Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.
Please contact [email protected] with any feedback.
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.
In footnotes or endnotes please cite AIP interviews like this:
Interview of C. Kumar Patel by Joan Bromberg on 1984 May 19 and June 4, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/31404
For multiple citations, "AIP" is the preferred abbreviation for the location.
The first session includes: his reasons for undertaking a Stanford degree in electrical engineering; decision to join Bell Laboratories in 1961; Gas lasers, CO2 and other molecular lasers; nonlinear optics in the infrared region; the discovery of the spin-flip Raman laser at the end of the 1960s; interactions with collaborators, colleagues, and management during the 1960s. The second session includes: his research in the decade 1963-1973; the carbon-dioxide laser; non linear optics at infra-red wavelengths; second harmonic generation; scattering from mobile carriers in semiconductors; spin-flip Raman scattering; the spin-flip Raman laser; techniques for the laser measurement of air pollutants; social aspects of collaborative research at Bell Laboratories; strategies for being a successful scientist.
I open my questions by a point from which you came from Stanford to Bell, and how you came to go into the optical laser group? And that’s I think where we might want to start now.
See, the reason why I came to Bell Labs in 1961, I think the primary reason was that Bell Labs was one of the very few places where I could get a job. I was not a citizen and quite a few places required a citizenship or a permanent resident status before they would hire, and certainly Bell Labs had the reputation of being open a very active place and most importantly, it was on the advice of my thesis advisor who suggested that Bell Labs would be a good place to at least start off if not stay there forever, and of course, he had quite a few friends here, so probably he was bias but certainly he gave the right advice. My advisor at Stanford was a fellow called Dean Watkins, who’s now the Chairman of the Board of Watkins-Johnson, the company that makes microwave and other components for the electronics industry. So, it’s not something that I may have actively sought out by myself, but suddenly all the people who had around me at Stanford knew more about Bell Labs than I did, and they knew the kind of things that happened here. As a matter of fact, I think, to be very honest, I’d known about Bell Labs only through reputation, but I didn’t have detail knowledge of how good a job that it was. And it sounds very strange that 23 years later to admit that, I didn’t know what was going on here, but I really didn’t. And had my thesis advisor not given me that advice I may not have come here. Essentially his advice was, there’s only place where you ought to seriously consider going. And so I think it’s in some sense, I’m here because of the advice that I got, I think just for local information, I think one has to go a little bit farther back because that might make my lack of interesting science, not necessarily interest in science but my lack of finding out more about science at that stage, may become a little bit more clear. I came to this country in 1958, under rather strange circumstances, came here to get a Ph.D. in some part of science or engineering. Primarily because, not because I was interested in doing that, but because when I finished my Bachelor’s degree in India, the only thing that I really wanted to do was to go into Indian Foreign Service, and in order to do that, one had to take competitive exams, but one couldn’t take competitive exams unless one was 21 years old, and I had 3 years to go, and so I asked myself the question what shall I do for 3 years, and the obvious question was you can get a Ph.D. in you know and at least do something worthwhile with your time.
You got your Ph.D. in engineering, didn’t you?
Yes I did.
And what was your motive in choosing that particular field?
Well, the kind of things that I was interested in which was in those days, microwave physics, microwave engineering, vacuum tubes, that kind of work was being done in the electrical engineering department rather than in the physics departments, and so that was an appropriate place for me to be there, but when I came here, I came to this country with a specific intention of going back to India after 3 years, I got my Ph.D. and take the competitive exam and go into the Indian Foreign Service and so while I worked hard at Stanford, having a long term association with science was not something that was in the back of my mind. And that’s why I hardly paid attention to what place I’ll be working at when I got my Ph.D. I knew about places only in so far as the work that was done at a given place and backed up on what I was doing and trying to find out who the others players are, but not really trying to find out what place would be a good home for me. So Addie, I think that it tells you why I didn’t have very much interest in finding out where I would be working when I got my Ph.D. and then when I came here I sort of was putting off the decision to go back to India by short amount, I said look I would be here for a year or 2 years, and then I’ll go back, since I had already invested 3 years of my time doing research and I was beginning to get bit by the excitement of doing research, so I said look, before I go into something different I would do a little bit more of it. And so I didn’t know what I was going to be doing, but certainly the excitement of research had gotten to me by the time I came here, I suspect that at the advice being something else, I go to xyz other places I would have gone there. Primarily because I didn’t know much about the reputation of a given place.
What kinds of things were exciting you as you were starting to finish Stanford? What were you chiefly interested in back then?
I did my Ph.D. thesis in the field of yttrium iron garnet. The thesis primary dealt with a very specific device that was made using the very high resolution, not resolution, very high Q resolution one can obtain in the small spheres of yttrium iron garnet. I believe that, that idea has been was subsequently exploited by Watkins-Johnson. I think they still do make commercial devices based on that idea. But by the time that I finished with my Ph.D. at Stanford, I knew that microwave engineering was not to my liking, and I wanted out of it. I had worked in it for about a little over 2 years and that was more than enough for me and just about that time I’d been hearing rumors about a marvelous device called laser. Something else that might sound very funny at this stage, you know after all these years, that until I left Stanford I didn’t even, I had never even read a general called Physical Dielectrics. I didn’t know it even existed. And you must understand, it all goes back to why I came here, and it’s, I really had only peripheral interest in science only in so far as getting through and spending those 3 years in a very consecutive matter. But I had heard rumors that there was something called a laser, and when I heard about that I went back and looked up in the Physical Dielectrics that was the first time that I found out that there was a magazine called General of Physical Dielectrics.
Where you had been reading the IEEE magazines or?
Really, I really wasn’t reading very much. I was reading the material that was closely tied with my research and those articles did appear either in IEEE or the proceedings of IEEE or in the transactions of microwave, I forget what MTT stands for, but beyond that I really had spent no time trying to broaden my interests in science because I wasn’t going to do that so why waste time. Anyway, but just about the time I was getting out, I think it was April of 1961 that somebody, you know one of these student groups that you get together and talk about what’s going on, mentioned that Ted Maiman just 6 or 8 months earlier had made a laser operate, and Ali Javan was on the verge of making the thing go CWO something like that, so that’s what excited my interest and that was the first time that I looked up things in Physical Dielectric, so when I came to Bell Labs, I talked to, I was shown various kinds of things and just to think that the kinds of things that I saw, didn’t see any work on lasers at all because the group that I which was the old carryover from vacuum tube research group and they were essentially winding down their effort in the area of vacuum tubes and they were going to start up the effort of lasers and I thought that that was a marvelous fortunately for me because if this effort was going to get started at least everybody else would be just as much ignorant as I am, as I was then, and the probability of being able to make contributions meaningful contributions was very high. The neutral individual that I talked to here, his name was P. K. Tien, he’s still here and I talked to him because.
He’s connected with Stanford wasn’t he?
Oh yes, I was going to come to that. Talk to him because my thesis advisor called him and said, I don’t know what he told him but when this guy comes to talk to you take good care of him, so that’s why I talked to him. And we talked about, what appeared to be very exciting work on vacuum tubes also a very exciting work that resulted a young fellow named Jim McFee had obtained at that time in the Amplification of Ultrasonic Waves by Interaction with Drifting Electrons and Semiconductors, the Acoustoelectric Amplifier, I believe that’s what it was called, it was McFee and Andy Hudson, and that looked like a very exciting field, but anyway, there was lots of exciting things that I saw here and then I talked to some people in the material who were working on the Microwave Laser. Now I get the idea of putting one of these things in a microwave receiver for communication with satellites. That works on a sound like a little too narrowly defined and just excited no interest, so when I asked him would he let me do what I wanted to do in the field of lasers, sure, you are just starting and it’ll be the right area to get into, because we are interested and if you’re also interested we would like to have you. I think your question number 2, and I’m just jumping slightly ahead, your question number 2 is slightly wrong. I did not go into (???)Department, I did go into laser materials, and so.
And Tien was a Group Leader?
Tien was a Department Head.
In the research?
In the research area.
That makes everything more comprehensive.
Anyway coming back to the earlier set of questions, as I said this was not a Laser group, Optical Laser Group, it was the old group that had worked on Microwave Tubes and was going to get into Lasers. That work had no connection with my graduate training, I mean apart from the fact that questions are the same, and propagation of electric doesn’t change, we can go from microwaves to dielectric and sauna apart from that, there was little or no connection.
But, I find it interesting also that the way in which laser works spread in the laboratories is partly through groups changing their direction?
That’s right. I think there is [???] story and Rudi Kompfner who was then the Director of the group in which I joined had just heard about the operation of laser by Ted Maiman and he came back to this group and they were working on trial tubes and vacuum tubes and the story goes something like this, he said, “Well friends, I think the days of vacuum tubes are over, wind down whatever you are doing and as we wind down we will increase our effort in lasers because of two specific reasons, there is a lot of new science to be done but in addition to that, when you look at its connection with the business of our company, its clearly, it’s very clear that 20 years from now our company must have a new technology for our business of communications and makes a very good likely from that light we will be the next carrier of information.” It’s a story, but you are right, your point is well taken that much of the effort that got started into the field of lasers around 1960-1961 came from individuals and groups changing over from whatever they were doing and going to this new field. And I think that is in some sense what brought me to this place, mind you, Bell Labs had the lowest salary offer of any offers that I had gotten so, but the thing that brought me here was that everybody that I talked to, the members of technical staff, the Department Head that interviewed me, the Director Rudi Kompfner, interviewed me, everybody had this notion that something is going to happen here and it’s this tremendous excitement that began wind down what we are doing and all of us new people as well as old people will have to [???] to go into any direction that they want to go into, in the field of lasers.
By the way, I should say that, since Kompfner is no longer living, I’m always very interested to pick up concrete memories of how he interacted with people, and as we go along you know.
But I think that the story that I told you I haven’t heard, but I have heard that story being told to me so many times that, increasing amount of embellishments every time I suspect, but anyway, I think that he was the kind of man who had a good idea about what things are going to be important in some sense, his own work on trial [???] tubes is clearly in the same character that when the rest of the world was going towards to [???]C, he branched out in there and [???] [???] it’s in keeping with what one would expect from that kind of individual. As I mentioned when I came here, the, I think the group was called Electronics Research Laboratory that was the laboratory which I came into, and the Department that I came into was, that I’ll never remember, I can dig it up, but its somewhere in the records what I came into. But the point was that, here was a group that was genuinely interested in getting into new things. I joined Bell Labs sometime in the middle of June, June 20, and 2 or 3 things I remember very well. I had shown up here after having taken a flight from San Francisco the day before and I showed up at work the next morning and never having worked in my entire life until that time, worked for a living, that was a different kind of experience coming in here and I didn’t know what to expect, you know, I had talked to all of these people, but that was in the role of their trying to track me then, rather than my doing something for them. One of the first things P. K. Tien said, “Look don’t worry about work, just make sure that you are properly settled here, and let me show you how things work within Bell Laboratories and I’ll introduce you to a number of young people and people who have been here maybe 6 months something like that they can tell you how things work here and maybe at the end of the day why don’t you come back and we’ll sit down and talk about what you might want to do?? And when I came back in the evening, he gave me a stack of reprints of things on lasers because I mentioned to him earlier that that’s what I would like to do, he said, Look, I think the entire laser literature, maybe about a dozen papers or something like that, it’s something that one can go through within one evening work, that kind of thing, he said look why don’t you look this over and maybe we can talk more about it tomorrow morning and I sorted through it that evening and the next morning, he said do you have any ideas to what you would like to do, now I think, I may have written this down somewhere in one of notebooks but I have to go back and look at it, one of things that I told him was, that what I would like to do is to do spectroscopy with lasers, now thinking back about it, I think it was a stupid statement to make because but I think that’s a statement that you’d expect somebody who doesn’t know very much about the field to make. And I knew what spectroscopy was but not in the gory details of what it was and secondly I didn’t even know that there was no such thing as [???] lasers, so I was making a very broad and big statement and Tien smiled and said
The research program for the next 15 years.
He just smiled and said that sounds like a good idea, he says, tell you what, that’s too big, why don’t you narrow it down a little bit so that you can get started on something more concrete, but anyway the point was that Bell Labs for the first time I had realized why my thesis advisor wanted me to come here. It was a place where people are allowed to talk about very long range programs, one may never get to the end but things are happening between us are so important that nobody would a stop to it. And, the question was would I start working on gas lasers or would I start working on solid state lasers, at that point it was immaterial, the point was that I didn’t want to get into lasers and why did I pick gas lasers as opposed to solid state lasers? I think one of my simplest reasons is that somehow, having something pulse once every minute or every half minute didn’t look like to me a good way of doing science, because in those days [???] lasers were the only lasers, and you know the flash lamp pump, you had to wait till then [???] , then the sample cooled down again, time of the order may be once every half a second or something like that. Gas laser had just been operated here by Javan, Bennett and Herriott, and this — physically present here.
Javan was just about to leave, wasn’t he?
Javan was just about to leave, and so I asked my Department Head, I said, I don’t know how people work here, but what I would like to do is to go and talk to these people and find out from them not what’s in print, because what’s in print anybody can read, I want to get some sort of gut feeling as to how things get done, and he said that that was a good idea, he introduced me to Javan, Bennett and he said look why don’t you go and talk to him, when I went to talk to Javan he said you know I’m leaving in ten days so you got exactly half an hour to talk to me. I said oh fine, so you know in spite of the fact that he was you know essentially packing, I recall that he did give me very much more than half an hour and we talked for a great length of time, I asked him all kinds of questions about if he thought helium neon system was going to be the only laser system that was going to operate [???], gas system or there were other possibilities, and I could be wrong, but I think he was very arrogant about the fact that he did not think gas laser with a single gas constituent was going to be the right direction to go into, for the simple reason that one would always need selective exhibition. I suspect that was a carry-over from all the earlier work that had gone on here both here and the clinics, primarily in terms of even for solid state lasers, the idea was to excite to a specific state and this also gave the idea that [???] — was deeply engrained in the community and I think that is something that came through very clearly. Another rather amusing thing that happened was, that was the first time that I got introduced to the Books on Atomic Energy Levels, table [???], in a sense that somebody had compiled this enormous and awarding three volumes that had taken, but so, Javari played a very important role, not in telling me how to go, but showing me that there were things called atomic energy lows which are compiled in all this goriness of whatever it is. Anyway the upshot of that conversation, I came back from him with a nagging feeling that, two feelings, one that had to be more to gas lasers than finding coincidences, accidental coincidences between the energy levels of one gas exercises of one gas, and exercises of some other gas in order for their system to work. And the second was that that was the only way gas lasers would operate, they didn’t have a great future, because [???] looked through atomic energy levels the table thing, you could show anybody and they should know that there will be a few accidental coincidences but not a whole hell of a lot, and
That’s a real dash of cold water, in a sense.
That’s right, and well it’s also, I came back and I told my Department Head, what I do like the idea of gas lasers it’s still gives me a slightly sinking feeling that maybe this is the last, the first arid the last gas laser system that’s ever going to work. Tien being a man of considering what experience that I had, and you know I was 21 years old when I came here, he said look, that can’t be right, I mean there is no such thing as the first one being the best, there are several being done and any of them, one has to take this thing on face value that you know, and so I decided that the thing to do was in addition to looking for accidental coincidences, one must also look at the details of how systems work, namely lifetime — lifetimes may turn out to be more important than accidental coincidences. And if you could in fact, find an upper-level which have a much longer lifetime than a lower level even a uniform exhibition to both levels will produce conversion for you and give you laser action. And I think, I don’t believe these are new ideas, I think people have recognized that, the question was that that may be, I thought if you cannot carry out all the calculations on the back of the [???] it’s only a few pages, maybe the right thing to do is to trans [???] up an experimental program which systematically looks at what idea of things, what idea of gases to find out if these ideas about only lifetimes are the mixed gases and it turned out to be that [???]. So beside that, I decided that initially I’ll spend much of my time trying start with what had already been done, namely first — helium neon laser, Ali, Javan, Bennett, and Harriott, to understand that it in fact works and do a somewhat detail study of the role that helium played in the operation of helium neon laser.
Was this not, already available, this task done?
What had been done by Javan and Bennett was that, anyway the optimum ratio of helium to neon and helium neon laser that was used in the first gas laser, and the gas laser which I think, the ratio has been subsequently checked to be correct. Was allowed by looking at, not by looking at the laser output but was that I read by looking at the lifetimes and the rated which helium energy was transferred to neon, and so, this was derived from essentially spontaneous admission studies from various states in neon Which was you know, was a perfectly good way of doing things. My approach was slightly different in the sense, that I said, look I believe everything that’s been said, but what I don’t believe is the requirement that the optimum will in fact occur at the point where the [???] spontaneous emissions results would tell you because the argument was that the earlier results were carried out in tubes, this spontaneous emission measurements was carried out in the discharge tubes which are very different [???] if it were compared to, to the [???] that you use for a laser and walls are going to play an important role and the question is this is a different system. Be as it may, what I decided was to do a really parametric study of the laser output power versus the helium neon, both the total pressure as well as the component, the mixture the ratios. One of the things that had come through rather clearly from talking to Javan and hardly talking to Bennett was that they believed that gas purity was of most importance. And they had gone through an enormous pains to make sure that they had done all they can to prevent any contamination of helium and neon from other things. And for that they used mercury, diffusion pumps, with nitrogen traps, liquid nitrogen traps, and so on and so forth. And what they had done were to make sealed off helium neon lasers. They would pump the tubes out quite extensively make it out for a long periods of time to a very good vacuum and then fill these tubes with the right mixture of the helium neon and then seal the tube off that no further contamination could take place. I argued that that’s one way of doing things; the second way of doing things is to not take the tube off the pump ever. And use breakable valves and perhaps even go to a system where you’ll floor the gas to the tube by some slow speed so that if there were any contaminates coming from say for example, out gassing from the walls of the tube this will be constantly swept out. And I believe this was the first time anybody had discussed the idea of using flowing gases through the activity of the tube. In fact that, that turned out to be a very clever idea, a very small idea, but it turned out to be very useful, because once you start flowing gases then you realize that changing gases in a tube becomes a sort of trivial matter, you don’t have to premix gases fill the tube find out what you’re doing, from an experimentalist point of view that was a much easier way of going about the problem, and I think, I had my equivalent of Javan, Bennett had [???] helium neon laser, going here sometime I believe in, by thanksgiving that year which was the end of 1961, when is thanksgiving? Thanksgiving is in November. Late November, that’s right, and there was another group of people who had reached about the same stage in the Department in which I was, but they were using sealed off tubes and that group was very much more interested in looking at the properties of the output spectrum of the laser, rather than the medium itself. This was called the Koegelnik and Rigrod were the two people looking the mold patterns of the [???] output, trying to cool their doors with what one Boyd-Gordon or Fox-Li calculations of residential. And so in some sense, there were power efforts but they were not directed towards what I consider the gain medium, they were directed towards medium. So what I am trying to bring up here is that in that very short period of time 6 or 8 months a viable group of people was emerging in that Department, where different parts of the laser physics were being attacked. Some people working the optical properties, other people looking at the gas properties, which is part of what I was doing.
Actually, I want to just interrupt you, because I’m a little confused. Now in these early researches you were working alone still, only in discussion with people like Bennett?
Is there anybody else who’s an important discussant that you were conferring with a lot at this point?
I think until the time that I had my own helium neon laser running, let me come back to that, I think that’s a good point. The only individual I was really interacting on some sort of informal basis was my department Head. Using him as a sounding board, more than anything else. Also, he had extensive experience in vacuum technology he came from the vacuum tube era so that he knew something about it which I knew nothing, so in some sense that was a good place for me to go and then he was right next door. The only other person I worked with was my technician then, and I think it’s not the same person that I work with right now, a fellow called Joe Hasiak. He was what you might call leftover from the vacuum era so he was a very valuable individual to my program to put all the vacuum system back together. But let me continue the story [???] that I was telling. Just after the helium neon laser was working in my laboratory using the different way of doing things are they flowing the gases slowly or changing them quite rapidly keeping the same tube on the pump all the time and so on. That would not be a very easy way of changing the gas concentration, and one of the first things I did was I started cutting back on the helium pressure and surprise of all surprises you didn’t need helium to make the neon laser work, it worked anyway. With zero helium the neon laser worked, that put a very important point out that had been bothering me but when I embarked upon the thing this program did have an open future or was it a closed story?
Now is this electronic citement in helium?
Yes, this is just electronic citement; this is the discharge of pure neon in a neon discharge.
Now that must have surely has been something that people reacted too?
Quite violently. The point was that the [???] wasn’t quite as what you got from a helium neon laser, almost a [???] smaller, but the fact of the matter was that the pure gas discharged laser worked, and this was a first time when I came to know of Bill Bennett, and I had known of him, but I had never talked to him until this time. The only person I talked to was Joe and he was leaving. I came to talk to him because I think the history now shows that he had essentially the same kind of concerns that I had about mixture of gases and in making a laser, but he was doing was measuring gain. Optical gain in gas discharges having different ratios of helium to neon gases. And I’m not [???] to his lab notebook, so I can’t tell you [???] but I think for all practical purposes, I sent the paper off for publication to General Applied Physics and Bennett gave a talk at APS meeting in New York, my first APS meeting that I attended.
That was winter of?
Winter of, January of 1962. He gave an invited talk and towards the end of the talk, and he talked about helium neon laser and towards the end he said, oh yes, and by the way I just made some very recent on the measurements of the gain in helium neon discharges and I find that even in pure neon there is some gain that remains, even when you take all the helium out. And that shook me because that was the first time that I was exposed to what I call competition, I didn’t know that there was such a thing. So after his talk, I walked up and I said, hey, Dr. Bennett, I’m from Bell Labs I introduced myself and he said I’ve heard of you, and I said I’m glad you’ve heard of me because I think we have something very much in common because just two weeks ago I sent a paper in for publication to General Applied Physics which showed and he said he didn’t know that, we’re not too far in this building and he was in the other end of the other building so we were 4 minutes apart but.
Both in the research department?
Both in the research area, but different parts of the research area. Anyway, that began a very fruitful collaboration Bennett was still ready to fill the tubes seal off the tube and do the measurements because his argument was that, that’s only where I know what’s in the tube. Because he had a slight advantage, he was an accomplished glass blower. And so he could do all of the processing himself, and so all of these things fall into place as to why I stayed away from sealing of tubes because there was no way I could seal off my own tube, I would have had to have a glass blower come and do that, which means that I could work only from 9:00 to 5:00 and I couldn’t work late at nights. To say, just aside, one of the first things my Department Head mentioned to me was that we work here normally 37-1/2 hours, but he says nobody is going to check whether you work 10 hours a day or 60 hours a day, but you can’t work 60 hours a day, but the point was that when everything is said and done it’s what you produce that’s going to count, and not how long it takes you to do that. And I suspect that I had not stopped being a graduate student when I came here, and as a graduate student, you worked long hours, you did that in graduate school because you wanted to get out quickly out of that jail, and I suspect some of that still got carried over, and I think that the, I don’t recall ever having not come back to work after dinner, I mean certainly in the first ten years but I am just talking about the early time.
Sounds a little bit as if Tien was treating you as a graduate student in some way, I mean, I don’t mean to say that he was treating you as a young man, but I mean that he was giving you the problems and giving you your own opportunity to find your way?
Well yes and no, I think with a small difference I think that the point was that he was not, he didn’t give me problems, I chosen the problem myself, that’s one, and all he was doing was making sure that I had the resources necessary to carry on what I was doing. Also, he was telling me that a very important fact of research life, namely, unless you try enough number of new things, you will not find the few of them that will succeed. He didn’t say that in those words, but, and second is that if you want to try a large number of new things, it takes time. Because some sizable fraction of those is not going to work out and as an experimentalist, time is the only thing that you have on your side, nothing else. Anyway, ideas count but the point is that even when after you have an idea an experiment has to be tried out because you don’t know all of the parameters that are going to go into making the thing work or not making it work. Anyway, coming back to the collaboration with Bill Bennett, I described to him, I said look why don’t you come over to my laboratory and let me show you what I have, I showed him, I said, this is a very different kind of approach from you have so we are not competing at least on an approach by approach basis but we may have some fair amount of power efforts which either we can join together or we can run against each other, I mean you know, take your pick and I said I’m younger than you are, and [???]. And he said, look there is no point in duplicating everything. In those days, for example, the dialect recorded mirrors that we used for making laser of service with them, were awfully expensive and if I remember correctly, one of these mirrors used to cost something on the order of $250.00 and in those days that was big money, because if you’re throwing the inflation into the picture they’ll cost a thousand dollars now and you wouldn’t think of paying a thousand dollars for a mirror these days but you know that was a very, and there was only one supplier maybe two suppliers but you had to use one of them. And, so I said look, the first thing we can do is to share mirrors, because you know, and I said look what I want to do, is very different from what you want to do, what I want to do is to really go after new systems, I have no interest in beating this system to death, because it can be done, but that’s not how I’m going to find new things, and how to establish myself and the only way that I’m going to establish myself is making sure that my name is associated with something new, [???] there is still a lot more to be done. And so what my approach would be to have a broad set of mirrors that would have reflectivity all the way from say, from the visible union to maybe some length of union, maybe 2 or 3 microns because that’s, if I understand the energy level diagrams correctly and if I look at all the previously measured lifetimes, there are at least a dozen or more than a dozen cases where having made the neon laser work I know that at least a dozen more transitions that will work in other gases, I know that because these things don’t look much different from neon and then to do this much work, and I need these kinds of mirrors I said I don’t know what your interests are but presumably you’ll be just as interested in that too, so why don’t you buy half of them and I’ll buy the other half, at least we know that way our bosses don’t certainly have a heart attack because I put in an order for $10,000 worth of mirrors, he thought that was a good idea. Just about that time, he was also working on come back, to the question of single gas sources, multiple gas systems. We all fall into a runt rather quickly, and he had fallen into the runt of two gas systems, even though he had seen the gain in pure neon he just wasn’t convinced. And so, he said look, there is another system which is the neon oxygen system, you know I think in some sense that was a very clever idea on his part, very, very clever idea, I think much more so than the helium neon system itself. Because the neon oxygen system required the transfer of energy from metastable states of neon to [???] oxygen. There the oxygen would be dissociated and one of the oxygen atoms will end up in the upper laser level, and that one will end up in the lower, in the ground state. And dissociated oxidation of molecules by collision with atoms was not a new idea, but the point that this may have somehow impact on the laser scheme in general, was, I thought it was a good idea for a simple reason that there are only finite number of atomic gases that I can come up with, where there are an enormous number of [???] gases I can imagine, and so the possibilities of this combination of an excite of an state and placed where you transfer the energy to are increased significantly, if one of them was an atom and one of them was a molecule, and so I said look alright I will agree to work with you on this, he was very much interested in measuring gain, I said, look in that case, because of my system I can make changes rather quickly I’ll take the tag off of making oscillators. And so, we started that way, and I think by April of that year we had the first neon oxygen laser working.
So you were working physically such that you were working with your equipment and he was working with his equipment?
That’s right, because complimentary, and as I said we somehow didn’t see eye to eye on this sealed off versus non-sealed off systems and, I suppose he just loved to do glass blowing.
Well that’s interesting. I didn’t know that collaboration sometimes preceded in that matter?
As a matter of fact most of the collaborations at Bell Labs do in fact start essentially the same way, by having complimentary ideas or complimentary techniques, or complimentary experimental apparatus. That’s precisely how things work here. In the research area we do not have large teams, you see, and so collaborations spring up because [???], I happen to listen to his talk or walking down the hall you suddenly find that somebody is working on something and there is a place for cross fertilization of ideas and equipment.
Now, how did McFarlane and Faust get in on this?
Now McFarlane, Faust, and Bennett were in the same department, and I think Faust and McFarlane had joined the Laboratories about the beginning of 1962, that may not be right. And their Department Head who was Geoff Garrett essentially must have [???] to Bill Bennett that here are two new people would you please take them under your wings and launch them into the [???] lasers or gas lasers, they came from the other side, essentially they were part of Bennett’s effort that already existed, and so we were all in the research department, not everything went well, not everything went well because, you must know a little bit about Bell Laboratories for you to be able to understand. When something exciting happens in a given laboratory, one of the measures, one of the internal measures of that work being significant, is that that work is described by the laboratory Director at the monthly meeting that’s held by the Vice-President, research area Vice-President, where the Directors get together with the Vice-President and essentially you give a report of what exciting things are that are happening that is in the past.
And that would have been Kompfner in this case?
Kompfner had just been promoted by the time I came here, so it was Cutler. Chip Cutler. And so he would describe the work at the Vice-Presidential Staff Meeting. And Chip had described my work on pure neon laser at one of his meetings in January. That was fine, but what happened in the next couple of months was while Bennett was still working on the neon oxygen and specifically neon oxygen system, I had expanded my gas laser experimentation to include other gases because as I mentioned I had an annual for at least a dozen transitions in other gases that would work, and so I’d known my system, I had argon, zenon and crypton and with the set of mirrors that Bill Bennett and I had bought together, and in fact I found that these transitions that we thought were going to work, in fact they did work all the way to about 2-1/2 microns or something like that, one of the strong transitions for the .02 micron in zenon, which really required only one mirror, there was such a high gain that the thing would just — no matter what you did. Which to me was very satisfying because of notion that you needed a second gas was proving to be less and less important. I told of these results to my Director and he reported this at the Staff Meeting, now of course had I been a little bit smarter a little bit more sorry I would have taken pains to tell my Director that I had used some of the mirrors that were used here, bought by Bennett, of course I forgot to mention that you know, this is called 20-20 hindsight but I didn’t mention that to him so he reported this work as being my own work, that got back to Bennett through his chain of command, and apparently he must have blown his stack because the next morning when I came to the Labs Bennett had taken back his mirrors with a little note saying that our collaboration is over. He was obviously hurt and so, because I didn’t know what had happened, so I walked up to him and I said, look, hey, what’s the matter, come on I mean, things don’t stop like that, what happened. So he told me what had happened, I said, look it was clearly a mistake and I had no desire to slight your contributions and I’ll be more than happy to put you down as a co-author because at some point we had made a decision that you try to take several federal tax to govern as much area in gas lasers as we can. But, you know if you want to continue separately it is fine with me, but I think it’s, you ought not to do it because of this decision, try to find a better reason to do so. Now, he calmed down after a-while and we continued on our path again. But I don’t think things were ever the same after that, I think that was a mistake, I learned a very important lesson there, namely that, what you have in your mind is not sufficient, you ought to convey that information to others also in the, I should have told my chain of command that, that I was collaborating quite closely with Bennett, I didn’t think it was necessary, anyway, it is after the fact, but when we operated the oxygen laser, again, one of the things that will come through in all these ramblings is that way back somewhere in my earlier childhood I had been taught to ask the question what if, and one of the questions I asked was, since I had other gases on my system anyway, Zenon, Argon, Neon and others, the metastable level of Argon still lies about the dissociation limit of oxygen, so the collision between Argon metastable and an oxygen atom will be associated where it doesn’t have enough energy to leave one of the oxygen atoms in laser, the place where both of them is going to end up at is in the lower state. So the next argument is that if pure gas, noble gases lasers, neons, zenons, crypton and so on, suppose I ought to start with an atomic oxygen gas, which is in fact what I get by using argon oxygen mixture, because it does produce atomic oxygen [???], then of course I should be able to get an inversion in oxygen by electron impact subsequent to its dissociation by golly it worked, and so the paper that we wrote, I wrote with Bennett in fact, has both neon oxygen and argon oxygen, simultaneously, he had not worried about argon oxygen because as I said, depending upon what mind set you start working with, and I had a very different mindset, now he did come and hurt me eventually, but it took a little while for him to get to that point where, whereas you see, I did come back to a mixture of gases and had I not had my mind set.
Your mind set which was led away from mixtures?
That’s right, and anyway, the I think it’s about this time that I started working with the technician who I’m still working with him now, a fellow called Rudi Ken, and I think it’s now what, 22 years that we have worked together, that maybe good or bad, but that’s immature, the point is that in some sense the two of us have grown older together.
Does it make a difference the point of view of your technician, I mean does that contribute and change the way in which you work?
Of course it does, because the technician brings a certain number of, certain amount of expertise with him or her, and so that in some sense, in a subtle way does make your experimental efforts stronger in one area as opposed to the other area. But the point that I was making was that, you very quickly recognize that after people work together for a few years, often you begin to think in pretty much the same manner, I mean, that’s good and bad both, and you get a lot done, but then you miss some obvious paths. In all of this, the work in the first year or so which included pure neon laser, laser action an ideal of noble gases and the argon oxygen and neon oxygen lasers. I don’t recall my management ever telling me what to do or what not to do. Remember this was a brand new field, and the management knew no more than I did, in terms of where one must dig to find the next gem. And so I think the question as to what is the role of the management, I think that the role of the management was probably in the area of providing support from the point of view of resources, secondly, support from the point of view of insulating the scientists from other problems which, problems that come from above, namely funding problems, if funding goes up or down or what have you, and I think the Bell Labs management does that very well, and did well even in those days to make sure that the projects, projects is not the correct word, people that the management believe are the promising people argument all the freedom and the necessary resources to carry out the kind of work that they’re doing.
I also get the impression that there was no pressure whatsoever for any particular applications at this point, is that correct?
Yes, that’s absolutely correct. I think if there was pressure, the pressure was not on people who are doing the fundamentals of science, of this kind, the pressure was on other people who were in the area of, In the earlier days using microwaves for communications for example, that pressure may have been there on them to look at the possible ways by which a laser source may be used for communications. Sticking to communications for the time being. But that’s not necessary a [???] group of people but certainly it’s a very different group of people. To them a laser is a black box, from which coherent electrode [???] comes out and as long as you give it to them they’ll be very happy.
Now who were these people in, I mean were they located physically here, were they?
Traditionally there has been a fairly large effort at Holmdel, in the Communications Systems area. It’s still in the research part of this entire Bell Labs enterprise. But in the area of communications sciences and for example, Stu Miller had, eventually Stu Miller suffered a light wave technology in the late 60s and early 70s. But clearly, these other people who have, who are much more detailed in knowledge about communications systems certainly more than I had, more than I have even today, I have some,
But were you, now were you talking to these people at this time or did you meet?
Okay, you just knew that there were some people over there?
They were there, and I think in some sense there was a responsibility of the management to have this information sort of filter down there, about from transmission of memos of papers or what have you, that traditional mechanism. But all of this research was funded by AT&T then and …
And you were at any rate personally not involved in this whole question of funding?
No. There is one very interesting story, I remember about funding. And I can honestly say that my management supplied essentially all of the equipment support that I wanted. Essentially and unflinchingly I think after the first operation of pure neon laser I think my creditability was established in a sense, that somebody was thinking in a manner different from most other people part, and so was likely to find new things. One incident that I remember I think sometime in the middle of 1962, I recognized that while I knew what I was putting in my laser tubes I didn’t know what was in it, in a sense that I didn’t know what their impurities were physically present because I had no way of measuring, and so I asked’ my management to, if I could acquire a mask [???] GE, I think at that time had made the first commercial magnetic deflection mask [???], until that time people used to make their own. And so, I said look, you know, it was $7,000 or $8,000 at that time, I said I think it would be awfully good because at least I will know what is inside, and if I’m to put down on a piece of paper, this is what works at least I can put down this works with impurities at a level apart a thousand apart, so another thing, when time comes for somebody to build this for actual use they’ll be able to reproduce what I have. Now this went up the line and that’s a lot of money, are you use you want it, I said yes I want it, and then well finally it went to the Executive Director which at that time was John Pierce, and well for whatever reason John Pierce said no, and the stories I heard subsequently was that my Director then which was Chip Cutler, had a major fight with John Pierce, he says look, man of this caliber comes only once in a great while and to not give him what he wants is a major mistake, Pierce stuck to his guns for a while, but I think 2 months later I did get my mask petro meter, which I thought was a good idea because I think it did not contribute materially to making better lasers, but at least when things went wrong, I knew why they went wrong, so that was a very, very useful direction. One of the things that, I’m walking slowly down this, I think this may take us forever to go through all of this, but, the next step in the fill of lasers, like I said, we had a number of, we had essentially all the noble gases lasing, we didn’t use it long, because we use radioactive gas, but pure helium, pure neon, argon, crypton, and zenon, they had ideal places and therein I published the paper called - techniques. Just goes to show how much I knew what spectroscopy, that clearly wasn’t spectroscopy, that was really finding a few lines which were not seen before, and we had all kinds of trouble with the reviewer, a physical reviewer, one of the referee, of the physical review letters, and obviously this man was an old time spectroscopic, or must have been, because see he really [???] to the title Spectroscopy. Spectroscopy in the title. And his point was that spectroscopy means doing a systematic study of all emission lines, not just things that lapse. I mean he was very armament about it and we were equally armament, and our view was that if you are able to identify new lines which were not seen in the earlier studies because the spontaneous emission is very weak and as you go to longer wavelengths just the simple fact that frequency cubed factor comes into spontaneous emission cross section longer and longer, you are going to get fewer and fewer plans comes out of [???] point in your submission, so our argument was, in order to do spectroscopy in the longer — you are going to have to use [???] emission technique, so I’ll make these transmission lasers. We stuck to our guns and it got eventually published, but there was a fairly acrimonious exchange of letters between us and the referee, the fact of the matter was I could have called the title of that paper, New Laser Transitions in Pure Noble Gas Discharges, I think the thing would have gotten published immediately because it would have gone to a different set of reviewers.
It’s a nice little illustration of the older specialist in contact with new kinds of techniques.
That’s right. Again, I use that as an indication of how people learn, I was only a spectroscopic by training, this exchange with the referee taught me a lot about what is meant by spectroscopy. I came to appreciate some of that and used much of that when I became, one of our spectroscopies in my subsequent years.
You’re getting initiated into the social conventions of spectroscopy along with everything else.
That’s right. Because by this time my idea of going back to India was all gone. Somehow it just did not make sense that, and so finally I had to learn what science was. And it was the first year of working here that taught me what it was. I suspect that if I were to do things over again I don’t think I would do it any differently. Maybe I would socially a little sorrier than what I was, but maybe not; it would be just as bad.
Well, that indicates it was a year of great excitement for you.
Oh yes, it was. I think something, again another side, I had gotten married in August of 1961, and you know my wife and I both were very young and I think during those first few years even my wife knew and could feel the excitement I mean each day there was something new that was coming out, there was not a single day that passed when genuine excitement didn’t, something really exciting didn’t happen here. And I think for that reason, she put up with late nights essentially in a day in and day out, all Saturdays’ all Sundays and I think, eventually all of that caught up with me, because you get older and you don’t have enough energy, but that’s different. But the collaborations with Bennett came to an end by the end of that year, because Bennett left to go to Yale. He left to go to Yale with very mixed feelings he essentially thought Bell Laboratories had not done well by him, and that you know, during the last couple of months I tried to find out from him several times as to why he was leaving, because I said, look you know we have a good collaboration going, and we have our hard times but I think the field is far from being just completely finished and it’s a lot to be done, well, but he just somehow didn’t feel that the management had everything done. Collaborations continued with McFarlane and Faust for a much longer period of time. By that time, I think that was the middle of 1962 or thereabouts, we had essentially run out of new laser transitions in noble gases. Now when I say, what do I mean by ran out of it. You realize that, in order to make, to go to longer wavelengths, either you have to have mirrors which you enclose in the vacuum of the laser tube because they don’t have windows, there are no transmitting windows in the long length region, or you have to find some transmitting material, one or the other. The Bennett, McFarlane and Faust, Bennett was the McFarlane and Faust group, coming from that side, they were used to putting the mirrors inside the vacuum model, because that’s how the first helium neon laser was built, it did not have blue strangle windows, the blue strangle window idea was invented by [???] in the laboratory in which I was. And so, there was a tact that I had taken, they were still using the enclosed mirror geometry, which has of course, advantages, one of the primary advantages that since you don’t have the blue strangle windows you don’t have any additional loss mechanisms, so you can look at even the weaker cane transitions and make them oscillate. The disadvantages are that, when you change the mirrors to go to a different wavelength region, you have to break the vacuum to put the mirrors back in the vacuum and the whole slew of things.
I guess I’m a little bit confused because, I thought that the chemist were doing infrared spectroscopy and had sodium chloride windows and all sorts of things?
Yes, but you must understand, that’s a different field, I was ignorance, I didn’t know anything about it, we will come to that eventually, but I didn’t know anything about it. See, this is how you learn, you learn because there was no way I could have learned about all of that if I was doing something at this end of physics. You are absolutely right, there was those out there who knew everything about infrared windows, except that I didn’t know anything about it and McFarland and Faust didn’t know anything about it so, we were doing things that we knew how to do. But the question about mirrors, still remained an important point. Because mirrors were essentially operated on dialectical mirrors were really operated on Pyrex substrates. And that gives you a limitation in terms of transmission. 3—1/2 microns or something, the longest length of that force. The tact that I took was to start using different kinds of materials, germanium and silicon, because they are transparent for the mirror substrates and the tact I took was, we’ll put down partially deflecting gold surfaces so that some light would leak out. Which was a good idea and a bad idea, good idea in the sense that, that does give you a broadband mirror, you don’t have to have mirror for a particle of wavelength, it has reflectivity. Bad because, a mirror that does not in the infrared region, the metallic coating that does not reflect some fraction of the radiation, that fraction is solved, that fraction is not transmitted, and you can show that frontier. So a small amount leaks out and that really doesn’t do you much good, what you are saying is that in order to get 1 percent transmission you have maybe only 60 or 70 percent reflection you are throwing away 20 percent of the operation. Anyway, that’s the tact that I took, and was in fact, we used both McFarlane and Faust at the other end of the building and myself at this end of the building, now we are gone to mirrors inside the vacuum chamber anyway, because we had to do that. And we used partially reflecting silver or gold coating on these silicon germanium mirrors. Something I should have mentioned earlier, how people get stuck to what they start doing. Bennett in his entire work here never used anything other than flat plain paneled mirrors. Very difficult to line up and everything else, while I never used anything other than Konfocal mirrors for the simple reason that konfocal mirrors invented here and flat mirrors are there, it’s simple, see, how you got stuck, but the world subsequently has learned that konfocal mirrors because of the advantage, many many advantages turns out to be the right thing to use. And by the time the end of 1962 came along, we were all changed over to using konfocal mirrors anyway, Faust, McFarlane, and me. At which point there is some discrepancy it you want to call it, I shouldn’t say discrepancy, but I have never been able to track down where the idea came around, but both McFarlane, Faust and myself, recognized that having a partially deflecting gold coating is not the right thing to use, because of the reason, that much of it is lost. And so, the tact I took was, I still stayed with the silicon or germanium substrate and had a mirror spectroscopy with a little bit hole left off in the mirror, small amount of uncoated surface, a circular surface left off in the middle. The idea was that the small amount of energy would leak out through that and of course what you do, is try to find out what the fundamental mold size will be at this long wavelength see. And make the hole much smaller than that, so that you would not perturb the fundamental mold inside the cavity.
This is still silver and gold or something else?
Silver and gold, so what you do is, you have a thick coating of silver or gold with a little bit of uncoated surface. Faust and McFarlane took a different approach, they said look, that’s hard, what I’m going to do is to drill a physical hole through my mirror, and of course both ideas worked, now introspect drilling a physical hole turns out to be the right thing to do, but I’m not sure where that idea originally came from to leave out an unreflecting surface in the mirror.
Now I’m also just wondering, this is the continued infrared spectroscopy paper?
Yes, that’s right. We are now to
And we’re about the late 1962, right?
That’s right, late 1962 is right. That is correct.
Well actually, we’re in a lot of stuff that I left out?
That’s alright, but I think I will come back to some of the other things. But using these techniques, using enclosed mirrors holes in the middle, either leaving out undeflecting, uncoating surfaces in the middle of the mirror we were able to reach very long wavelengths. This is that 133 microns in the infrared. And, you know, the number of lines we solved were just enormous, these were the days when if a day went by when either myself or Faust, McFarlane, didn’t find 10 new lines each day, there was something basically wrong with us, I mean, it was just of question of looking, it was there for you to find, and that is to my mind, was such a big change from the first helium neon laser. That what we essentially proved wasn’t a gas discharge, in version was more of a norm than not, some levels we were always getting murdered because simply of the lifetime considerations and nothing else.
Now the infrared spectroscopist, were they getting more and more interested, or what was the relation to that group?
I suspect I never kept track of that; I will come back to the spectroscopy I think the real interest from the commercial spectroscopy in ours came when we got into the system of [???] systems. It did not come until that time.
Who was reacting to you at this point, do you have any recollection?
I think probably the laser community. A little bit of people who were trying to do exciting kind of spectroscopy’s using the lasers as sources, but not very much, we are at a stage where all of this is in very much competition with solid state lasers. And gas lasers were not the favorite of the community. It’s a good toy you found lines but its ultimate usefulness was seen to be limited probably if for no other reason lasers were so good because they were going to give an enormous amount of power, and how anybody in his right mind could think that a gas at a few [???] pressure would have, can give you a lot more power than a solid where the density is, where the density of active irons is 3 orders or higher, it just was not the way one would do things, and so gas lasers, while they were providing all kinds of excitement to us a small community here, were not considered the main stream of laser activity, not even in Bell Laboratories, mind you, not even in Bell Laboratories. I mean certainly the outside world was going elsewhere, I mean, everyone was making higher and higher power of relay of glass lasers, you know, but nobody was spending their time trying to make gas lasers. Because these were never going to be high powered devices, they were clumsy, glugy, big, I mean you attach every project objective and gas lasers were them.
But you’ve got people like spectrophysicists working with Perk and Elmer to turn out commercial gas lasers, does that enter into your world at all, this kind of attempt?
See that happened early 1963. Now see, one of the other things I missed here was, I consider a significant achievement and this happened in the [???] Rigdon and White made the helium neon laser work in the visible and that in fact was in late 1962. That was a major boost to the field of lasers; I mean that at least a use was found a practical use was found for gas lasers. The point was only can you make this small enough and portable enough and make them last long enough so that they could be used as pointers. Essentially a straight line. And that didn’t happen until much later about 2 or 3 years later when finally people learned how to process tubes and so on and so forth to make them last long enough. But the gem was born there, that in fact gas lasers had some applications.
I had a completely different
Anyway, in all of these running after new transitions there are a few things that happened this provided considerable amounts of excitement, original excitement to me? One was the field of how to restrict the number of longitude long [???] instead of oscillating into a gas laser. These lasers were a meter long; a helium neon transition will have maybe 6 or 8 modes oscillated at the same time, so these are not single frequency devices. And, what I discovered was that if you couple two separate cavities, we are still on the first page, this is number 6, if you couple two cavities with different length cavities one very short cavity one very long cavity, then the combined modes of this compound cavity has much smaller number of modes than one single long cavity, and you can in fact make the helium neon laser work on a single transition. The coupled cavity approach, now that, well that remained we obtained the pattern and that remained a laboratory curiosity, until 1982. Until one of the fellows in I can call my group here, applied that idea to semiconductors and invented what is called the Cleaved-Coupled-Cavity Laser, this is semiconductor laser, what is involved here, is one takes a whole semiconductor laser and breaks it up into two pieces, just cleaves it without completely separating them, so that now you have two separate cavities, two separate lasers, but they are optically coupled together.
Now Kogelnik was working on mode suppression independently at this time and you got together with him or what?
No, he happened to have the only operational 6328 angstrom the visible helium neon laser in my, in the department in which, but that shows again very graphically how collaborations get set up in Bell Laboratories.
Yes, it does, it’s a nice example.
And again, as I pointed out that idea took 19 years before it became practical, but you know, we had done all of the ground work in terms of what was needed to make single frequency lasers. There is another thing which happened to me in those days was Dick Fork came to work at Bell Laboratories, and he came from the field of having done conventional spectroscopy, not spectroscopy on a large number of lines but looking at an almost dispersion in gas absorption lines, looking at magnetic field effects, looking at zenon effects, and things of that kind. And one of the first things that he wanted to do was to look at, search for the same kind of effects, magnetic field effects, zenon effects a transition that wasn’t working, see in other words on a laser transition. And I had been working on the high gain 2.026 micron zenon laser and it turned out to be we decided that that would be a good thing to try it on, because as your plan [???] you would split the lines, so if it was a weak gain transition what you would find is that it is not a laser at all, because the gain is split, and since the zenon is a very high gain transition we were not able to do that. So we tried that out and it worked and it worked very nicely, you could fairly sizable tuning.
So, now you’re working with a spectroscopist, a properly trained spectroscopist?
A properly trained spectroscopist, yes, that’s right. And, that turned out to be awfully good. I think, in some sense we were the first, Dick Fork and I, were the first to show what could be considered in those days broadband tuning of gas lasers. You’re using a small electromagnet and we could tune over maybe 30 kHz it doesn’t sound like very much tuning these days, but in those days when tuning was zero this was an enormous amount of tunability.
Broadband referring to the 30 kHz?
30 kHz, that’s right. Life proceeded along those lines of using lasers to do different kinds of things, one of the things that I think subsequently has been forgotten is that if you really go back and look at which, at what is the first experience that was ever done, where somebody used a tunable gas laser, a tunable laser of any kind to do real spectroscopy, spectroscopy of something else, I think it was the work that I did hear on measuring the [???] profile of zenon gain transition by using a zenon laser which was tuned magnetically, so we would have one laser which was tuned, another discharged, in which we measured gain, and of course we obtained a very accurate [???] profile we measured the [???] so on and so forth. The important point is that I think in some sense, if when asked the question, there was a first time and somebody used a laser as a spectroscopy, I believe it was the first time it was ever done.
You know, going down this list, I am curious whether the, when you went to Paris for the Quantum Electronics Conference whether that conference was in anyway had an impact on you, or whether anything went on in the informal sessions which is worth talking about? That seems to be the first of the Quantum Electronics Conferences that you went to.
Yes. Two things happened. The first thing that I learned, how not to give talks. But, you know, here I was with something like 400 new laser transitions that had been [???] since the previous Quantum Electronics Conference, and so I reassured every single one of them, you know that’s what you do. Anyway, the two or three things happened. First, I met Nico Bloembergen. He was one of the organizers of that meeting, and I think I also met Charles Townes, I think those are the two people that I remember at that meeting. And surprising they were human people. Anyway, that’s one. The second most important thing that happened there was that I came to the conclusion that finding just another new line was not the important thing to do in Quantum Electronics anymore. One had to do something beyond that, namely the question of, lasers that had something special for them, either high powered, something other than just yet another line which produced a few million watts of power.
Now, why did a conference induce that?
Because the Quantum Electronics by that time, was beginning to move away from just lasers to obligations of lasers to science, not only the optics had been around for about a year and a half or two years, a year in a half roughly, lots of experiments had been done in second harmonic generation, 2 for 1 absorption experiments had also been done, and it was very clear that if one were going to use the gas lasers for that kind of application, that kind of scientific studies, milliwatts won’t get us too far.
So I get the impression that this really in some way expanded your horizons to lasers, what was going on in lasers?
That’s right. That’s one. And secondly I think it allowed me to put into perspective as to what the role of that kind of lasers we had been looking for until that time was. A spectroscopic tool for low par applications what we were doing was fine, but there was a whole class of new types of studies which really required high powers, and gas lasers would never make it, not the gas lasers that we had at that time, and that’s what really soared the germ of high powers into my mind. And also, told me also, I think I learned with some conviction that there was a good reason why the world did not pay that much attention to gas lasers. The point was, that things that you could do with it, were very much limited. But we didn’t stop looking for longer living transitions but I did go on into looking at other things, and until that time we had only looked at what you call, noble gases. And the point was there was a reason to go on looking at other atomic species. Such as carbon, nitrogen, sulphur, and so on, and these if you look at the energy level diagrams and the lifetimes and the kind of electron temperatures you would have in a conventional discharge, you can show that if you had atomic species of that kind in this discharge, you should be able to get inversion. And one of the things that it was, it start with miraculous gases and to hope to dissociate them in a sufficiently hard discharge. And it did in fact work out quite nicely. But one of the crazy things that we found was that, when we use [???] in a mixture of gases we found the same transitions that we have seen in the case of neon oxygen and high one oxygen gas, up to a few years ago, except that the frequencies were slightly different. What you found was, instead of having a single transition which is what we found in oxygen at the 8446 angstroms what we would get out would be 4 or 5 4 different lines near 8446. Closely spaced so that your high resolution spectroscopy to be able to separate dissolve. And there was no such transition in [???] so it was a mystery to us as to what this was. One of the experiments that we did was to measure the splitting very accurately. And, Bill Sharpless was making germanium point contact diodes, those days, he was at Holmdel and he was making these diodes as detectors for microwaves. My argument was that, if this detector is about to detect microwaves, so it has a frequency response up to 10 kHz, germanium will absorb the radiation of 8446 angstroms. So one should be able to do hetro dining studies. Take these various lines that you saw from the [???] laser coming out and look at the beat frequency, because here all you are using is really photoconductivity, using photoconductivity of germanium to do that. Surprise of all surprises it includes germanium photodiode in fact detected 20 kHz beat free it seemed. The first time anyone had done optical hetro dining experiments at that high of a frequency. And we did measure the splitting quite accurately. I don’t think that the story is closed, but I suspect that those transitions don’t belong to Blooming, I think they still belong to oxygen, which was a small impurity and there are a number of possible explanations but we won’t go into those explanations as to what is happening. The point was that…
So the investigation of broming really what it yielded was the use of this new technique that you were using to?
For doing optical hetro dining. I mean that’s what it really turned out. It was good for that, but not good for anything else.
Let me just ask about that. In that, that’s in one of the articles that I didn’t read. And I think. And you title it Optocohetra Dining, was it that that came out as a conclusion and you emphasized it in the article but you went in really with some different object in mind? Like explaining what was going on in the brome?
No, I had twofold objectives. One was, the first one was to find out what was going on in brome, the second equally important one, was to look at the frequency response of some of these so called microwave detector diodes. One couldn’t have done that because nobody had two optical lines separated by 20 kHz. Here nature had already given them to us, and so therefore it was on the one hand we were going to check out if these diodes worked at this high frequencies or not and secondly it would have given us some a better feeling about what exactly these splitting were between these lines and might give us some idea about how it connected with brome.
Now I do want to just interject just a couple of questions now. Did you also go to the meeting at Polytechnic Institute in the spring of 1963?
No, I did not. No.
Okay and another question I would just like to put in now is, were your reading habits changing during this time? When we left you in graduate school you weren’t reading much. Were you still not reading extensively or had you begun too, I just want to get an idea?
No, I think the, when I came to Bell Labs, like I said the entire literature of lasers was only about 1 inch thick and then it was very easy to start reading because you didn’t have to go through 20 or 30 years earlier to find out what the hell was going on, so in some sense that was a gradual learning process of keeping up with what else was going on. Yes, I think the reading habits change over a period of time and sometimes I wish I didn’t read as much now. I do more and read less. Anyway, coming back to the question. So I came back from Paris with a definite feeling that had to go to a different kind of system where powers were high. Second aspect that was pointed out and I don’t recall whose talk it was, but probably one of the high power lasers, solid state laser talk where somebody emphasized the question of efficiency. Efficiency of lasers, how much optical power came out and what went in, and that both power and efficiency concept is something that I came back with from there. On both those grounds the lasers that existed until that time, another conference that I remember and had a major impact on me was the Solid State Meeting in Baltimore, APS meeting in Baltimore which was March of 1963, and sounds about right. And the specific session that had a major impact on me, was the session in which there were a number of talks on nonlinear optics. Even, though I didn’t do anything with nonlinear optics for many years, subsequent to that, I think to me that was an important meeting to have gone to, and I think if somebody asks me even today, is there any particular APS meeting that you remember, that’s the only one I do remember. I think I genuinely felt that there were lots of new things that I could still learn and have a good time and someday I would go into nonlinear optics, so that was a time when I thought that there was a good field to go into. Coming back to the question of higher powers and what not. From the middle of 1963 on, I think much of my activity in the area of going towards longer wavelength transitions [???], I mean it went on much of the work either was carried out by Faust, McFarlane or by my assistant. But I was spending a lot more time in the laboratory trying to understand what limited the efficiency of gas lasers. And how can one go about getting higher powers. One of the first things that I did was, I definitely came to the conclusion that all the possible atomic species that were going to laser had been lased off and on, that there would be a few more, but they would have the same problems, that we have faced earlier with any of the noble gases or with carbon nitrogen, sulfur and so on. They would work, but they were going to be low powered devices, because you could never get high enough density of this species. We did miss out on the argon and crypton iron lasers. Eugene Gordon in the [???] material had been working on that, Eugene Gordon had been working with Bill Bridges and I shouldn’t say misstated, we knew what was going on, but I could never get excited about passing 40 ms of current through a small tube, it just somehow sound like, it wasn’t a very pleasing way of doing things, and then the point was that somebody else was already working on it, so what one had to do was something different anyway.
Were you much in contact with Gordon?
Yes and no. I think we had our constant run-ins about what was right and what was wrong, and you know, we all get very buried to our children. My child was low pressure gas discharges, and low power gas discharges and his baby was he was going to drive as much current as he could through this temperature.
This kind of lunchtime conversation?
Yes. Mostly lunchtime, that’s right. But the, this was when in going through, you know I was trying to learn more about spectroscopy and I came across the first two volumes of a book by Hertzberg. That was my first exposure to, my real exposure to molecules. I mean that I had learned about molecules in the Quantum mechanical courses at Stanford, but you know that’s different. The first time, I realized that there are different kinds of energy levels. And the thing that struck me, absolutely, I just wasn’t thinking, the thing that absolutely struck me was that energy levels of molecules were packed a lot closer together than those of atoms. Everything happens in a, of course these are vibration energy levels as opposed to the electronically energy states, but that is something that you can work with. You can put your hands on something and begin to ask important questions about can you get inversion amongst them, what are the transition metric elements, what sort of gain you would have and so forth and so on. And, it was that, that, I don’t recall but sometime second half of 1963 that I really started spending more time looking at properties of molecules, and in the early part of the game I convinced myself that a diatomic gas would not be the right thing to use. Why? Probably because you can show rather very simply, it doesn’t take, even for me it was possible to do, than it comes to Hertzberg, is that the metric celliment gets larger and larger as you go to higher and higher miraculer states, vibrational miraculer state for a diatomic molecule. Because the molecules, the atomic species vibrate more, so it has a bigger [???]. What it says, is that the lifetime gets shorter as you go to higher vibrational states. Which is backwards of what you wanted for making a laser? For making a laser you want the upper level laser lifetime longer than that of the lower level. So at least for the time being I dropped diatomic gases which means I had to go to the next volume of Hertzberg, and the simplest system there is, is a next level is a triatomic system. And having recognized much earlier in the game that if you have a gas discharge with carbon monoxide for example, and run a discharge through that, you get a fair amount of dissociation. And so one of the things that I had learned from going after the atomic transitions and carbon nitrogen, was that I had to look for very stable molecules. And carbon dioxide is about as stably a char molecule as you can possibly make. Maybe there are a few more, but not too many more. As in additional to that it was a molecule that is simple enough to do back [???] of lifetimes. You can do that. It doesn’t require a whole hell of a lot of detailed knowledge; you can copy the order of [???], to the numbers. I think sometime in October or November, I remember coming back from the laboratory just totally excited about it, because I thought that I had found the system to do work on. Because this was carbon dioxide, because it had the right kind of lifetimes, you had the upper level with the longer lifetime, as compared to the lower level. But more importantly using complete wrong arguments I convinced myself that in an electron, in a discharge which in value obtained extradition by electron impact, there was a possibility of getting selected extradition of the upper state, simply by electron impact and carbon dioxide, because the upper state is semetric vibration. If you take a [???] just something coming in bumping something, the probability that you’ll excite the [???] is much higher than the probability of exciting an [???]. Because semetric vibration is going to hit the thing right, exactly in the middle at right angles. The other way you can hit at any angle and that is a reasonable chance. I mean that was the wrong argument but that’s not the point. The point is that I was becoming convinced that CO2 is going to work. More importantly when the draw the energy level diagram of CO2 the states which I thought were going to lase, which is the A semetric stretch to the semetric stretch of semetric bend, you find that that occupies lower 5% of the energy level diagram of a helium neon laser. Everything happens downhill, helium neon it happens up there, sitting way up there, [???] is here. And so the argument is that if I excited neon atom from here to here, get that much out, small amount as a laser radiation, has to come all the way down before that atom can be reused, so that gives you a lower efficiency. In carbon dioxide 40% of the upper [???] would come out as laser, I didn’t know if it would come out or not, but would come out as laser energy, so it has a click, a potential for high efficiency.
Well, we ought to talk a little bit at some point about where that paper is, in which some of these early things were?
Yes. Just about that time I think the middle of October or something like that, I put the gas into the system.
This is the first time you’re going into the lab with these ideas?
Yes. After about spending 3 months just trying to understand, trying to, spending 2 months for ruling out diatomic gases. And then spending 1 more month on trying to understand which triatomic gas I was going to try first. And further carbon dioxide was the first one that I was going to try. Put it in and sure enough the lasers lit up light gang busters.
What a moment. Now you were alone in there?
Yes. Either alone or with my assistant. Because gradually, our collaboration with McFarlane and Faust was falling apart, they had different interests, I had different interests and so we were sort of gradually going different ways. This happened I think sometime at the end of 1963, and I think there is a question of why did the thing appear in print until 1964?
Well, I just wondered about that you know.
No, that’s a good point. I had a badly infected appendix and I had to go into the hospital, and things didn’t go as well as they should have gone and I was just in the hospital for a couple of weeks and the doctors said, don’t go back to work for at least two more weeks, and so I said that’s a whole month gone so that happened, which was good, because what I had seen, I had gone to the hospital with the notion that I had seen these large number of lines both near 10 microns and near 9 and 10 microns, I identified what they were. The thing that wasn’t very clear to me was why so many lines would lase at the same time. I mean you know these are vibrational, rotational lines and had to work out the details of what the shedding of vibrational energy amongst rotational states was so on and so forth. So that I worked out after I came out of the hospital, and sent for publication at that time. That thing appears in print in two fashions. One may argue that it appears first in print as an abstract, for the APS meeting; this was in Washington in 1964, that abstract had to be sent in sometime in December, and that abstract has three authors I, and Faust and McFarlane. Apparently because for the old time sake I didn’t want to get into the same kind of problems that I had run in with Bennett, so I decided that even though they had little on the thing to do with that work, I’ll keep them on. And the first paper which was the explanation of what was going on there, which has only me as author. So it’s the end of 1963 the beginning of 1964, the time difference had more to do with the fact that I was not available to work on it. But that to my mind was tremendously exciting. Now this is what I meant earlier when I said that I wish that I had not gotten so much [???] to using a single gas system. Because had it not been that [???] to using a single gas system I would have immediately picked up on the fact that one can use something else to transfer energy to carbon dioxide. Pure carbon dioxide worked very well by itself and we were getting hundreds of milliwatts of power just from pure carbon dioxide, so I mean there was problem there. About that time I read an article by F. Kaufman, I don’t recall his first name, I think Cincinnati, and does that make sense? Somewhere it matter. What he was working on was looking at vibrational energy lifetimes and storage in nitrogen. And he was carbon dioxide as a means of diexciting this vibrational excite molecules. And so the argument that I used was that, if that’s the case and mean obviously nitrogen must store that energy, because he claimed lifetimes of the order of the second for the first vibrational state of nitrogen, it matches very nicely with the upper laser level in carbon dioxide. And I made a few notes in my lab notebook and I asked my assistant to build the flowing gas system where I would be exciting the carbon dioxide separately, I mean nitrogen separately from carbon dioxide. This if you recall is a direct carryover from what Kaufman had done was to look at the energy that is carried by vibrational excite nitrogen molecules with and without carbon dioxide injected some point down the line. My argument was that, if I was ever going to be able to show that the N2CO2 system had any advantage you had to excite it separately so that you had a laser region in which there was no discharge, that’s why it was [???] that way and I gave this sketch to my assistant for building and I went off to India for a vacation. For six weeks. I came back and two things happened when I came back, the system was ready to go, which we turned on and it worked. Again, you know as expected, much better than expected because very quickly we recognized that if you didn’t do that and if you run the discharging laser region with nitrogen and carbon dioxide together it works even better. Which is what you would expect, if for no other reason, for a simple reason that you do not lose any of the nitrogen molecules as they excited nitrogen molecules, as the flow from the discharge region into the laser region, that’s one, and secondly, you also take advantage of the fact that electron impact excavation of carbon dioxide is also then defected, so you have two mechanisms by which will be exciting carbon dioxide. But, another thing that happened at that time when I came back, and we had made the thing work and we sent the thing into the patent department for a possible filing patent. This was after we had sent the thing for publication, which is what we do very often when things are scientifically exciting. We will send it out for publication at the same time make sure that our patent department will, be able to file if necessary by the time the publication appears in print. The thing that he brought to my attention, this paper by Legay which had appeared in July issue of Com [???] News. Where he had made this very specific proposal. Was that why not use the transfer of vibration from nitrogen to carbon dioxide?
Which you had already done?
Which we had already done. But the date on that paper was July 1964, which was when I was gone, but that’s doesn’t matter. The point is that we had done this. The powers at this point really started jumping and what was pure hundred milliwatts up to 4 or 5 watts, at which point we were convinced that we were on the right path, that this point there was simply nothing wrong in that system. One of the things that happened and I don’t recall where it had happened, but somebody pointed out to me that we probably still had a bottleneck in terms of the lower level de excitation in carbon dioxide. Because it is not [???] relatively.
Somebody, you mean somebody when you talk to…
Talk to somebody here or at a meeting or somewhere there, I really do not remember where it happened, but you know, there may be a genuinely bottleneck in the lower state and so if you find some way of de exciting those things faster, lower levels faster, you might be able to get more power. About this time I came across a paper written by it somehow skips me. Somebody from Phillips Laboratories in Andover, who had been looking at the de excitation of carbon dioxide by addition of water vapor. The argument was that the lower laser level matches very nicely with the first excited state of carbon dioxide, I’m sorry, of water vapor, the lower level and carbon dioxide matching first excited state of H2o. And that if that were the case, water would act as a very rapid relaxer of the lower laser level in carbon dioxide. And so we added a little bit of water vapor to it, I believe that the power went up even more, at which point you go through the literature and find that helium does equally well and helium is such a, so much nicer gas to use than water vapor. And that was when we added helium and we had the first laser that produced more than 100 watts of CW power of any kind, and I think that at that point laser community started to think about the fact that solid state lasers may not be the only candidates for high power systems.
Now we really, I would like to stop here, we’re now right in the middle of 1965.
We’ve been speeding along and we should go back and pick up some of the context in that, who was saying what around you and what was the reactions of various people?
I will take a 3 minute break at this point.
I think the first public place where I talked about this work; the high powered work was the APS meeting that was held at Columbia University in a session chaired by Schawlow not Schawlow but Townes.
Now this was the work already including the helium or?
No. This was the sort of work that included water vapor, but not helium. And this was in June 1965, when the power outputs were in about 20 watts region. And this was the summer APS meeting which was held at Columbia University given an invited talk there, and a couple of weeks later I gave a similar talk at the Puerto Rico Meeting. Now Puerto Rico meeting turned out to be one of’ the more exciting meeting I had gone to for two reasons. First, very selfish reason. I was one of the 2 or 3 heroes of the meeting, that’s sort of an exciting point, something that was genuinely new and unexpected that had happened at that meeting, and there was an old contingent from the Soviet Union who had not heard about the carbon dioxide laser until they came to the meeting, and there were a whole number of people who had not known much about it. So that was a good meeting for me. Secondly, I for the first time met Basou and Prokhorov and my comments about them shall remain to myself. But this was a meeting which was held by almost by invitation only, which means that it was a small meeting and that allowed an enormous amount of time and scope for discussions with others. And the second thing, the other hero at this meeting, or other two heroes at this meeting were Giordmaine and Miller; they had just made the first optical parametric oscillator work. And so, that was a good meeting. Another reason why this was a very good meeting was I learned how to sail. I didn’t know how to sail, and I learned how to sail there, so I came back both intellectually as well as physically restocked with new things to do.
Well listen, were there any people you met there for the - first time that had any significant impact, or who enriched you, I mean you meant Basou and Prokhorov, but what about other people who were?
I think, by that time you know this is not 1965, I’ve been in the field for years and so at least I knew almost everybody who was in that field by that time. And much of the time that I spent there was spent not on trying to find out what more one might want to do with carbon dioxide lasers but more with trying to find out directions in which one might go in using carbon dioxide lasers for scientific studies, and again I think this is where I found some resistance on the part of the people there. In the sense that most of the people who were there, were used to using lasers in the short wavelength region where the technology is a lot easier, detectors is typically where one might either use a formal or a photodiode for detectors, windows are simple, because by that time I had learned about sodium chloride and potassium chloride windows. I knew they were transparent materials in infrared. But certainly for example, fast detectors at 10 microns had to be cooled with liquid helium, so there is a general degrees in complexity in terms of what instrumentation you need and I think those are clearly indicated as obstructions or something that might slow down if somebody wanted to do science using this long wavelength lasers. Although at the same time it was very clear that the resistance to extending spectroscopic nonlinear and other studies to long wavelength regions was becoming obvious to lots of other people and so I did come back with the idea that time was now or never, in terms of making the first jump into doing solid state physics using carbon dioxide lasers. One thing we jumped over was that while all of this high powered work was going on we did look at other triatomic molecules. N2O and CS2 specifically, and both of which turned out to be not rather useful systems. Not quite as useful as carbon dioxide because neither one of those molecules is quite as stable against dissociation in discharge as this carbon dioxide. To the point where you cannot make sealed off lasers using N2O or CS2, but you can make sealed off lasers using carbon dioxide, and so if you want purely practical matter, N2O and CS2 aren’t used that much, but they are still candidates for [???] and a number of people at NRL have in fact made very high powered pulsed CS2 lasers there is one big advantage in CS2 the rotational lines are spaced very much closer than they are in carbon dioxide and so you don’t need the same kind of high pressures that you need for carbon dioxide to make that gain distribution completely uniformed and so that you can make a completely tunable CS2 laser or a very broad region using CS2 than carbon dioxide. So it has a different twist, we have not done much work other than show that it in fact does work and give our indication of the lines.
Now, I also want to find out, did the Department of Defense come to you at this point. You get this very high power.
Can I interrupt? I was going to say that, I was going to say that in connection with why we stopped at 120 watts or whatever it was a watt of CW power from carbon dioxide. But before I get to that let me point out something else. That’s a fact that sometimes you learn by mistakes how not to do things, and one of the things that looked very exciting in the beginning of 1965, that about was, that oxygen carbon disulphide system would be awfully good, because the levels are matched much better with oxygen and carbon disulphide than with nitrogen and carbon disulphide. We had made nitrogen and carbon disulphide but that’s with a fair amount of energy discrepancy. Oxygen is a lot better. Not being experts on vacuum technology, so you charge ahead, so you know this is a continuous flow system and so this discharge in which oxygen and carbon dioxide are being pumped out. Now anybody could have told me that, if I’d asked them, but if you do that your pumps will blow up and our pumps blew up, it went Pafoom — a major explosion. How not to do things. Anyway, but nobody to the best of my knowledge has made an oxygen carbon disulphide laser work yet. For many reasons, but certainly for the reasons it’s not a system that you can just play with very lightly. Coming to the question about high powers. The important point there was, I think within 3 or 4 months after we had announced the upwards of 100 watts CW from a carbon dioxide nitrogen heating laser, at that time I was not a citizen so I could not have gotten a clearance heard rumors that the Defense Department was essentially duplicating what we had with a much longer tube to get kilowatts a par upward. And it was classified at that point and at which point I made the decision, I can’t remain in the game, because if I remain in the game I’m always at a disadvantage. And so I decided that the time had come to do things with the carbon dioxide laser then continue to work on it.
That’s a very interesting point.
I don’t feel sorry about it; I think it was just as well because you ought to make a change. And I think any mechanism that you use for making that change faster is good for you.
Well, but you know I’m always interested as to whether the existence of classified research makes a difference in the way people are carrying out programs, it’s just an interesting little
Well yes and no. I think in this particular case, going to higher and higher powers had 2 basic things against it. One, I couldn’t see how that made sense in the context of Bell Laboratories. That’s one. And secondly, with the enormous amount of resources that the Defense Department can bring to bear upon this problem, it can get wiped out in half a day, I mean; I could never mustard up a group of about 100 people.
Okay, so it’s not the classified?
No, it’s not the classified. Classification in this case was only one of the things that were a problem, but when an army starts working on it then you better get out. And so, there was a point when I made a decision that is clearly easier minds to be digging in rather this one. So that’s the direction I decided to go into, but before then, let me come back to one question. I think there was a question somewhere here I had said why did you really decide to come back to carbon monoxide after deciding that it was no damn good? And the reason was twofold; we had shown that under false condition, carbon monoxide will lase. That you can show, because lifetimes don’t kill you there because if the excidition occurs over a very short period of time, for a short period of time you can get inversion. Our colleagues in Japan, in France, the Legay’s group, had picked up on the nitrogen carbon dioxide and they had shown that if you cooled down the laser tube to a liquid nitrogen temperatures N2CO laser will also in fact work seductive, So we did come back to that for a little while just to see that it worked and to get some idea about how to make and to CO and CO laser CW for a future reference, in case we needed it. And we’ll come back to that and you’ll find that we came back to that after many years to use that laser as a source for some nonlinear optics experiment. So in some sense, when we came back to many years later it was because others had carried it much farther than what we had done here.
You also have the chemist, are really getting very busy at this point, I wonder, you know getting chemical lasers going, I wonder if that intersected at all?
Not very much. Because, having made up my mind at this point that I was going to get out of making new lasers, I didn’t want to turn back at all. There was an earlier situation which doesn’t appear anywhere in the print, anywhere here, but I think might be good for history was that one of the earlier interactions with a chemist, which I found was very exciting was just after we had made the high power CO2 lasers work, I was visiting Berkeley, specifically, I was visiting Palanti’s.
He’s in Toronto. Pekatean? And Pekatean was out, but he was out, and one of his students Casper showed me around and he had showed me the system that had just made up and this was the 2H5I dissociation laser. This is dissociation with a flash lamp and the [???] lasing at one point 3 microns. And the second system they showed me was the HF system which was dissociated of hydrogen and I forget, but whatever it was, but it is flash.
I guess a uranium fluoride.
I forget but whatever it was, dissociation of some flurene containing compound and that was a chemical laser and I think we had a rather lively discussion about the fact that the system is very similar to CO, the HF system, and many times of its inversions in watered lifetimes, which also applies to that, lifetimes gets shorter as you go to higher states and that does require something very special to be done and the chemical dissociation and the real association clearly provides that pattern. When I came back, a new individual had joined Pekatean’s group a fellow called Marty Pollock, and he was looking for things to do and I suggested, that why don’t you look at other sorts of chemical lasers that you might be able to make using the flash dissociation and the combination technique. And one of the things that he did was the dissociation of carbon disulphide in presence of CO, in presence of oxygen, which gives it as to fairly intensive laser action on carbon monoxide lines. It’s interesting, when the work was done and he wrote up the paper and he gave me the paper to read through, and he had put down my name as a co-author, I said look I didn’t do anything, you know, I told you what to do, and everything was done by you and I took my name off and I think that was the right thing to have done, but the point is that, that’s about the only association I have with chemical lasers, I tried to start an effort here and Marty Pollock’s interest changed subsequently so we, nothing much got done in that field.
By the way, all this time you were not head of anything; I mean you’re this entire time still just scientist?
See that’s the tremendous advantage of Bell Laboratories, and I think by this time I probably launched four or five other people on projects that came out from things that are happening here. And, Bell Laboratories allows that kind of leadership without striating people with [???] duties. It’s a defector leadership without, and so that’s very good, because what counts is the ideas, when that happens. But a person who played a very important role in my getting into the field of nonlinear optics was Gary Boyd. He had his lab right across from my lab and one day I walked up to him and I said look, I’ve made up my mind, and he had been working in nonlinear optics for maybe four or five years, fairly long period maybe three years, I said look you know a lot about it, can you give me some basic things that I can read and also get some ideas as to what I should be looking for. And he said look, this, this, this and I think he was very instrumental in telling me two things, he says look if you want to do second harmonic generation, the material must have, must be transparent at both frequencies, that’s one, for the second harmonic, so you better start with something phenomenal, secondly is that it should not have centroff in version symmetry and third if you want to do phase matching it must be by the birofringent, and go look for it, because that’s how all of their materials are found too, and I think this is the only way we can do it. And some years earlier, this is something interesting, some years earlier when I was looking for materials which would be transparent in the wavelength region for windows, one of the materials that I had come up with was tellurium, tellurium is transparent beyond 3-1/2 microns. The thing that is wrong with it is it has an enormously huge reflective index and it is birofringent, it is very strongly birofringent. And I remembered that from way back when, that was his material, so I said hey this is a good idea, I looked through the book of crystal structures and surely enough the crystal structure of tellurium is D36 or whatever it is, which is a no central symmetric structure. The question is where do you find tellurium? And so, I found that there is a company in Germany called Wacher Chemical, used to make it, so I got a hold to them, they sent us boule down, I did the second harmonic generation experiment in that phased match, and it worked the first time. That material turns out to be the material which has the larger known, known material with larger nonlinear coefficient in any region of the spectrum.
Now until this point is practically no infrared?
There is none, this is the first experiment. This is the first nonlinear physics experiment of any kind that had wavelengths longer than 1 micron. 1 micron being the medium glass laser. And it turned out to be a very, when you start off in a new field with a bang, at least I find that to be a lot of things to carry on and do other things, and of course, I found lots of good materials which are equally good some are better for reasons such as they are harder than tellurium but not quite, but do not have quite as high nonlinear coefficient and so on.
Miller is one of the people that you mentioned this connection in your paper?
Right. See Miller was here and he was not using any regular contacts, I would consider him the grand-daddy of nonlinear optics at Bell Labs, he and Giordmaine. If you remember that I mentioned to you that he and Giordmaine were the other heroes at the portico conference because they had just made the optical paramedic oscillator work. And I talked to Miller about how to estimate the nonlinear optical coefficients of materials, and he had an imperial rule which connected the linear reflective index to the second order of nonlinearity and so I had used that as a guide to find out how good tellurium would be, the tellurium had not looked that good I would never had tried it, the point was that it was very clear even before we had tried it that there wasn’t going to be another material with that high nonlinear index anywhere, because it has the highest linear index. It goes to the third power of the index, so. But it is at this point that the CO2 laser was you know, after the experiments in tellurium CO2 laser was being taken very seriously, began to be taken seriously by people who were doing physics here, at Bell Laboratories. Specifically, Peter Wolff. Seeing that in fact that the CO2 laser is good for something other than making holes in anything and everything, he started working in the field of response of electrons, free electrons, in semiconductors when subjected to a high powered infrared radiation. And what he was working on primary was in the field of doing Ramad scattering and of course he showed that application of magnetic field will give you discrete energy levels for the electrons from which you can in fact scatter with fairly high efficiency, so this was the article work. Just about this time, Peter Wolff, then was a Department Head, and hired a new MTS a fellow called Dick Slusher and
Member of Technical Staff, which is what I was too. He had hired a new member of technical staff, a new scientist, Dick Slusher did his Ph.D. under Alan Portis on nuclear magnetic residents, he came here and he said he had heard enough of nuclear magnetic residents and he was going to do something in optics. Peter Wolff had never clone anything in optics and so he says, why don’t you go work with Patel he knows all about optics and the C02 laser has just broken ground in nonlinear optics also and maybe we can find something else to do. So Dick and I got together, and Dick said look, I know some solids for physics I can teach you that, if you teach me everything about lasers and maybe we can get together. The first experiment we decided to do was to do, light scattering from plasmas in the semiconductor, such as emitting light or [???] this is ramas scattering from plasmas. And when we stuck the sample into the beam, and looked for an approximate place where we thought we should see the rama scattering is in spectrometrist, we found a large number of very closely spaced lines and not one but many, many lines. And, this was in lead telluride.
I’m sorry, I missed that.
In material called lead telluride, and what you found was of course that what we were doing was, we were optically pumping the material by twofold on absorption. Twofold on absorption of the 10.6 micron photons giving rise to population inversion in lead telluride giving rise to laser action and the band gap. That was not what we were really after, so we decided that that was a bad material to look at, because the band gap is so small, that you always get twofold on expedition no matter what you do. So we went to a larger band gap material in the emisanide, again we looked in the same place again we found different set of closely spaced lines, now what’s going on? Of course what is going on was that we had discovered the third order of nonlinearity of free electrons. See, the earlier nonlinear experiments that anybody and everybody had done, involved second order of nonlinearity which arises from the lattis vibrations from the lattis okay, not from the free electrons. Now free electrons because they have an inversion symmetry, I mean electron react the same way whether you plant that way or plant another way. Can have the lowest sort of nonlinear which is a third nonlinearity and what we had discovered was the third nonlinearity of free electrons. Wolff carried out the theory of that and we published a paper back-to-back physical of the latest experiment in the theory.
I’m kind of impressed that you’re suddenly jumping into this whole new field of semiconductors was that something that you had to spend a lot of library time on?
That’s right. Library time, buy books, read, change your profile in what you read in journals and things of that kind. Change the kind of people that you talk to down the hall.
So, you really kind of changing your universe a little bit?
That’s right. I told you the decision was made when I found out that the defense department were building a thing which was 20 times as long as mine, I didn’t even have space for putting in the lab. So I had to do something different. And that field of optical nonlinearities, from electrons has turned out to be a rather very interesting field. We didn’t do very much after that point, after the initial beyond the initial experiments, but it has turned out to be a good way of measuring a variety of parameters of free electrons. Because you can measure these parameters without making physical contact to their sample of semiconductor. Light goes in and lights come out, so you can study the properties of electrons without touching them. But in all of this what we were trying to was in fact, was to look at drama sketching from plasmas, it turned out to be a lot more difficult than what we thought it would be. Probably because you can’t tune plasmas if you find one sample I got only one line essentially, because the line which is shifted by the plasma frequency of the semiconductors, because tunable it is not there, so if you are working in a very low noise, very small situation, now remember that the detector is here they are invisible. The detector’s here are made 7 orders are made to do worse. So, and unless you’re going out for small samples, if you, you can tune your frequency of what comes out you will never find it So we began to concentrate on Peter Wolff’s idea on looking for life scattering from landauables of electrons in semiconductors. You take a semiconductor apply a magnetic field, electrons are in discrete levels and you look for a scattering from landauables of electrons and semiconductors. And at this time we picked up another colleague Paul Fleury. We picked him up because his lab happened to be next door to mine, and he had done light scattering for his thesis, he had just come from MIT, and so he knew a lot about light scattering, Dick Slusher knew a fair amount on semiconductors and I knew a lot about lasers. And so that collaboration turned out to be exceedingly useful.
Now again, there are a couple of things that I don’t understand that I would like to pick up on if we can make a pause. One is that, you did this one experiment on tellurium at this point, you told me before that other people were really ripe for going into the infrared; did a lot of people start doing nonlinear optics of the infrared at this point?
Not, at that point, I think the nonlinear optics 10 microns really erupted after we published a paper on the third order of nonlinearity and semiconductors, at which point it became clear that its only one kind of nonlinear laser, a whole slew of nonlinearities that there are going to be studied, and specifically the group that might, jumped into it quite vigorously because he has much higher magnetic fields of a level at the national magnetic laboratory and so he did many of the experiments, say for example the twofold experiment which I mentioned in lead telluride in high fields and so on. I think that program might even had been done for almost five years, looking on twofold of an idea semiconductors, using CO2 laser as an expedition source.
Now another thing, is that I had the impression that all this time you were looking for tunability, now is that just a false impression?
No, it’s not a false impression, I’m still looking for tunability, I haven’t gotten there yet. It’s a goal which is very far off, and we are very close to getting to the tunability because the idea that Peter Wolff had proposed, namely using landauables, probably light scattering, which is your plaminated field, and the spacing between these levels are now proportional to their magnetic field their supplied, and so the frequency of the scattered line is tunable by changing the magnetic field.
Now I guess what I’m really asking you there is that in the account that you’ve been telling me so far, the motivation for getting a tunable laser has not played a part in this account and I wonder if it played any explicit part or if it was just one of the sort of background things that you had in mind or?
I think there was a goal that was set in the background and broadly directed me in specific fields, nonlinear optics is one way of getting tunability, parameter oscillator is one example, light scattering from an energy levels whose spacing’s can be changed is another example, but all of these then required first the high powered laser, second required studying of nonlinear properties, none of which will source will, in some sense with 20/20 eyesight one can argue that the course of action was absolutely correct, at that time we did not know if you were ever going to make it to the final goal of getting a tunable source.
I guess really I’m just trying to see as we go along without the hindsight you know, how one experiment may or may not have.
You know I suspect that this kind of work could never have been done at any other place. Because we may have had shorter goals but its connection to our long term goal was best tenuous. It was in generally the right direction but you couldn’t show a connection one, two, and three as to how those were connected.
I see. But there was a context here?
There is a context right, that’s right. And so, eventually we got there, but as long as what came out in between were good signs, I was happy and my management was happy.
And that’s still Cutler or somebody else?
No, it’s still Cutler, no wait a minute, by this time it is Solomon Buchsbaum. I’m sorry. By the time we’ve come to the high powered lasers and the nonlinear optics.
And he’s the person that would be your management figure or there would be somebody else in between?
No, there is Tien in between. Tien was still my department head until 1967. Anyway the point is that, we did the experiment on scattering from electrons in a magnetic field and of course found the one process that was proposed by Peter Wolff and the other process which is a spin flip scattering proposed by Yako Yafet here at Bell Laboratories, and it turned out to be that the spin flip frosts was very much stronger, at which point I decided that my interest diverged sufficiently from slusher and floodies that I was going to deward all my time to making a tunable laser out of that using the process and this was in late 1967, other things happening between and one of the things that, happened that is rather, interesting has to do with the work in the transparency and photo in gases. One summer, I think it was the summer of 1967, Dick Slusher’s professor Irving Hahn came and spent four weeks or something like that here. Irving Hahn and one of his graduate students Sam Mc Call had just discovered certain new transparency in a solid material, I think the sample of ruby or something like that, and one day, Dick Slusher and myself and Irving Hahn were sitting in my lab, talking about things, the experiment was running on the landau scattering by itself, so all we had to do was make sure nothing broke down, so meanwhile we are talking and I asked him the question, I said, Irving, how do you use transparency you are talking about passing the light through your sample and then looking for the reshaping of the pulse and delivering everything else. But you have really no control on the parameters that determine the delay and the reshaping, the parameter is the reassure of the pulse with to the videas lifetimes of the absorbing the medium, because once you have a sample of ruby you are fixed, I mean you are stuck, you may take another sample and put it down there but that’s not the same thing as having a continuous. Wouldn’t it be damn good if you can think of a system where it can change the properties of your absorber in a continuous manner? He said that’s a super idea, I said how about doing it in gases, is anything wrong in doing transparencies in gases. He thought for a-while and says that would never work, I said how come? He says, look the point is the following, that in order to use self-induced transparency, first half of the pulse defacing your various absorbing dibels, and the second half of the pulse has to rephrase them, but if the absorbing molecule is physically moving, which is in the case of gas, he didn’t think that the ideas of self-induced transparency applied. So we battered back and forth, we couldn’t convince him and he couldn’t convince us, so I said look there is something called nature, you try it, I mean at this point we are not going to convince ourselves, we tried it and it worked, and now of course you could change the pressure the gas and thereby change the lifetime, the collisional lifetime of the molecules, so that you have a continuous tuning of the lifetime and the pulse too.
Tell me when you do something like this, how long does it take to put an experiment like this into actual practice?
It depends upon how difficult an experiment it is. The experiment on sullivent news transparency took probably about a week.
Set up the equipment.
Set up the equipment, because we got the laser running, all we had to do was to have a long enough pat containing sulphur which is the absorbing gas and set up property [???]. So that takes a week. Experiments like Raman scattering from electrons in a magnetic field, six months, because it requires a super conducting magnetic which we did not have, which we had to order. So it depends upon the experiment. The studies of nonlinear effects in tellurium took us about three months from the time I thought of it, the delay was primarily getting the material from Germany.
And in these things, do you do sort of quick and dirty run through first and then a careful one?
Oh absolutely. I think that’s the way that I do it. When I say quick and dirty, it’s got to be sufficiently good, so that the measurements that you obtain can be used as a basis for other continuing or not continuing. This means that you cannot take shortcuts in terms of the physical experiment that you have set up, you may make shortcuts in terms of the amount of sophistication that you may build into the experiment. For example, subsequent to setting up, let me give you the example of magnetic Raman experiment that I was talking about. The first time when we carried out, we did not use a spectrometer for analyzing the scattered light that came out from the sample. What we did was, we used fixed band with filters and tuned the magnetic field so that we could tune the scattered radiation through the band pass of that filter. But that’s not how you normally do it, but that’s what I call quick and dirty. Because, it does not give us results that we can publish, but tells us whether we are seeing something that ought to be investigated further or not, and that takes a lot shorter time than setting up a spectrometer, because there is a factor of 10 loss in light going through the spectrometer, so you need better detectors and you need longer integrity times and so on and so forth.
I have to say that I’m very interested in this kind of material, which I don’t think frequently enough gets on the record.
No, what you write down is that this is a typical experimental data which may have taken them 20 years to get it, and that’s the only time it has happened. Anyway, the point is that, we did see self-induced transparent in gases for the first time, and once you saw that, it was quite obvious that photo-experiment, and in that particular case, Erwin Hahn was elementally against it that you would never see it because there is a physical time difference between the first pulse and the second pulse. And then there is a time separation before the pulse reforms, he says, this molecule that was here, let’s say that there are two or three molecules in one physical space, and those have been coherently excited by the electromagnetic wave, first pulse. Second pulse these guys have moved away, and you come with a second pulse, and they are still motivated by the third pulse when the reform. How do these guys remember that they belong to this group of guys that were together at first point?
Now that’s all happening in this first conversation, because there is a year between these two papers.
Yes, that’s right, this is all happening in the first conversation and of course, he says, no wait, no because by the time it comes to reforming it will all be gone. Well, the argument is that he now remembers now where he was really wrong, the place where he was wrong was that you can’t talk in terms of molecules, all you can ever talk is in terms of polarizations, and a polarization does not move away, things can move away but the polarization still sticks then in the physical space, it is moving in the velocity of light waves there.
That’s a most interesting case, where an experiment leads to a real understanding of theory. Just sort of a throw away, I guess?
That was a throw away experiment, it was exciting, interesting, and did not require tremendous amount of either in physical or experimental investment to get it done. Coming back to the case of light scattering from Ramad scattering from free carriers in materials, lots of other things that we did, and I think that you can look through yourself on different kinds of materials we finally did see Ramad scattering from plasmas also, but that came much later. But, from 1968 on, in some sense, I spent a fair amount of time trying to make the [???] laser. About that time a post stock Earl Shaw, he was also from Berkeley, and so we could spent a lot more time on it. About this time I also had become a Department Head, so I had other duties in addition to just doing my own research. The interesting thing about being a Department Head in the Research area, is that principal number one, a Department Head is evaluated at least 50% on his own research, so continuing personal research is an integral part of being a Department Head in the Research area. Of the remaining 50%, maybe 30% is determined on being able to provide technical guidance to the people who work in your department, but that is something that I have been doing for the last six years anyway, now for the young people that came into the system, and the remaining 20% is mishmash and everything else. So, you can see that 80% of the work that I started doing as a Department Head was something I was already doing, so being a Department Head did not significantly change my mode of operation, except certain times in the year when you have to do evaluations and things like that, but that happens only once in a year and one can take care of that on an ad-hoc basis. We were about to give up, Earl Shaw and I were about to essentially about to give up on the probability of making spent thrift Ramad laser. I mean essentially we had come to the end of the rope, nothing we did seemed to work.
Was surely the tunability that was exciting you, or were there other things that were good about it or what?
There were no tunable sources in tunable high powered sources in the infrared region, so that was a big thing to go after. And you know, if there were other sources, I don’t think I would have walked on it. Apart from the fact that this would have been a brand new source and nobody had ever made a tunable Ramad laser. This would the first tunable Ramad laser, so there are all kinds of firsts attached to it, and finally I said, look we don’t seem to see, when you want to see some fact, what you can do is to change the input power and plug the output power and see a break in the linearity, because at some point your experiment has gone over the threshold and starts lasing, so there should be increasing power output. And we couldn’t see that. I said look, let’s give that up, there is something that we are doing which is not right, and I don’t know what it is. Finally, we tried to do spectroscopy with whatever comes out and maybe at some point we’ll start seeing nettling of the lines, because spontaneous emission the line with spent thrift rained, spontaneous line width throws off the order of two wave numbers. Now you know that if it starts lasing, line width should go essentially to zero it is certainly smaller than anything that you can use for measuring it. So why don’t we do the following, why don’t we stick in a cell of containing ammonia, ammonia gas, which has all kinds of absorption lines in all the region where we do see tunable spontaneous Ramad scattering, so let’s measure the absorption spectrum of ammonia using the tunable spontaneous Ramad scattering, and change the input power levels and see at what point the line would start narrowing. And we tried that for a few months, and we got some inclusive results that perhaps we were seeing some narrowing towards and so we said alright look, let’s just give it one final try, we’ll do everything right we’ll tear up our other experiment, put it back together everything absolutely right and if it doesn’t work we’ll just go out and do something else. It worked eventually.
I think that’s a very different kind of experience, then some of these others where you walk through the lab and everything.
It took a year and a half, I mean myself and my assistant and because, you know a threshold experiments are very hard experiments to do, I mean, because if it doesn’t work you have nothing, you know there is nothing to fall back on, we have already done the spontaneous Ramad scattering experiments earlier, so we have gotten as much out from the laser as we could, so if it didn’t work there was nothing to show for but the time that you spent.
It would have been particularly bad for your young man?
That’s right. It would have been harder on him than on me, but the system has a way of taking care of that sort of thing. You know if somebody does a long term experiment the system will make sure that he gets a second chance. But it worked, and we talked about this at the [???] meeting in Dallas, I believe, this was in 1970. It worked in the late 1969, the thing did laser, and we talked about it in the meeting at Dallas as a contributed paper given in written paper at the APS meeting in Washington and the [???] the APS meeting in Boston.
What kind of reactions were you getting? Were there anybody’s particular responses to this?
The comment that I heard was at the end in talking at Dallas was that somebody not from Bell Labs he said, you know, after your talk people should have stood, up and given you a standing ovation, people knew that we were working on this and just not getting anywhere, and the fact that this was a first tunable infrared laser of any kind of high power output, first Ramad laser that could be tuned, I mean everything had just jelled together, this was a marriage of two very divergent fills, high powered lasers and really fundamental solid state physics had come together.
So the solid state people also are going to be very interested in this?
That’s right. And this was a solid state meeting this was not a laser meeting.
This job was given at a solid state physics meeting. And of course at that point the question came to how do you, so far that laser was using pulsed laser C02 laser the question how do you make it CW what kind of additional work do you have to do, what other materials can be used, what are the new phenomena we can use. It really responded a very broad field of very lively interaction after the talk and so on. That went exceedingly well, one of the things that we did very quickly after that was to do, show that in fact spendthrift Ramad laser is good for doing spectroscopy and infrared. And that paper and those results got us into a lot of trouble subsequently. What we had shown was that the language was very narrow and rightly some of our colleagues at MIT that language were narrow but not as narrow as we had claimed they were, because what we were using was absorption line widths in ammonia to get a handle on how narrow the lines were, the lines were clearly narrower than what our spectrometer will tell us. So we had to do [???] spectroscopy. This is what finally what I told my Department Head, what I wanted to do, I wanted to do infrared spectroscopy using lasers. Finally it was happening ten years later. That’s why that was exciting, because finally I could go and tell him, here it is. It so turned out that under pulse conditions the tunability is not as continuous as one would like, because the cavity has a spendthrift Ramad laser cavity has cavity moles and you get a small amount of continuous tunability, see what you are, doing is, there are fixed cavity moles you know separated by roughly 2 wave numbers and a spontaneous line that sweeps past this cavity moles as you tune, which means that if things were really against you, all you would see is these individual moles oscillating as you sweep past, and the thing in between. But in reality but happens is that, you get a fair amount of mole pulling and in addition to that at some points two moles will be oscillating at the same time so it appears like continuous tunability but not the same line width that you saw that you thought you heard. And so what we were seeing was somewhat broader line than what we thought we were getting, it really was not working in a single mode situation. It was a tunable laser but not running in a single mode.
By the way, who were the MIT people you were interacting with on this, particular?
We were not interacting, they were criticizing.
This was going on in the journals?
No, not in the journals, but at the meetings. After the meeting after the talk, there would be questions indicating what you do.