfe and I started growing crystals. Chuck Nuese, I started him on making junctions and devices and all that.
Then quickly one thing leads to another thing, leads to more people, more work. I also got back into some tunneling experiments for many reasons. In the meantime, Charlie Duke, who had been one of Wigner’s latter day students, Charlie was in high energy or something, or theory, wound up then at GE working at some tunneling things. He wrote a major book on tunneling. He got involved in tunneling and surface work, wound up here, and we know about one another, but we hadn’t actually worked together. But then we worked together and had some people working across that boundary. He had a theory student. I had some good experimentalists, doing some interesting things and that led to some, to a lot of stuff. But then also Charlie was finding out more about 3-5s because of the stuff that we were doing with luminescence, so he got involved with us on stuff like that. You see, one item leads to another.
Then he got disaffected about support and labs and stuff like that, and left and went to Xerox. But that didn’t change things for us in what we were doing. We just kept going from one problem to another one. And at some point, I refereed a paper for Applied Physics Letters which dealt with MBEs layered structure of Bell Labs, where they had done a photo luminescence measurement and had shifted the energy up from the band edge, and I knew that was quantum size effect. I knew that from something I knew about that, all the way back to Bardeen’s lab. That’s when Schrieffer was measuring some of the surface properties of GE with the post-doc, there was an inversion layer, and that’s so thin that’s confining the carriers, and they knew it, and Schrieffer had written into some of the papers about the quantum size effect of a semiconductor inversion layer.
And so I knew about that, way back when. So when I got this paper to referee, I realized immediately that, yes, in fact, that’s true. That you could shift the energy up from the band edge because when you start to confine the carrier, it has to go up to a higher confined state. Then I quickly dismissed it, because I thought, “Who’s got a million dollar machine to be making these kinds of things?” I knew right away that meant that, uh-oh, it’s possible to make extremely thin layers into these structures, and that immediately means something big. But I ignored it for a little bit. But then I think about it some more and I realize that you don’t have to grow these things with an MBE machine. You could do that with a vapor phase-epitafier machine like Monsanto’s got, and I consulted for Monsanto in St. Louis.
So I tried to get them to modify a simple machine for that purpose, and one of the principal guys wanted to do it, and one of the higher ones yet wouldn’t touch it because Management had spooked him that they were going to leave 3-5s and not to touch anything new. So they wouldn’t do it, and so I came home and grabbed my graduate student who was far enough along, doing liquid phase epi, and I said, “Here, Ed Resek, I want to describe this experiment to you. We can do this and all that,” and I described to him what we were going to make, and make thin layers by liquid phase epi. Our first experiment worked, but we made the structure a little bit complicated. This was in 1977. And the second one, we simplified a little bit and actually we had quantum size effect running as a diode laser. The first laser diodes with quantum wells are ours, made here very cheaply by a liquid phase epitary process.
We’re off and running then. From 1977, from that point on, we were in the realm of quantum wells and devices. Ours were the first devices to run with a current that are quantum well lasers. Not Bell Labs—that’s me and my grad student. So it’s one thing that leads you to another thing, leads you to another thing, leads you to another thing. It depends on what you’re thinking and what your background is and what you see and what you think can be done. Which is exactly how Milton and I get into the business of realizing his transistors can’t be doing what they’re doing without also having recombination going in on them, and in that circumstance, there has to be light. And now you can go ahead and manipulate that and do something else with it. And yes, it’s correct. And we, indeed, make the transistor into a transistor laser.
Ashrafi:Did GE ever complain to you about taking some of the stuff you did for them, and redoing it here?
Holonyak:Nobody ever said anything, because I don't think they really cared. Because they were in the process of trying to determine what was happening in the competitive world of semiconductors and what their role would be anyhow, and they weren’t really paying that much attention. For example, immediately when gallium arsenide phosphide occurred, a group that Dave Packard had set up, that was initially Hewlett Packard Associates and it later became the stuff that became now his LumiLeds thing that exists with Phillip’s Lighting, they had some people in there that were smart enough right away to realize that gallium arsenide phosphide was immediately the way to do a LED, and they came in right away. And Monsanto, who had been fooling around with 3-5s right away got into that. So immediately Monsanto and Hewlett Packard began to make gallium arsenide phosphide light emitting diodes. And it proliferated. GE had never made an attempt to—they continued what I was doing for awhile, with some other people, but they never really got on top of it the way Monsanto did and the way Hewlett Packard did. Because what I was doing with these things was Air Force supported, it was in reports. So I wasn’t taking anything proprietary out of GE and giving it to someone else. I wasn’t carrying any GE documents away from them or anything like that.
Ashrafi:What happened to your notebooks and records from GE?
Holonyak:They are there somewhere at GE. They have that. I Xeroxed, primitive as the Xerox machine was, I Xeroxed the page of the notebook where I had written up the fact that we had run a laser in 1962 in October. I got that page of the notebook and Gabriel scanned it and tried to fix it up a little bit, because there were some rough spots in it because of the reproduction. I got a few pages like that. But I don’t have the notebooks. They’re, I don’t know what they did with them. They maybe threw them away. I don’t know.
Ashrafi:I wonder if somebody has them right now. When were you married?
Holonyak:When I was at Bell Labs, before I went into the Army, I was married. In fact, my wife and I have been together now something like 50 years. She was a graduate student at NYU in Nursing Education and going on towards a PhD, but then if she got married, she had to return some scholarship money to this hospital sponsoring her in Chicago. So I got drafted in the Army and went off to military camp, and she finished her Master’s at New York University and then went back to teach at that hospital until I got out. And then we went to Syracuse.
Holonyak:No. It just happened that way. No, it’s how it worked. But look, I’ve had a lot of good students, and so I’ve got other kinds of kids. As I told you, eight were members of the Academy of Engineering. One died. Three or four were physicists, and the rest were like electronics engineering people.
Ashrafi:So, do you have a list of your students? You had 16 of them?
Yeah, I’ve got a list. Yeah. I’ve got some stuff in a filing cabinet on the other side of the wall. Bobbi can send you anything that you need later that you contact me about. That list is not up-to-date in the sense that they’ve sense moved around here and there, but yeah, there’s very prominent people there. And my own reaction is, “Look. One of my students, Morley Blouke is the guy that made the big CCDs that are up on Hubble. When Morley made these things, these were the biggest integrated circuits on Earth. They were 2,000 or 3,000 pixels this way, two or three thousand that way, about two inches across—huge, and of great sensitivity. And every pixel worked.
How do you take a piece of silicon and when you’re done processing it, have that many pixels on that big of a thing, no pixel being bad in there, and all of them capable of seeing at the level of one or two photons. Boy, that’s a fantastic thing. His CCDs are in the big telescope in Chile, and I don’t know where all. And some French astronomers have named an asteroid after him. I’ve nominated him for academy membership, and don’t ask me why, but it hasn’t gotten past the politics of committees yet. But I look at Morey and I think, “My God! How can this man not be a member of the Academy of Engineering?” In other words, what I’m saying is there are eight there, but I can see at least eight more who were doing some magnificent things, really nice things, and younger ones who will be heard of before it’s all over. See, if you ask me what has been the big thing about Illinois, it is I had access to some of the best, brightest young people that you could imagine. Well, plus, in fact the years that I had with my association with Bardeen and so forth. There have been some things like that, that are on the grand scale.
There have been some other things that are, ah, you know, the usual academic politics and other crap that’s who needs it. In other words, this is not a pure, pristine, perfect place. It’s not even close to that, but it is a place where there was a lot that we were able to get done. So I don’t regret the fact that, “Gee, the years slipped by, and now what?” I just wish I didn’t, that my arm were in better shape, and my back. But so what? If I started off and said, “Well, could you do any better,” sure you know of some mistakes here and there but you can’t stack it all up so it would be perfect. So generally speaking, I don’t think I could do any better than some of the things that I’ve been able to see, that are really nice, fun things.
Ashrafi:Were you ever tempted to leave Illinois after you came here and…?
Holonyak:Well, I had plenty of offers to do that, including from MIT. There were some of my friends there that said, “Take your pick. Come to MIT or to Lincoln.” And in fact, when the offer to go to Schenectady is put out there, “Nick, bring it all out to Schenectady,” also, the Lincoln guy said, “No, Nick. Let’s bring it all here.” And there have been other kinds of things like that, but not really seriously, because like I told you, this one contact from the guy out East that, “Take retirement and come,” and that’s just a case of saying that I could bank more money. But I don’t care about that. I’m not getting rich from this. That’s not the point. The point is to do what we do and try to get it done, and do something that’s fresh and creative, and it’s a fun thing to do. You don’t start off thinking, “Well, yeah, and in the end there’ll be something that’s on the scale of an ultimate lamp.” See when we were doing this, at some point I got nervous about the fact that, maybe that isn’t a lamp—
Ashrafi:You devised a proof?
Well, you can prove that you can start with a pristine pure piece of semiconductor and you might, for quantum reasons, technical reasons, want to specify that you’d like this to be a direct gap, in other words have the conduction band in K space at the same place as the valence band so that the electron and the hole have K=0 at their two minima. You may do a few things like that for good technical reasons, and then you look at this thing and say, well, I’ll start the problem very simply. I’ll put light in and see what happens. I’ll put in a photon and generate an electron whole pair.
Now if I run this up in intensity so that there’s enough of them (carriers) so that I can talk about a chemical potential for electrons and a chemical potential for the holes, then I can ask myself what do I need to do to be able to access this voltage, this difference in electron chemical potential and the hole chemical potential. Here’s where it proves nicer to be a physicist than to be an electrical engineer, because you know something about thermol dynamics and understand that if you can build me a world of electrons, you can talk about a chemical potential for those electrons.
If you build me some holes you can talk about the equilibration among those holes and talk about a hole chemical potential. And I can talk about there being a difference of potential here that I should be able to access. But when I try to do that, I can’t just put two probes in there and have access to that. So I find out that if I’m willing to dope one side up N type, I can connect to the electron chemical potential. I can keep the light shining on here and implant or something and have access to that chemical potential and come out here and contact that and have contact to those electrons. Similarly, having done that, I can do the same thing on the other side, dope with an acceptor, and have access to the other one. Now, out on the outside world I have access to one chemical potential and the other one.
That’s a true voltage difference. I can arrange a battery around there that bucks that, that’s opposite in polarity. I have a switch, and if I close the switch, nothing happens. Absolutely nothing happens—they are two opposing voltages. So I close the switch, now I turn my external light source off. This thing is still sitting in this configuration, it’s flat band, I haven’t violated neutrality anywhere or anything like that, except now there are no photons coming in now to generate electron hole pairs. Yes, I’m in the same configuration. The battery is now going to keep it in that configuration and in current flows and there it is. Instead of this thing having photons coming in and having a voltage that it can deliver externally, an external voltage is delivering a current and the process is no longer photons coming in, it’s photons coming out. I have been led into a PN junction, just from a fundamental argument. Some tricks of fundamental arguments, but nevertheless, I have been forced to make a PN junction.
I knew that for years, taught it to the students, but never wrote it up and then finally in September of 2000 I published it in the American Journal of Physics. Is the PN junction an ultimate lamp? It is! Now, there are technically things that you ought to do to it involving doping and have barriers and things like that because nothing is ideal when you begin to realize it and all that. But in fact, I have a flashlight up there that a fellow sent me in 1999 inside of that cabinet, and it had a 50 or 60% external quantum efficiency with some tricks of assembly and how they arrange getting the photons out, getting the absorption off of parts of the structure and fixing the geometry, the edges and all that. The damn thing puts out 50 or 60% external quantum efficiency out of a spontaneous source. That’s within closer than a factor of two of ultimate lamp performance. I don’t need it any better than that, and the argument I know is valid. And so I know that the people working on an LED lamp, you could come in and say, “Wait a minute, I’ve got this polymer, this organic or something, and it’s very good and all that.” And I’d say, “Okay, Babak, it may be also another ultimate lamp, but it isn’t the ultimate lamp.” In other words, you are not going to exceed the PN junction. In principle, in principle you are not going to do it any better.
There are technically a lot of things to make and to build to get it all done, and that doesn’t happen overnight. It takes a lot of people a lot of work and much effort to get that to all sizes and scales and speeds and everything else, and it’s going to go on for quite awhile. You’ve got all the colors and costs and everything to get done. But there’s no stopping it now. It just can’t be done. If you said, “Well, is that what you started off to do?” No, of course not. I started off just because these were interesting things we could do and as we were doing them we began to realize something further and further and further. But you can see that it all goes back to understanding what atoms do and stacking atoms and what quantum ideas allow, and then what you can do with materials and electronics and other things, and little by little something comes out that was worth something.
Ashrafi:Did you come here as a professor of electrical engineering or physics or both?
Holonyak:I came as a professor of electrical engineering. I was a full professor, tenured, right from the beginning. I think John knew enough about me to realize that I’d be able to handle it, and so there were no further arguments about that. He knew that what I was doing also involved physics people that I had always been around the physics of this stuff and that there would be physicists interested in not necessarily high energy, but in solids and semiconductors, and that they ought to have a chance to work in this, and he was right. I was not originally in physics, but I had full standing in physics. In other words, I could take and guide physics students and not have supervision from someone else and someone else signing for me or anything. Then with further things, with chairs being awarded and all that, when the Bardeen Chair was created, that was a chair of both electrical engineering and physics, so I have had an association with physics, essentially a professor’s association with them from the beginning.
Ashrafi:Are you the first Bardeen Chair?
Holonyak:Yes. And since then there’s been enough money to give a second one and that’s to Paul Euiat, who’s involved in quantum computing and quantum interference and that kind of stuff. That’s a game for one of the younger guys anyhow; that’s not my stuff, it’s someone else’s stuff. I can understand why physics wanted to try to get a person like that and apparently he’s a man of some attainment in that area. He’s a young guy, and I don’t think he’s got much in the way of awards or recognition, but presumably that’ll come.
Ashrafi:So how has it been, being in both physics and electrical engineering? How is that different for you than everybody else?
Holonyak:When I was younger and faster I could move around quicker and they’d see more of me. Now I don’t care. I sort of live in my corner and do what I do and I don’t think it makes much difference at all right now.
Ashrafi:But if we go back to—
Holonyak:I think if we go back into Bardeen and other’s time, that means back in those years I had lots of contact with people like Bardeen and Seitz and a certain amount with Charlie Slichter and the others. I had much more with Bardeen. John was apt to come to our lab to talk about—when they wanted photographs of the lab setting and all that, he always came to our lab. He was more comfortable in and around our lab than any of the others. I saw him pretty regularly. He either called me about something and I’d go over to his office, or frequently he was walking and he wanted to walk over and talk about something and we’d sit and talk about whatever was on our minds. Then there were other kinds of occasions, seminars. I don’t know how many times I heard him give a seminar on the so-called early days of the transistor. Do you have this?
Holonyak:This is a talk he gave and that they wrote up that you folks ought to have there somewhere which you folks ought to have there, somewhere.
Ashrafi:Thanks. Thank you.
Holonyak:Yeah, you should take that. Maybe there are some things in here, that you don’t know or haven’t seen and ought to take. This is a review article that I wrote. There were some other ones that the pictures in here are not particularly good. But presumably, Gabriel would be able to run off better ones of these. This is in an IEEE Journal, and there are guys maybe you don’t know about. And then, I think Bobby sent you this stuff.
Holonyak:She sent you probably this, I’m assuming. And then this is a two-page version, a shorter version, and then this is essentially a take-off on this, but without the references.
Ashrafi:Which ones? Are they identical, except one’s shorter than the other?
Holonyak:This one is shorter than this one, which has references.
Ashrafi:Do you mind if I take them all?
Holonyak:You can take them all, and then you can see which one turns out to be handier for any purposes of yours. You can take all of these. You can take all of those and then you may run into some other stuff, and say, “Could you send me a copy of…” like, if there’s in that paper’s list there’s something. See, some papers are obviously much more important than others. For example, if someone says, “Well, let’s make some of these papers go away; will you scream very hard about this one?” “No, no, no.” But you hit certain ones and you’re going to scream. For example, the Phys Rev Letters (1959)about the inelastic tunneling. There’s none that precedes that. That’s it. That’s the beginning of tunneling spectroscopy. That’s the beginning. So there are certain papers which are fundamental, that are clearly first.
Ashrafi:How about the memoir, you mentioned it early on, in the first tape. It was a memoir that you wrote about your early work that you were hoping to print. Would be published by the Chemical Society. Do you have a copy of that?
I have that, but I’ll have to send that. And you know why? Because that stuff has some important material (Bardeen, Barttain, Shockley). There’s a lot of stuff about that first silicon work. April, 1992, Physics Today had an article, had a special issue on Bardeen. And a lot of us contributed articles. In 1991, I was recovering from cancer surgery. And that’ll spook you because you realize that you’re not here forever, and I thought, you just can’t have these things (surgery) happening very often, and some time you’re dead. And so I started writing stuff that I knew, that I knew that others didn’t know. So I started writing what I knew about the picture, and Bardeen and all of these things that I knew from the inside. And then either Schlictor contacted me or Gloria Lubkin of Physics Today about writing a piece about John about, “who knew that John was going to die early in ’91 and that these other things (surgery) would happen,” and all that. So I wrote that manuscript.
I knew that they would do what they did, which was to send it to other people to read. One of the guys that wound up getting it was John Moll. The first thing he did, was he knew something was wrong and he asked me about my health. That’s my buddy, and he asked the right thing—not about this stuff, but about how life is treating us. And when we got over that, he says, “You got some powerful stuff in this article.” And I said, “Hell, yes, John,” and we start talking. And I said, “I know you’re bothered. I know you’re bothered about what people think about silicon and how Silicon Valley got started and all that. See, Moll is a little bit like John Bardeen. John would not talk about John Bardeen. John is not going to try to make—see, he knew he was John Bardeen, as I wrote in that article. He doesn’t care if you know he’s John Bardeen.
He knows he’s John Bardeen, and what John Bardeen is capable of and what he’s touched (done). And John Moll knew that silicon didn’t start with Noyce, and crew, in Silicon Valley. It started at Murray Hill, and then he was one of the architects, maybe the principal architect of putting that together. And he knew that I knew that story. I told him, “Look, John. I’m home, writing anyhow, and I got some of the artifacts and things by accident of what happened then. I will put together a front end on it, and then I’ll send it to you, and you put together some more of it.” Well, he always was slow. And over the years, he got slower and slower, and when it was time for him to have his piece in there, there was no piece. While I was doing this and getting the figures ready and all that, I sent him the stuff, so he had my slides. So in 1994 or ’95, somewhere along in there, he gave his a Si talk at MIT, and then later Hewlett-Packard sent me a tape of the talk he gave at MIT.
Ashrafi:How did Hewlett Packard get into this?
Holonyak:Because he (Moll) worked then for Hewlett-Packard. I guess what happened was, MIT cut a tape of his seminar, and he wanted to send it to me, and so they (H-P)sent it to me. I got it somewhere. It’s probably at home in a stack of stuff that’s growing. Gabriel hasn’t re-cut that one. But in fact, here’s something that I’ve got a spare of that you can take.
Ashrafi:Oh, very good. Thank you. I’ve heard about these. They used to be on the website, but I couldn’t find them.
Yeah. Well, there’s another one, but that’s one of them. And maybe we’ll find the other one. But at any rate, Hewlett Packard sent me this tape, and I could see, “Oh-oh, his (Moll’s) memory isn’t working very well.” Some stuff he got wrong, like how the oxide occurred. The accident Frosch had about how—You see, it wasn’t John Moll that was etching the crystals and then taking them down to Frosch’s place for the oxidation and diffusion and all that. I was doing that, and I saw what happened and he didn’t. Then when he reconstructed some of what happened, he didn’t get it right.
So I know there are some memory things that he didn’t have right, in there. But I also know that what’s valuable about that manuscript. I don’t think there’s anything that anyone will show anything is in there that isn’t 100% factual about how that early work was done on diffused devices and what his role (Moll’s)was and why we were doing it, and how it got done. Now, I thought that would get published in the Proceedings, the IEEE 50th anniversary issue of the transistor. It didn’t, and I think the politics of the situation says that, “Holonyak is telling Holonyak’s story and he’s not telling Bell Lab’s story.” And see, that’s one thing that I’ve learned about Bell Labs that I never did appreciate; how they contrived to control information, and about how the business about this Schon thing (the falsification of data) is another thing like that.
In other words, when all is said and done, the person who’s culpable is that dumb guy down on the bottom and idiotic thing he did, but what about these guys that were so happy to ride on the coattails of that when all of it’s being put out, and then later, when there’s some serious problems with that, they vamoose and he’s the culpable one. Why? There’s only one thing that works, and that’s the truth. If you don’t tell the truth, you’re in trouble. And as nearly as I can tell, there is no early accurate Si front-end story that can leave out John Moll and his role in how that got started. And I also know, and he’s reported this, that when Shockley was given this stuff, Bell was helping Shockley. Bell, at that point, thought it was so powerful and so big that the people going out, they would help, they would bless him and they would be Bell Lab’s licensees and all that. I don’t think they could foresee that this stuff was bigger than they were, and that when the chip stuff got rolling, it would be a bigger thing than even they knew how to handle, and were capable of handling. It isn’t something, then, that is exactly just the way they were putting it out; it involves a lot of other people.
Moll had a very important part in that business of pushing the button on how silicon got to be what it is today. But his MIT story is lacking because his memory was already beginning to play tricks on him. He’s got it right mainly, but in detail there were some little wrong spots in there. So that manuscript (my Si story, 1954-55, that didn’t get published and is now at ECS)goes into matters like that, and before they screwed me up on my filing cabinets, I could pull that right out. Gabriel’s got it in the computer. He could print it out. In fact, if you’ll leave me your card, I will have Gabriel email you that. And Howard Huff, with some kind of involvement with Michael Roirdan (physicist author) is looking into—the Electrical Chemical Society puts out a thing like Physics Today. They call it, Interface. Right now is the 50th year anniversary of the discovery of the oxide on silicon, right now, (Spring 2005, 50 years after Spring 1955). And he’s trying to see if they can’t compile a special issue, dealing with that oxide, and he wants to get that manuscript of mine, which is in some meeting report, in a published form so other people can have access to it. I was getting all that stuff ready back then before Nebeker’s time, and you can’t accuse me now of saying, “Well, you wrote it now and you’re playing games with it.” Hell, no.
I wrote that thing basically way back then, back when Bardeen died. And why John Moll got hooked up in this was, I knew that he was sensitive about—people all fool themselves about what they did for silicon, and not even knowing that this guy had an important role in that. See, it’s okay for me to point at him. It’s not so okay, and he’s uncomfortable pointing at himself. And that’s what my point is. I have to point at the fact that Moll was a Si “hero”. That’s likewise (the same) about Bardeen. If Bardeen came through the door, he wouldn’t tell you, “I’m John Bardeen. I invented the transistor.” If Shockley came through the door, he’d let you know right away, he’s Bill Shockley, and the reason electronics looks like it does, is of course because of Bill Shockley. And I said, “No, Bill, it’s not like that. There’s so many people that have done so many things for it, that it’s something that all of us were lucky enough to get in on.” Moll was a very important part of the business of silicon, of that button getting pushed. And see, that’s what my manuscript is all about. It’s the account of what happened that year. What did we do? Why did we do it? How did it happen? Gabriel’s got it in the machine, and he’s real good at this stuff, and just seeing that will remind me to send that to you.
Ashrafi:Great. Thank you. Is there any major thing that we should have covered, that we didn’t?
Holonyak:Well, you know, this is ’05, and I’m learning from Bardeen in January, 1952—53 years ago. I’m getting my nose into what the semiconductor is, and here it is 53 years ago now, and I’m still trying to find out what it’s all about. There’s a lot that’s happened in that 53 years. So I’m sure there are things that we’ve not touched for one reason or another.
Ashrafi:Yes. But we’ve got five hours now. Thanks very much. It’s been very interesting.
Holonyak:Well, it’s sort of where we’ve been, what’s happened and all that. But if there’s something when you’re scanning through this that you need, and you know it’s something that I’ve got on a piece of paper, we’ll send it to you. Or if there are questions you want to raise about it, when Gabriel sends you this, he’ll have an email number you can pick up from him and ask questions.
Ashrafi:Thank you. Oh. This is what I meant to ask. I knew there is something here, at email. How come you don’t just email him?
Holonyak:Oh. Ah, you know, in my time, when you went through school the way I did, the people who were going to wind up being secretaries and stuff like that did shorthand and typing. People doing what I was doing, didn’t do that and the secretary always typed things. So there was always a secretary to type manuscripts, so I never sat at a typewriter. Then when I thought about it, then, “Why the hell should I mess around with that?” John still uses his slide rule. I still use a calculator. I go and open up Bobbi’s email and look at some of the emails. But now maybe I’m going to get forced into it [chuckles].
Ashrafi:Did you ever use computers more than that calculator?
Holonyak:No. No, no, no, no.
Ashrafi:So computers haven’t really affected your work?
Holonyak:No, no, no. Not much. Because sometimes, when we calculate something— I remember I worked out something about a diffusion problem and one of my students threw it onto a Wolfram’s [?] Mathematica thing. He threw it in there. It was a thing that I realized, at some point, I knew how to do this integral, and did it, and he did it one way on Mathematica and I did it another way with a pencil and paper, and so I never—When we ran into certain problems they would push it through the computer, but I would do what I could do on a piece of paper with a calculator. And at this stage, what do I care about heavy computation? I’m never going to go do heavy computation anymore. That’s what someone else’s thing. If I were starting over, I’d have to do all that. No, I think right now, it’s a matter of for video reason, tape reasons, converting stuff, visual information, maybe emails, I have to stop and do that. But I’ve been busy doing these other things with these other guys and writing, and so when do you stop to do that, including sort paper and other stuff? And so I just haven’t bothered to sit there, getting good with the computer.
Ashrafi:And it’s never been a problem for you?
Holonyak:Not basically, because there’s someone who’s around who’s doing the computer stuff. And I’m not doing any heavy computation anymore anyhow. That’s simple stuff way back when, but not now. It’s too late for that. I mean, right now, it’s more important to talk with Milton about what we’re working on and what won’t work, and to try it. So I would prefer now, stuff that I’m reading on—For example, do you see that book there, right there, that sort of pink one right there, that World and other things? The one by Vitali Ginsburg (Russian).
Ashrafi:Yeah, about Science, Myself, and Others?
Yes. Pull it out there for a second. I’m starting on it. I’ve read one of his other ones. But here, look. See, I get involved with this other stuff. If we look here, “Inner weakness. What has been forgotten by academicians, Zhores Alfërov 465.” This is the Nobel Prize winner from 2000 in Leningrad, in St. Petersburg, now. Here’s some stuff about all this under the Communist Era, what happened? So, see, to me this is fascinating, because during the Cold War, Zhores was late ’70 and early ’71 here in our lab.
When Ginsburg gets to here, he says, “Alfërov’s parents belong to the minority of old Bolsheviks who did not become victims of Stalin’s terror. And Alfuro himself, started working in the post-Stalin years and must have been able to go abroad freely enough.” I don’t know if that’s translated quite right or not. And I think what he means here is, “And was able to go freely abroad.” Anyhow, “He worked for several months in the USA.” That’s here, with us. “I do not know whether he would have won, together with two American colleagues, the Nobel Prize, which he quite deserved, as far as I can judge, if he had been born earlier and in general had had better acquaintance of Stalin’s freedom of movement.” I think what he means here, “And had better acquaintance of Stalin’s non-freedom of movement.”
In other words, what he really understood was, how under Stalin, you were going to be living on the edge, you know. “Since what has been said might sound as if being prompted by some personal offense, let me note that there is no such feeling in my heart. My destiny in science turned out to be quite happy.” In other words, look, Ginsburg is really a renowned physicist. He was Bardeen’s friend. And Ginsburg, Bardeen, and Bernt Mathias, the experimentalist, were keeping alive the idea that superconductivity involved other kinds of coupling schemes and issues than just the BCS phonon mediated affair. And this Ginsburg is quite a top-notch physicist, of some renown. So it is a little bit peculiar that Zhores winds up doing hetero-junctions with the Nobel Prize and Ginsburg was for a long time without a Nobel prize. Except now suddenly when Leggett gets a Nobel Prize, you can talk to him about Ginsburg and all that and now a Ginsburg Nobel Prize. Well, this is fascinating to me, because right here, when he’s talking about this, he’s talking about “here.”
And the circumstances of Zhores’ work entered around the kind of things that we’d been involved in. So to me, right now, dealing with some of the history and sociology of where we’ve been and the times we’ve been in and the circumstance and all that is important. And if you’re from a Slavic group like my folks, they were one of the little stepped-on groups. And the Russians were the big group. They had more literature, more culture, more everything else, but also created greater havoc in terms of the world. There’s a connection here. And so knowing about some of this stuff and reading some of this, at this part of my life, is what the hell do I care about poking keys on a box on a keyboard? I don’t care about that. I care more about where have we been? What’s happened to us? What have we seen? What does it mean? What do I care about all that? I’m still working things out with pencil and paper. But I may be driven to it (a computer) because the world has gone that way, and how else do you—
Ashrafi:I’m actually more interested in the fact that you don’t feel that you need computing, computation.
Well, because a lot of what I do now relies on the fact that I know something—that I know something that comes out of the world of an idea and not out of something that I’ve calculated. In my early life, I would have calculated because that’s how we’re taught and that’s how we understand, and that’s how we perceive, and that’s how we see. And I still love to read math books, because math was something that seemed to me, it didn’t lie. It was what it was, built on the kind of firmament it’s on and kind of ideas it’s on and all that. And there’s something still nice about certain kinds of things when you look at them and you see what that expression is saying and how you can go from one to another and all that. So I still keep calculus books and other things around because I find them interesting.
But relative to, let’s say, whether an idea is going to work or not, in the part that I’m around, I probably can think and see whether that’s a good prospect or not. For example, I get into discussions with Milton about his HBT’s, his hetero-junction bipolar transistors, high-speed. And that gets us into some discussions about the current densities and other things. And I know he’s breaking some rules relative to what I know about some of those things. So then when I think about it, I realize, “Milton, there’s something more going on in there. You’re broadcasting some of the energy of recombination that’s going on in there, and there’s got to be photons there. Let’s go look for them.” See, that isn’t something that I computed for him and said, “See. Here. Look. This is what’s happening.” I can tell from the numbers and all that, “Oh-oh. There’s got to be in there.” So we go look, and yes, indeed, they’re there.
So then I know from the prior work, about what size quantum wells we ought to stick in there and what we ought to do and all that. So we immediately get to something that’s working, that didn’t involve a Matrix element and all that. As a matter of fact, we’ve got a colleague who’s very good at quantum computations, Sun Lin Chuang who’s an MIT PhD. Chuang is good with calculations. He immediately made some mistakes that Milton and I knew were wrong, simply because we knew what some experimental facts were, that he had wrong, and that no equation would save him from. We’re not led into the next thing now, by heavy computation. We’re lead into it by something we know led from experimental events. So I don’t really care now. Little by little by little you don’t compute anymore. It just doesn’t happen anymore. The younger people are better at it. They’re just, they get more out of it, and they’re quicker with it. But there are things they don’t know. So I mainly am working from a different kind of a knowledge base now than everything being worked out.
Ashrafi:That’s a good place to stop.
Yes, okay. Well, I enjoyed it.