Elias Snitzer

Bromberg:

What we'll do is tell the tape that this is the 6th of August 1984, Hiroshima Day. I'm talking to Dr. Elias Snitzer in his office at the Polaroid Corporation. Well, I sent you a series of questions and we may or may not find it profitable to follow them. How did you find them as relevant to your memories?

Snitzer:

I found it very helpful in terms of reminding me of some of the activities in glass lasers. Yes, they certainly were worthwhile and I think they would be a useful basis for covering some of the material, though there may be some things that will go beyond what's on the questions.

Bromberg:

Why don't we start then more or less with the questions and then diverge as it seems reasonable?

Snitzer:

When I started at American Optical, I was hired by the Director of Research, Dr. Steven MacNeille. Dr. Steven MacNeille had been brought to American Optical to be Bryan O'Brien's assistant. American Optical at the end of World War II had been fairly successful financially. They made a number of important optical components for various optical instruments. Their traditional business had been the ophthalmic business, which was lenses and frames, but also ophthalmic instruments, microscopes. They also had a theodolite. So they were looking to expand the scope of their business, and they felt they should get a broader, more lucrative — well, not necessarily more lucrative, but a bigger scope of business because they had money to invest, basically. And they thought that the way to do this was to bring in an accomplished and well-known optical scientist, and that was Bryan O'Brien. After Bryan O'Brien was hired, he didn't last there very long because the top management of the company had changed. There's always a very close relationship between the top management of the company and the Director of Research. He was hired by one president of the company, but that president was changed; it became someone else. And that other person was not that comfortable for one reason or another I wasn't there at the time. They just didn't hit it off to the extent that one would expect for the relationship between the Director of Research and the chief executive officer. Steve MacNeille was on board at the time, and a man named Weldon Schumacher was president then. And so Steve under Schumacher was made the Director of Research. And Steve had been in that position for perhaps three or four years, something of that sort. He had a research program going which included a variety of things which I would say by and large were kind of typical of what you'd expect a research of a company like that to be involved with. They were a company that was in transition in the sense that they had an old established product line relating to ophthalmic instruments; microscopes, ophthalmic products in general, and then a desire to get into the new things, which tended to be more tied to what you might call electro optics. They had a program there with some of the optics on the Sidewinder missile, for example. These tended to be more involved with the military hardware, and had a fair amount of electronics as compared to the previous types of instruments that they worked on. There was more electronic involvement and more of this being of the nature of hardware to be supplied to DOD.

Bromberg:

MacNeille had what kind of background? Was he a physicist?

Snitzer:

MacNeille was a physicist. He graduated in the middle of 1936-37 from MIT, and worked at Eastman Kodak, first as a researcher on his own, and then later he had some administrative responsibility, I believe. He is generally regarded as a person who is very level headed, he was a clear-thinking guy, didn't stampede easily, and was well liked. He was respected generally by anyone who had contact with him. These are the attributes which led Bryan O'Brien, who was at the time a world-renowned optics man with a lot of influence in Washington because of the work he had undertaken during World War II on various optical devices that particularly the Air Force was interested in. So, Bryan O'Brien, because of this falling out with the head of the company, who then was Schumacher, I guess his original contract called for him being a consultant if he were not to remain as the Director. I don't know the details of it; it may have just been an automatic sort of being bumped into that status of consultant if you were not Director of Research. And Steve MacNeille became Director of Research. And then Steve had a very good relationship with Schumacher; certainly a better relationship.

Bromberg:

So when you're coming in, and this is quite late, this is '59(they're still actively in an expansion phase?

Snitzer:

Yes, because just before then, Bryan O'Brien had been brought in. The major activity he had undertaken was to start and essentially complete the TODD-A-O program. The TODD stands for Todd, you know, the former husband of Elizabeth Taylor. Was it Michael Todd? The wide-screen panoramic Sound of Music was done with the TODD-A-O procedure. There were some concepts that this guy Todd, who was then the husband of Elizabeth Taylor, had about movies how should be made. He got together somehow or other with American Optical — I don't know who sold who, but I assume Todd sold American Optical, so AO started this program with the idea that they were going to revolutionize the movie industry by having wide-screen, wraparound, and various other features that would be part of the TODD-A-O. I guess they did South Pacific, not Sound of Music. Yes, South Pacific was the first one. And it was called TODD-A-O. A lot of money was spent on that, and it probably would have been a successful financial endeavor if it weren't for the fact that television was coming on so strong, and that television really sopped up most of the popular interest in entertainment in that period. So they were in this expansion mode, and there was a feeling of "Well, that didn't work out as it was hoped. Perhaps we ought to get back a little closer to the things we know how to do." And in that context, there may have been some pulling and shoving for different projects to be selected. I believe that Steve MacNeille's personal preference was to go after the more demanding areas of electro-optics with the technology that AO had. You see, AO was not particularly strong in electronics; however, it's unusual capabilities were that it had very good optics, both in grinding and polishing and lens design capabilities through consultants that could do this. So they could design very complicated systems. In addition to that, they had their own materials capability; they made their own glasses, and not many companies did that. And I suppose another significant factor was there was a tradition of making a commitment in an area and working at it for a reasonable length of time. So it was not like what many companies do today: as long as they've got a contract, they'll work on it, and as soon as the contract goes away, they stop. That was not the case at AO. There was an understanding of how long it took to develop a product and, if there was a conviction that the product was a sensible product to pursue, a willingness to spend whatever money and time was necessary to get that far.

Bromberg:

So it sounds to me(is this correct?) that both of these were part of what attracted you to come.

Snitzer:

Not quite. I have to say something about my own personal history of involvement. I was teaching at Lowell Tech. I taught in the period from 1956-1958, and in the spring of 1958, I was subpoenaed to appear before the House Un-American Activities Committee as a consequence of the fact that I had been very heavily involved in left-wing politics as a student at the University of Chicago. I was called before the House Un-American Activities Committee when there was the last big sweep on the part of that committee. And they called up everybody they could. My understanding of it was that what they were attempting to do was to save an investigating committee in the state that was about to be de facto eliminated by having its funds cut off. And the net result was that they subpoenaed everybody they could to make a big show in Boston. And I refused to testify on the grounds of the First Amendment concerning my political beliefs and associations. There was a little bit of bombast in what I did. If I had to do it again, I probably would have done it a little differently.

Bromberg:

That was a very courageous ground.

Snitzer:

But essentially I would have done the same thing. Anyway, the net result of that was, I was teaching at Lowell Tech and I was suspended first at Lowell Tech at the time that I was subpoenaed, and then subsequently I was in a hearing before the Board of Trustees of Lowell Tech. I lost my job. I was canned. That case was taken up by the AAUP and it was finally settled some years later. The [Lowell Tech] Administration had to change their procedures so that whatever had transpired that was considered a transgression on academic freedom would not take place again, and they had made proper procedural steps and commitments to take care of that. And then they also had to have some kind of restitution to the aggrieved parties, and in this case it was myself and another fellow named Dave Fine. Officially it's to allow the person to have his job back. But neither Dave or I were interested in going back. At least I wasn't. I can't speak for Dave; I don't remember Dave's attitude. I don't think he was interested in going back. But I had no interest in going back there because things were going very well at American Optical at the time all of this was finally cleared up. There was a cash settlement. Not very much, a few thousand bucks, that was all. Anyway, that's how it went. So I was unemployed! And I was interviewing in various places. I must say at that time that I got a lot of help from friends. I got calls from all kinds of people who told me the job possibilities that were available. I got a lot of consulting work; I consulted at a number of places.

Bromberg:

What were you consulting on? What kinds of problems?

Snitzer:

Well, I had a little job at High Voltage Engineering, where I did metal filing of matching networks that go on microwave plumbing. And then I did a little theoretical work for them where they wanted to beef up a proposal on the Stanford Linear Accelerator. They wanted a calculation done in the case where you have just barely relativistic electrons and what happens to the space charge distribution. In short, whatever I could do. I did a lot of tutoring. The word was sort of out by then to a lot of people. I must say that year, before things settled down where I actually got a job at American Optics, was a very chaotic period, but there were an awful lot of people who just kept bringing work to me, so to speak. And my income actually went up slightly, not that I made that much at Lowell Tech. But it was just the combination of various things. But it was a very aggravating time; it was a very difficult time because my wife was pregnant with our fifth child. So anyway, when the job at AO came along I didn't hesitate about taking it. I felt pretty comfortable about taking it.

Bromberg:

Am I right in saying that this was an area that was pretty new to you?

Snitzer:

Well, I had been to a talk where I had heard about masers. And I was very interested in it, because that was the new device that everybody was talking about as a low-noise detector, it was actually a low-noise amplifier in a detection system. There was some kind of excitement associated with it because it was a new device involving a fair amount of interesting new physics that people had begun to recognize had some commercial relevance as well as having significant scientific content.

Bromberg:

Now I am puzzled. Did they actually hire you to work on masers and lasers here?

Snitzer:

No, they didn't. I just knew about it.

Bromberg:

Okay.

Snitzer:

And before I was hired, I had an interview. It was probably more than one occasion, I remember a couple of events that had taken place. One was, I met Steve MacNeille. I did discuss with him the fact that I had appeared before the House Un-American Activities Committee and the consequence of that. And I did it on the context that, from what they had described of what they done, I felt that they should know that I would not be able to get government security clearance. So to the extent that they had a position in their organization, which did not require clearance, then I could be considered. That was the general policy I had if I were interviewed by someone who seemed to be serious about me, and who seemed to be a decent guy in the way in which they treated me and so that there would not be any unrealistic expectations about what part of the organization's activities I would be involved with. That's as much as I would tell them. Usually, in fact in all cases, it was enough. I remember Steve not feeling that there was any particular problem. I think part of that was because of his own sentiments. It sounds kind of corny, but I always considered him to be a very noble guy, and I think he was very distressed by all the government security clearance stuff that had transpired and a lot of the injustices that had taken place as a result of it. He just was a very, very decent guy, and I think he felt genuinely distressed about the fact that people who might have a contribution to make in technology were excluded from doing so because of government security requirements being inappropriately applied. It wasn't that he was opposed to security requirements; he just felt it was not done fairly. So apparently he saw the basis for proceeding with American Optical making an offer to me. That was one event that I recall. The other event was that they were looking to do more in fiber optics, because they had just ramped up the fiber optics work. The leading guy in the fiber optics work, a fellow named Will Hicks, who incidentally is here now at Polaroid, had recently left to set up his own business, and they were sort of regrouping and kind of redirecting what they should be doing in this area because they had built a substantial patent position in fiber optics at that time. I was at lunch with a number of people, including Steve MacNeille, and I was shown a picture of the end of a fiber bundle where you can clearly see wave-guide modes. I have that picture. It's just my luck that I won't be able to find it. (Rummages through belongings.)

Bromberg:

This is a picture that they showed you?

Snitzer:

They showed me a picture of that and they said, "Do you know what this is?" And I looked at it, and I said, "Well, this is interesting. It looks like the end of a fiber bundle. How did you take it?" And they explained how they took it. And I said, "Oh my God, those are wave-guide modes."

Bromberg:

Did they recognize that those were wave-guide modes at that time?

Snitzer:

I think they probably did. But it was a recognition that was perhaps not immediately obvious to them or they thought it was perhaps some kind of a special insight or that it reflected some understanding of wave-guide mode propagation. Anyway, I do recall at that luncheon when we were sitting around that restaurant, I remember the picture was taken out and shown to me, and I said, "That's wave-guide modes." Now at that time I had been working at one of many part-time jobs, this one at High Voltage Engineering, where I was working basically on wave-guide mode propagation in linear accelerator-type applications. And so I said these are wave-guide modes in a dielectric wave-guide, and then we talked about it. Later on I was told that my identifying these as wave-guide modes was one of the things that made them favorably disposed towards making an offer to me. I don't know the exact chronology, but basically it was the following: there are a number of government contracts that American Optical had in-house that required clearance. I think most of these were through the Navy. Steve MacNeille he had informed the Navy, or whoever else it was, that they wanted to hire me, and the word that came back from the Navy was yes, you could hire him, but he could not work on classified work, which was standard. And then Steve said "Well give it to us in writing," and they did. I don't know whether they had to be pressed to do so, but they did. The significance of that is that there was a legal structure, all of it said yes a person could work in a place. If you didn't have clearance, you couldn't work on classified work. But in fact what happened was, if there was some classified work being carried on in the area, that person would not get a job anywhere in that area. One reason why I say this is because after I had left Lowell Tech, I then got a job at MIT. I was told I was hired there; this is in the Computer Components and Systems Group. At the time a security officer at Lincoln Lab was on vacation. This is a real wild story, it's not relevant to lasers at all, but anyway, I'll tell it.

Bromberg:

One wonders whether that means that the population of scientists in lasers would have been skewed against people who couldn't get clearance. It may have distorted in some way.

Snitzer:

I think that probably was the case. Well, what happened in the case of MIT was, here's this Computer Components and Systems Group, and they had done some very nice things. Dudley Buck was one of the people there; he had invented the cryogenic memory element, which I don't think is an important device today, but at the time it was viewed as a very important potential device, and was being pursued actively from a research point of view. And then the Security Officer from Lincoln Lab came back from vacation, and apparently he said I had to go. I discussed it with a person who was my boss at the time, who I must say, was a very nice guy. I suppose he felt he was not in a position of being able to influence the course of events very much. He was more a guy who viewed himself as someone in the middle, and didn't have much latitude but to respond to whatever pressures that were brought to bear on him. At least that's the impression which he gave me, and I believe that that's the case. People always have choices, could do other things, but in fact it was not something of his making, nor was he prepared to stick his neck out in an extreme way like Steve MacNeille, who stuck his neck out in an extreme way, in a sense that he pressed the Navy to give him in writing that there would be no objection to my being hired provided I didn't work on classified stuff. I remember at the time then going to visit various people. I went to see Victor Weisskopf. Weisskopf apparently raised a lot of hell. He was very willing to shake up the system for me to keep the job. There were some other people who were not as good. I won't go into that, but Weisskopf was really very good. And then I had some potential jobs. Just at the time I started at AO, I had a potential job at Harvard. I actually had two places. There was a possibility of a job with Gold in astronomy. And then there was another one with Livingston on the electron accelerator — it's not a linear accelerator, but the accelerator program that's right near the applied physics, Pierce Hall.

Bromberg:

I don't remember him being there so late, but yes. That white building.

Snitzer:

It's sort of set back near where the museum is for glass flowers. Anyway, that required consideration by the Harvard Corporation. And then I saw McGeorge Bundy on a couple of occasions. McGeorge Bundy, I must say, was very willing to hire me. He felt that there was no problem. He struck me as a very savvy guy who understood the political situation he was dealing with. He was not dealing with a potential spy or anything of that sort, and he made that very clear right off the bat. However, right at the same time, they had had some incident of somebody who was in either Slavic Studies or Russian Studies or some social or political science activity who had been identified by the FBI as someone who was engaged in some sort of espionage. So there was a dismissal and a lot of publicity about it. And that's about the same time that my name came up, and of course he had the obligation to relate to the Harvard Corporation people who were making the decision that I had "a political past." And they turned it down. And he apparently fought very hard to get me appointed. And then, a curious thing happened with him afterwards. He felt very badly about it, and he wanted to give me a hand. So he said to me at one point during a telephone conversation: "You know, to help you out, I'll write you a letter and say you got a job at Harvard. Provided you don't take it. And you can use it maybe to get a job elsewhere." I should have said yes to that, but by that time I had already firmed up the job at AO and I said, "Well, I don't think I need it." It would have been a delightful thing to have had. He's a decent guy. And a guy who appreciated my situation, and certainly a person who was not swept up with any hysteria about the Russians infiltrating all of our institutions, etc. Anyway, I'm drifting off. I was hired by AO.

Bromberg:

And you were hired specifically to work on fiber optics at this point?

Snitzer:

To work in general. The way Steve did things was very much to his credit. I've tended in my work, to the extent that I'm involved with administrative work, to sort of emulate it. He would say, "Well, these are the things that they have identified as important to work on. See if any of these things strike your fancy and if you feel you can do some work with them, we'll begin to shape up a program appropriately with additional people." He made the selection, but it was not an autocratic management. He made the selection of the program the result of, uh...

Bromberg:

A negotiation.

Snitzer:

Right. You negotiated the thing with him. Negotiation is perhaps a little more formal description than what actually took place.

Bromberg:

Do you have any recollection of what the whole range of problems were as you came in?

Snitzer:

What I was told was AO had a done of work in fiber optics, and was probably the leading company in the country in that area, but they did not have a sufficient number of papers published, particularly careful measurements and, most important of all, theoretical descriptions of propagation of light in fibers, so they would like to beef that up so as to establish fully their position in fiber optics and have some publications to the organization's credit.

Bromberg:

I see. So these publications are really part of the image of the company as much as anything else.

Snitzer:

That's right. They saw that as important. Remember, the fiber optics program was mostly company-sponsored. They had some government contracts, to be sure, but the company had put in a lot of money and they had paid a lot of attention to establishing a good patent package, and that patent package really stood, so they could license [others]. Tape one, side two

Snitzer:

There was a possibility of getting started with a program that was involved with a high-resolution cathode-ray tube, that is the way in which you make a high-resolution, settled screen, phosphor. That was something that reflected this thing that I mentioned earlier, that there was interest in getting more involved with electronics and electro-optics, types of things. I had actually started something along those lines. I had some apparatus built, but I never used it, because in the meantime the question of what was called an optical maser came up, it wasn't called a laser at that time, it was called an optical maser, and I discussed this with Steve about the possibility of starting a program of that type and he was all in favor of it. He said, "Yeah, go ahead. Go with it." In the meantime, I had made some contact with Hans Zucker who was here at RADC — I don't know if it was called RADC at the time Hanscom Field around the wave-guide mode propagation studies.

Bromberg:

I'd like to really find out about that. Also, it somehow sounds as if your work on the transmission of wave-guide-type modes in these fibers is all involved here, that you're doing this at the same time you're doing the maser work. Or did the optical maser come in first?

Snitzer:

No, actually the wave-guide mode stuff started first, because I was shown this picture before I was hired. After I got on board, of the various things that were presented to me, I could get started with this most readily. Part of the reason for that was everything was there to make the fibers and to make the studies on this, and the equipment was there for observations, etc. Harold Osterberg was there, and Osterberg was a superbly competent optics man who could set up almost anything from scratch for making observations, so he was just immensely helpful. I made the fibers myself. Before then, I wanted to get some fibers made for me, because I didn't have the familiarity with making them, and that was at a time when there was a lot of pressure to get some things out on some contracts on the part of the people who were making fibers on a regular basis, and I had special-size fibers to be made, which was a little bit of a disruption to what the other people were doing. Plus probably just the normal degree of reticence about getting involved with a program with somebody who's new on board, etc. But it turned out about that time, I had met Will Hicks who had just recently left AO and set up his own business, Mosaic Fabrications, and who was very clever in making fibers, and could make these things very easily. I met him socially through the Walkers. Gordon Walker subsequently left the job at AO and became the head of the American Mathematical Society, which operates out of the Brown campus in Providence, RI. He's since retired. I'm not sure; he may not even be alive. So anyway, I met him through them; actually it was his wife Jacqueline who was the person who was the major source of introduction to Will(and I got to see Will quite a bit, and then after we moved up, Will and his family and my family socialized together, because the wives became good friends, and then it turned out that his children and our children were about the same age. He had worked at AO, so he knew a lot of the people that I knew. I had mentioned to him on one occasion to complain about the fact that I wasn't able to get the fibers that I needed, not because of any reluctance on the part of the organization to supply me with them, but just the normal course of events where people don't get to know one another that quickly. So he said, "Well, come over and we'll make the fibers that you need." There were a bunch of special sizes and various shapes.

Bromberg:

You were already trying for smaller-than-normal fibers at this point?

Snitzer:

It was a smaller core, but I wanted a range of fibers and then I wanted to try twin-core fibers for cross-talk studies. Because you see, this enticed me that there had to be cross-talk taking place here.

Bromberg:

I see, from looking at this picture of an end of a fiber.

Snitzer:

That represented cross-talk which was a particular interesting question because there was the kind of prevailing sentiment at the time, I think both by Narinder Kapany and by Will(was that cross-talk would not take place. The idea of cross-talk is that if you took two cores, light will go from the first to the second and back to the first. The idea that if you had just two, the light would go back to the first from the second, was not that well established. There was some skepticism about it, because of a feeling that you had many other modes to which coupling could take place, so that eventually what would happen is the light would not all go back to the first. At least there was some skepticism about whether that was the case, and at the time I said, "No, if there are only two cores, the light should all go back to the first." That was the subject of a paper, which was given up in Ottawa, I think. That was probably in the fall of '59.

Bromberg:

Yes, there was this Ottawa paper which seemed to me (???)

Snitzer:

On cross talk?

Bromberg:

Well, it was on optical wave-guides and I didn't go into the paper sufficiently to be able to tell you whether cross talk was one of the important things. In most cases I just skim these papers.

Snitzer:

It was either that or American Optical entertained the New England section of the Optical Society at Southbridge at AO. Then I know cross talk was one of the topics presented.

Bromberg:

This is the bibliography [of Snitzer's publications] in case we need to look at it. So anyway, cross talk was one of the initial things, and Hicks then made you these fibers.

Snitzer:

I went by Mosaic Fabrications to see Will, and he made the special fibers for this, and he was able to make it actually quite quickly, because he's very skilled at doing things, particularly doing clever things. So, I then used the fibers as part of the study. And then afterwards, I felt it was an imposition on Will to be getting fibers from him. He was also ramping up his business, so that it was more difficult to get things made from him. I built this small set-up at AO just for making experimental fibers to be used in wave-guide mode studies. And most of the subsequent wave-guide mode stuff was done that way.

Bromberg:

When you were giving these papers on fibers as wave-guides, how surprising or not surprising was it? I mean, did everybody know for sure that these were wave-guide modes, or was this a big surprise to them? How did people react in meetings, for example?

Snitzer:

I would say that everybody said, "Yeah, of course you'd have to have wave-guides." There were some striking differences, though, in the characteristics of dielectric wave-guides as against metallic wave-guides. The people in the microwave region, back during World War II had done a lot of work on metallic wave-guides and on dielectric guides for the so-called end-fire antennas, because they were part of antenna systems. An awful lot of work had been done on dielectric wave-guides, mostly on the lowest order mode. They were called polyrod antennas, ferrod antennas. But the higher-order modes, how the modes were described, and in particular, the consequence of a low difference in index of refraction between core and cladding had not been investigated, because those were not particularly relevant issues for dielectric wave-guides in use then. I remember one reaction by a fellow named Strandberg here at MIT. His reaction was, "Oh, my God! We're going to have to get into even shorter wavelengths. It's going to involve all kinds of complications!" Because he had just gone through the phase of having to go from centimeter waves to millimeter waves. He considered that to be a pain to scale down. And here we were scaling down by another couple of orders of magnitude. But it was in a jocular vein that he stated this.

Bromberg:

Is this a situation where you were well-read in these dielectric wave-guide studies in the millimeter range?

Snitzer:

I read what I thought was all the important stuff, I was aware of the important stuff.

Bromberg:

You read it as you came into AO and started this work?

Snitzer:

Yes, that's right. No, I didn't know anything about it before.

Bromberg:

Okay.

Snitzer:

And then along the way what happened was: well, what can you use this for? And I remember that at this meeting that was held at American Optical where AO was the host for the local Optical Society, I presented this talk on wave-guide modes in fibers, and afterwards discussed with various people what was going on. And then I expressed the opinion that this could be a useful host for an optical maser, because you've got a wave-guide structure already. Work had already begun on attempting to make a potassium vapor [laser] at Columbia, and there was kind of a general feeling that, well, if you get a coherent system, you're going to have wave-guides, because you want to keep coherence. wave-guides in general were considered as part of this, although people did not see the precise connection because a lot of things had to fall into place. It was viewed as part of the whole area of coherent optical systems and that it would eventually play a role somehow. Now I remember there was a conference held in Berkeley where I gave a paper.

Bromberg:

That's already '61.

Snitzer:

Okay, that was later.

Bromberg:

That was the one that Singer organized, the second.

Snitzer:

That's right.

Bromberg:

You didn't go to the first Shawanga Lodge conference, did you?

Snitzer:

No, I was not involved in any of that stuff at the time.

Bromberg:

So, at this point, you're thinking about these optical wave-guides and you're thinking about masers.

Snitzer:

It seemed as if it was the right kind of cavity to make, and you take a long length of fiber and you terminate the ends with reflectors, which is easy to do, and then you just pump the center. And we made attempts to do that, and we thought also by having the fiber it would be an optically thin material and so we should get the maximum pumping per unit volume which you could get from a flash-lamp or light source. We had started doing this before the announcement by Maiman of optical maser action in ruby. However, we were not using a flash-lamp; we thought that we could use one of these high-pressure mercury lamps. At that time it was well-known how to describe theoretically when you expect laser action to occur because you get losses involved in the system, then you get gain, and if you talk about the equivalent temperature of the light source, that in turn allows you to determine whether you should lase or not. And the calculations indicated that laser action ought to occur, but it neglected one major consideration, and that major consideration was the business of re-absorption from the excited state to a still higher level, and we couldn't evaluate that at all.

Bromberg:

What kinds of glass were you working with at this point? Are we still in late '59 or early '60? Maiman's announcement came out in July of '60.

Snitzer:

That's right. Well, we had started before then and we had started with the visible rare earth fluorescers. These were based on line width, because the great concern was the question of line width. I gave a colloquium talk at Columbia at the invitation of Townes and Schawlow and that was sometime early in that spring.

Bromberg:

Were you working alone at this point or were you part of a team?

Snitzer:

I'm not sure exactly when the people came on, but first there was Dave LaMarre, who was interested in pursuing lasers or optical masers, and the possibility of getting a Ph.D. thesis in it at the University of Connecticut. He had done all the work prior to that. And then there was also a young fellow who was starting to work on a Master's degree and did some work on wave-guide modes. I've forgotten his name, but I have him referenced in one of my articles. And then finally, later on, Saul Bergmann came on board. Saul had done some work on microwaves. I noticed there was a reference to Saul Bergmann here in one of the papers [with questions].

Bromberg:

Yes, I just wondered.

Snitzer:

Saul came on as an additional guy in the wave-guide mode stuff. He did not bring the idea of using the dielectric wave-guide as the structure. He was hired as an additional person. There was a certain psychology that existed that time, and that was that technical people were very scarce and it was very difficult to hire technical people. That was the aftermath of the Russian sputnik, and technical people were hired very readily, and particularly out in Southbridge, which was sort of like out in the boonies, it wasn't that easy to get people to come out.

Bromberg:

So LaMarre and Bergmann are really sort of technicians?

Snitzer:

No, not technicians. They were...

Bromberg:

Junior scientists?

Snitzer:

No, even more than that. The group was such that the three of us were working together as collaborators. Because I had sort of founded the group and had the interaction with management about supporting the group, I was a little more influential than the other two, but we basically acted together. I was officially the boss, but de facto it was pretty equal in terms of responsibilities that we had on a technical level.

Bromberg:

I guess I want to ask another organizational question before we go further. Am I right in saying that trying to get this stuff to lase is one of the number of things you were doing? Because somehow in here is this contract for simulating end-fire antennas.

Snitzer:

I had two activities then. The laser program and the dielectric end-fire antenna simulation, wave-guide mode stuff.

Bromberg:

Which was, I assume, non-classified Air Force-Cambridge research.

Snitzer:

Yes, that's right. It was just basically to elucidate the properties of these things, and make calculations of them. We didn't deliver any hardware. It was a study program. Just like what we would now call 6.1 money.

Bromberg:

Okay. So you were actively working on lasers and since you gave a colloquium at Columbia, the whole community was well aware of this work.

Snitzer:

That's right. And furthermore the feeling was that the wave-guide mode stuff would somehow or other be relevant to lasers, maybe as the lasing structure, maybe as the means of transmitting the light from one place to another where you could keep track of phase. It wasn't clearly understood where it would connect, but there was a conviction that it was part of the whole ball of wax that was associated with this new field of coherent optics.

Bromberg:

And you had already been investigating the use of this fiber as a resonant cavity and its advantages over the Fabry-Perot cavity as you saw them, and so on.

Snitzer:

Right. And then things went very slowly, because we first started out looking at the visible fluorescers. We decided first of all if you go through the calculations you have a certain radiative lifetime. From the radiative lifetime, you can then infer what the oscillator strength is. From the oscillator strength and the line width you can infer what the gain coefficient is, which in turn tells you how much population you need in an excited state to get it to lase. So what that all amounts to is, you want as narrow a line as possible in order for laser action to occur. And there had been some work done at Bell, because Bell had previously gotten some laser action going, when they did this stuff in ruby after Maiman. They concluded that out of various materials, glass would probably not work because the line widths were too broad. But if you look carefully, the line widths are not that broad, and the lifetimes, etc., and I think that was a rash conclusion on their part. It was probably meant in the spirit of it not being the most hopeful candidate. However, those things sometimes get overstated. Anyway, there was an overstatement, and I've forgotten who at Bell was responsible.

Bromberg:

Of course there was a lot of competitive pressure to get the most hopeful candidate, I would guess.

Snitzer:

That's right. Everybody was scrambling and making exaggerated claims, I'd say, for their own systems. There was a tendency to say "Glass is not it; it's not going to work." And some very prestigious guys said this.

Bromberg:

I was just wondering if this general sense of pressure in any way was communicated here at AO.

Snitzer:

Well, it got communicated to my boss, Steve MacNeille, and I remember Steve asking about it and I said, "Yeah, he's right. If you had something that had a line width that was a tenth of what we can get in glass now, it would be a lot better. However, if you just go through the numbers, it still looks very good. It still looks like the system ought to work. I don't understand why it doesn't." And Steve's attitude was, "Yeah, it looks good. Keep working on it." He was just a very understanding guy. And I think it also reflects his experience at Eastman Kodak, which was another company that was a successful company built with its own funds, so it was used to making commitments over long periods of time and getting results, and ended up being a leader in its field as a result. Whereas if I had been working at a more traditional, what we now call aerospace, company, there's a good chance that that would not have happened.

Bromberg:

It's kind of an interesting contrast with Hughes, because of course Maiman was also working on an unpopular material. But I get the impression that he was under a lot of pressure from his immediate people.

Snitzer:

I would think so. Because the prevailing sentiment was that ruby was not going to lase. And of course the red ruby didn't, but the pink ruby did. So anyway, I first looked at the visible fluorescers and we spent a long time on it, and the visible fluorescers were europium, terbium, samarium, and dysprosium. And in order to maximize the probability of our success, we worked at that time initially with small fibers thinking they would be optically thin.

Bromberg:

How big do you mean when you say "small"?

Snitzer:

Well, there were probably a few mils diameter for the core. As fibers go, those are big fibers, but they're not rods. And then maybe five or six mils for the cladding. And then we wound around tight helices and put them near lamps and tried lasing them.

Bromberg:

Is there anything, by the way, that we ought to record about this activity? Any experimental methods that you had to invent?

Snitzer:

No, I don't think so. Well, I guess we were the first ones to use clad configurations in solids. That was a big help, because if you clad something, you get the equivalent of an immersion effect and you concentrate the light. (Draws diagram.) And so we very quickly realized that if you have a core of glass and there's a pump light that you put on the outside of this here, the light that would go in grazing incidence gets refracted in. So the net result is, whatever light was incident across here ended up being focused into this smaller region. So that was an immersion effect which concentrated the light. And the immersion that you could get from this was in the ratio of the index of refraction of the material to air. So it increased it by 50%.

Bromberg:

The index of reflex ion of the cladding to the air?

Snitzer:

Yes, because this index difference [between core and cladding] is relatively small. It's just n1 sin theta1 = n2 sin theta2. You come in at grazing incidence so that theta1 is 90, so sin theta1 is 1. But anyway, you can see this is an immersion effect.

Bromberg:

So when you say you started this with clad solids, you mean something much bigger than a few mils?

Snitzer:

Even with big rods, if you want to get a lot of focusing of light in a cylindrical configuration.

Bromberg:

Now that's a rather new thing. I mean, nobody that I know of was doing this kind of thing with lasers at all.

Snitzer:

Virtually all of the rods from that point on, not all of them, but a large percentage of them would be clad rods, because you had various things. You had an immersion effect, which you could get from this. You could make this out of samarium, which is something that came later. The samarium was transparent to the pumping light, but was absorbing to 106mm light, so that any spontaneous emission going out the side, instead of going out the side and rattling around and coming on through and robbing from your inversion, would be absorbed out and taken out of the system.

Bromberg:

You say the samarium came later. In other words, these claddings were not samarium at first.

Snitzer:

Initially, they were just clear glass for an immersion effect cause the problem that was perceived at the time was how to get a bright enough pumping source.

Bromberg:

And one of the things that is going to happen later is that you begin to understand that the cladding should also be such that it should absorb the lasing line. Do I have that right?

Snitzer:

Yes, that's right.

Bromberg:

Okay, so we have you here in the lab and you're working with big rods that are clad and then you're working with rods a few mils that are clad and then, to just return to that question a little bit, this is the only group, as far as I know, that was working with clad rods. Is that true?

Snitzer:

That's right. We were the only ones that did any significant amount of work. I'm sure others did some cladding later. At that very early stage, no, because everything was crystals, and you didn't clad crystals in the early days.

Bromberg:

One of the things I wanted to do is to try to get the sequence of novelties as they get developed.

Snitzer:

Okay. So what happened then was we worked very hard on the visible fluorescers. How do you establish when something is lasing? There were several ways: you could look at the output suddenly going up and you get a spiking type of behavior. No, the output increase, but not a spiking behavior at that time, because we didn't know about the spiking behavior in the early days.

Bromberg:

Of course the ruby did spike.

Snitzer:

We started to work on that stuff, but really didn't crank up the program in a vigorous way until there was that Coherence Conference at the University of Rochester, at which I gave the paper on wave-guide modes, and the motivation for that was that wave-guides somehow or other would be a part of this whole business. And there was a front-page newspaper article written by Malcolm Stitch on wave-guide modes and he was enthused at the time and I think he probably had wind of the fact that Maiman had lased ruby.

Bromberg:

Maiman did it actually in May, so he knew it.

Snitzer:

And that took place later. So one or two weeks after that Coherence Conference, there was the announcement of ruby.

Bromberg:

What was the Coherence Conference like? What was the ambience? I mean, one can read the proceedings, but (???).

Snitzer:

A lot of excitement. The feeling was a lot of new things were happening here, a lot of stuff that was sort of unconnected, perhaps, was beginning to come together. There was a lot of interest expressed in my paper, not because people could see precisely where it would fit in, but just in general that it fit. And I think the attitude was that there's no reason to believe this is any different from optical masers or masers in general or microwaves. It's an extension of microwaves to shorter wavelengths. A different technique was required; the starting point of the technique was induced emission. There was a lot of concern about cavities: what constitutes a cavity, how do you make a cavity? The Fabry-Perot cavity was one that was considered. At the time, I remember I had an argument. Often when you take a cavity you can make the argument that the cavity is an idealization. You have to get in and out of a cavity. You idealize the cavity from the point of view of the field distribution based on some other structure like a wave-guide. For example, [draws diagram] if you had a cavity that was in the form of a cylinder with a little hole in the center and then the little hole in the cylinder allows you to couple light in, but you manage to excite a mode which had a field distribution which was like that in intensity, in the z(direction, or any of the field components which have a large amplitude in here. What you normally say is, "Well, seal up the hole, and look at what the field distribution is inside." And then you could say that the field distribution in that cavity is basically like a particular mode propagating with a certain number of half-wavelengths between the reflectors. Tape two, side one [The point of this discussion is that the Fabry-Perot modes are the same as the dielectric wave-guide modes far from cut-off with end reflectors, JB.]

Snitzer:

(for the dielectric wave-guide, and then took it far from cutoff. Then that's the same field distribution as you get in a Fabry-Perot interferometer; it turns out to be essentially the same problem. So anyway, there was interest in those things, and there were a lot of discussions of that type of stuff. When we came back from that coherence conference, or a week or two later, when it was announced that Maiman had lased ruby, then we got flash lamps right away. We figured our technology was inappropriate. We really should go into flash lamps where you get much brighter. And we immediately ordered some flash lamps and got busy setting up the flash-lamp configuration to do the lasing, whereas before we had been working with high-pressure mercury lamps.

Bromberg:

And you got better results? How did the flash lamps change that?

Snitzer:

Not right away, because we certainly were able to get brighter light sources, but we still did not lase it, because we were dealing with these four visible fluorescers. And we saw no other glasses except for(oh yes, we thought that, well, since ruby was the right stuff, we tried to get a hold of some ruby fiber. It was a ruby spongy material where we could pull out little fibers, and that was hopeless. We tried cladding with glass and we couldn't.

Bromberg:

What was the psychological atmosphere like after Maiman?

Snitzer:

A lot of excitement. It was demonstrated that it could be done. There must be other materials that make it possible to do. It's tough to do, you need a bright light source, and it's pushing the limits of what you have. There was a general feeling that it's hard to do, because it requires this brightness, but it can be done. And for that reason you just get out there and work at it, and other systems will follow along. Then, while we were working on that, Dave LaMarre and Saul Bergmann were in the group, and then I remember Dave and I got a little discouraged because we weren't getting any results. We were working hard. We were typically coming in on weekends and on evenings, because the feeling was that Maiman had just announced about ruby and in order not to be completely scooped or preempted in our own efforts, we should proceed as quickly as we can and try to make a glass lase. And we just were unsuccessful.

Bromberg:

Were other people getting into the glass lasing business at this point or not yet?

Snitzer:

Not that we were aware of then. I understand that later on people had contracted to get some glasses and to start looking at glasses. But at that time, no, and in fact, it was quite the contrary. What had happened was that there was one guy at Bell who reproduced Maiman's stuff and reported on the spiking behavior of the output.

Bromberg:

Well, there's a paper by Collins and Nelson.

Snitzer:

That's right. It was Nelson who basically had done that. So we knew there was spiking. And the reason why I mentioned that is because if you look for various evidence for lasing action, you look for kind of a spiking behavior for the fluorescent output. You look for a substantial increase in brightness, which is sort of related but not completely so, and then finally it was felt that the most convincing evidence was to look for a line narrowing. So we decided, well, we don't know how much light we're getting from these little fibers, and we have to go through a spectrometer, etc., so we would look for line narrowing. So we got set up and we were just flashing and exposing plates, and every once in awhile we'd see something that looked like a narrowed line and get all excited, but it wasn't anything. It was an artifact. We made a lot of glasses and tested a lot of glasses. By this time we had gone through around 200 glasses.

Bromberg:

All of these are coming from your lab or you're making them yourself?

Snitzer:

No, they were coming from the lab at AO because AO had a glass fabrication facility. And these were one-pound melts, typically. And then we had to prepare them and make fibers or make rods that were appropriate for lasing. So Dave LaMarre at that time decided that things were not going all that well. There was in the meantime a ramping up of activity to pursue ruby. We had managed to get a contract on a sun-powered ruby laser, which involved an end-fire pumping configuration. That looked like the hottest activity in the place and (???).

Bromberg:

That was also your contract?

Snitzer:

No, it wasn't my contract. It was AO, but I did not figure prominently in that. I think I had made some contribution to the calculations on the original proposal, because whenever proposals came in house at that time, there were so many new things that everybody who had any contribution to make would all get together and have these meetings and thrash out what the approach ought to be and then assignments were made for people to do either writing or managing things or whatever. So I was involved in it but did not have a major administrative responsibility. I was involved technically in participating in the technical decision.

Bromberg:

Because I noticed you have a patent. One of your patents concerns a sun-powered laser system.

Snitzer:

It's sort of typical of most of this stuff. There were a number of people who were technical contributors. We tried to formulate what the approach ought to be to pursue a given contract. And by that time, we were pretty strongly oriented towards picking up any contracts we could in lasers. It was a separate operation that was established that had clearance, because the building I was in was an unclassified facility. It was known to everybody that I didn't have any clearance.

Bromberg:

So you would pick up those DOD contracts because the sun-powered laser would certainly be a DOD contract that would be an unclassified contract.

Snitzer:

That's right. I was involved in it but I did not have administrative responsibility, because there was a fair amount of administrative responsibility, like writing reports and making sure people got things done on time.

Bromberg:

And that was ruby.

Snitzer:

Yes. And Dave LaMarre at the time decided that he wanted to stop working on glasses. I guess he became discouraged at the prospect of getting a glass laser and he saw Steve MacNeille and requested to Steve that he'd like to be transferred over to the group that was working on the sun-powered laser. And they had begun to do some other things that were related to it, but basically it was the sun-powered laser program. It was a disappointment because Dave was one of three people, but that's what he wanted to do, so the intention was to continue on. And then shortly afterwards, there was another disappointing event that took place. I don't know if the two are related or not. Steve MacNeille told me of a conversation that he had had with Saul Bergmann, where Saul Bergmann had gone in and spoken to Steve and told Steve that he felt that Steve should have made Saul aware of the fact that I had government security clearance problems, because that created a problem for him in his future because he was originally from Europe. He basically blackmailed Steve into getting a raise. So Steve saw me and Steve said, "You know, we've got this problem." I said, "What's the problem?" And he told me about it. He wondered if I had any conviction about it. I was teed off about it but I didn't know what to do. Steve said, "My suggestion is it's a dangerous thing to have him working together with you under the circumstances. And it's also a dangerous thing to have him leave the corporation now. So the best thing to do is just to transfer him over to this other group, and let him work there. Whatever he manages to come up with, fine." And that's what they did at that time. So then I was left alone.

Bromberg:

At about when was this? In the fall of '60 or later than that?

Snitzer:

Probably somewhat later than that. No, fall of '60. That's it. Maybe late fall. So what I had decided was that the visible fluorescers were not going to work at all, so I should forget about it and really start working on the infrared fluorescers. And the infrared fluorescers were neodymium, praseodynium, holmium, erbium, and thulium. There was a Perkin-Elmer single-beam spectrometer where one could change things around and instead of the glow bar, put in a fluorescent material and then just illuminate it with a light source; then a chopper and everything else could be used. It was set up in the old vault, which was used to keep all the records for AO originally. One of the benefits of that was that the room tended to be thermally well-insulated. I started collecting data, just looking at fluorescence, and then I was getting weak signals from things like thulium and holmium, but then neodymium just went clear off the chart. It was just an outstanding amount of fluorescence by comparison, because I had to use wide slits.

Bromberg:

In other words, this was not in the literature, this kind of infrared fluorescence. You had to make your own measurements.

Snitzer:

Well, the interesting thing about this was...I did not look at it carefully at the time, but I should have realized that that was the case, that neodymium, now this was neodymium before neodymium-calcium tungstate was lased. There was a famous article about the rare earth energy level diagram. In the energy level diagram, they show the various energy levels of neodymium, and neodymium is like this. 4I 9/2, 11/2, 13/2. And then the level from which fluorescence occurs is 4F 3/2. And what was very interesting was that these two levels (4I 13/2 and 4I 15/2) were not shown on the diagram. I don't remember whether this level (4I 15/2) was shown or not, but the ground-state level, the 4I 9/2, and the level from which fluorescence originates, the 4F 3/2, was shown, but not the J=13/2 or 11/2. Now anybody who looks at the diagram and is familiar with it, would say immediately "Well, there has to be a level in there," but if one didn't know that neodymium was going to lase or was a lasing candidate, if one was not really careful in thinking about these things or looking at it carefully, the tendency is(at least the tendency I have(to look at the energy-level diagrams which had been put together by somebody else, and I think "Well, those are the levels. That's where they are." And if you look at the levels there, there doesn't seem to be anything there that would suggest it as a promising candidate for a four-level system.

Bromberg:

I see. Now this is the article you did not know, so you hadn't(

Snitzer:

McClure, that's who wrote the article. It showed the energy-level diagram but did not contain the terminal state for where lasing action did finally occur at 11/2. He's a very competent guy, and I'm sure the only reason he didn't put them all in was basically for economy of what could be presented in a nice clear way on the assumption that anybody who was knowledgeable would realize that there are levels in between. But at any rate, the fluorescence looked very good and then I immediately decided to take all the rare earths that showed some fluorescence and I had made up five glasses of various compositions. There was a 2 weight-percent of the oxide, 1 weight-percent, a half, and I remember it went down to a tenth. And made them in the form of a one-millimeter core, about that, and the OD [outer diameter] was maybe three millimeters, something like that.

Bromberg:

How come you chose those measurements?

Snitzer:

Just for convenient size, because I had a little fixture. See [drawing] this is the laser rod and there's a little fixture that fit over the top that could then be held in a tube affair, and this could be inserted down into these helical flash lamps.

Bromberg:

See, one of the reason I asked is that when you were working on the fiber optics stuff, you had very much smaller cores, didn't you?

Snitzer:

I decided by that time, if you look at the absorption that takes place in the neodymium, or any of these that we were considering now, the millimeter-size core is small enough so that it's all optically thin. This isn't anything optically thick. So there's no real loss in pumping light associated with propagating through the material.

Bromberg:

What about the wave-guide properties? Would those be relevant?

Snitzer:

No, because the conclusion that we could reach then is...You see, it would be more favorable to have total internal reflection, so you use wave-guide properties to get it going so you have less loss. But it was not necessary to get to a very small fiber to take advantage of that. So this relieved the requirement of having to have the end reflectors perfectly parallel and all that. It could be a pretty crummy cavity.

Bromberg:

Which relieved the requirement? The total internal reflection relieves the (???).

Snitzer:

Yes, see, if you've got two end reflectors, they have to be very parallel. But if you have inside here a reflecting mechanism on the sidewalls of the core that bounces the light off of here, then that's a pretty good cavity. By that time, we got a little bit sophisticated and realized that putting cladding on and having a lower index with a good interface being core and cladding relieved the requirements of having to have good parallel end reflectors.

Bromberg:

This is what you understood as you were going along.

Snitzer:

That's right. And that was I suppose a small insight that came from that, that clad rods with a lower refractive index, with a good interface between core and cladding, could be made to lase more readily, independent of what the ends were, provided that the ends were not too extremely bad. So at that time, essentially when that work was started of looking at the infrared stuff, I was working alone and Saul Bergmann was off on the other stuff. By that time, they were getting contracts for various things, or seeking contracts to pursue other programs than the sun-powered laser. There was another program that was started. Chuck Koester had been working on a retina L coagulator based on a high-pressure mercury lamp and in discussions with him, I think I may have suggested it, but it was suggested in a "How about so and so" kind of way(we both looked into it and then concluded it made sense, and so we submitted a patent disclosure on the use of the ruby laser for a retina L coagulator. And that became the AO retina L coagulator that we worked on with Dr. Charles Campbell who was at Columbia Presbyterian.

Bromberg:

So here you are and you're working alone on these and getting very good fluorescence from the neodymium. So what happened between that and November, when you put in your letter to Physical Review? I guess the letter went in, in November '61, or was it published in November '61 in Physical Review Letters, when you actually got the lasing?

Snitzer:

I don't remember the dates.

Bromberg:

But it really isn't the date that I want to get at. It's how you got from that bright fluorescence.

Snitzer:

The way that we established that it was lasing, we had made up a bunch of rods, four or five rods, of each of the different compositions, and then I put it in this little holder, and the first time I lased it, I blew off the end reflector. You see, I didn't know what was going on. And then on the other rods, I more carefully examined what was going on, and clearly established it was lasing. That took place at a time there was an Optical Society meeting out in L.A., so I called Steve MacNeille out there right away. He was out there, but I was staying back. I had not intended to go initially to the Optical Society meeting, because I considered it to be very important to crank some stuff out. And then I got a hold of Steve and told him about it. And then Steve asked me to come out right away, which I did, and then told Steve and a few others from AO what had happened, that we had lased the glass. And then shortly afterwards, various meetings were held with different people from ONR, then I was invited down to an IDA meeting, the Institute for Defense Analysis, where they questioned me on this glass laser, what's it all about, what it's behavior was. I remember Bloembergen was one of the people who were on the IDA panel, and he asked me what the connection was between fibers and this glass laser. And I remember my answer, which surprised some people: "Nothing." There was no relation in the sense that you didn't have to make it into a fiber. If you wanted to get a lot of power out of it, you make a big glass laser. On the other hand, if you wanted something that would have special properties that could be provided by fiber, you'd use a fiber. So at that point I would say the business of a glass laser and a fiber were really completed decoupled; they were two separate endeavors. You could, to be sure, make a fiber laser from glass, and it might be the only way you could do it readily, at least at that time, because for anything else you'd try to do, it would be difficult to control the size and to control the interface.

Bromberg:

They gradually became decoupled. I mean, you sort of began by looking at them as all part of the same ball of wax. I'm not sure I understand completely. But I do want to make a pause here on the tape. (PAUSE)

Snitzer:

You don't need a super optically thin element.

Bromberg:

So that's one of the steps that leads to this decoupling. What else happened?

Snitzer:

And then the other thing that came along later on was the realization that here you have a laser rod like any other laser. By that time people had a lot of experience with the ruby laser and then, while we had started to work on neodymium in glass, before the time of the lasing of the neodymium glass, the people at Bell had lased neodymium in calcium tungstate.

Bromberg:

Who was that? I don't remember. It's not an important question.

Snitzer:

Greidenbahn or Geusic. Or was it Bloembergen?

Bromberg:

I don't know.

Snitzer:

That was a significant laser. That was the first neodymium laser system going.

Bromberg:

That didn't affect your own work, did it?

Snitzer:

No, it didn't, except that it gave us more confidence that yes, that's the right thing to work on. We had concluded that neodymium was really the stuff to look at, because neodymium had the other advantage. If you look at, say, [drawing] a four-level system here, you're going to go (pump) to here, non-radiative to here (level 3), then you lase to here (level 4), then non-radiative back to here (level 1). But there's a possibility that from this level 3, you can go to a still higher level and this will give you more absorption to here than you get emission down here. So that's a difficulty. And that's probably the difficulty in the case of the glasses that are in the visible region of the spectrum, because you get a lot more energy levels and they're more densely packed and the oscillator strengths to them are greater, so essentially if you populate an excited state, you're increasing the absorption, not decreasing it by stimulated emission downward. In the case of neodymium, because it was shifted into the infrared, you knew all these energy-level diagrams, and you could look at an energy-level diagram or look at an absorption and you knew that this region up here, you didn't end up in the middle of an absorption band. It was a transparent region of the glass. So the one element of uncertainty that you didn't know anything about was eliminated. So that made neodymium particularly attractive to go after, and then particularly there the strength of the fluorescence was very, very strong.

Bromberg:

So the guys at Bell independently may have looked at some of this.

Snitzer:

They may have independently done the neodymium in calcium tungstate. I think it was first ruby, and then Sorokin and Stevenson did the divalent samarium and trivalent uranium, I believe, in calcium fluoride. And then it was while I was working on neodymium in glass, the Bell stuff on neodymium was coming out.

Bromberg:

How did the company react to the glass laser? Were you only in touch with MacNeille or were you seeing people higher up in the company?

Snitzer:

Well, I had contact with Weldon Schumacher and with another guy. Weldon Schumacher was the president of the company, and the second guy in command was Vic Niss, but I didn't have a great deal of contact with him. And then there was the guy who was the head of the instrument division, Kelly Hannan.

Bromberg:

Were they principally interested in turning this into a contract with DOD or were they principally interested in the possibility of commercial products?

Snitzer:

I think their motivation was commercial products. However, there was a willingness to be patient about how long it would take to get there, and Steve MacNeille fairly consistently cautioned them about trying to proceed too rapidly to try to make a profit out of some of these things. And there's evidence to go along with that. There was one member of the board of directors who had gotten the ear of Weldon Schumacher and had convinced Weldon that if you really want to take advantage of this new laser technology, this is the way you do it, because when he was in Union Carbide, that's how they did it. Now Union Carbide, would that have been where Gus Kensel was from? I think it was. Anyway, Weldon Schumacher got talked into the idea of opening up a research laboratory. See, there was a lot of difficulty in hiring people to Southbridge. There was difficulty in hiring in general because there was a shortage of technical people. And so they opened up a research lab in Briar Cliff Manor, which is in Westchester County on the north side of New York City, a very attractive area. And then it was a very attractive site that they got. And then the guy who was hired to run that I've forgotten his name now, but he was a man of considerable experience technically and administratively had been in the Hudson Laboratories, he had worked there in some significant capacity. He had some patents that related to some of the nuclear energy developments. He had been at Columbia. I've forgotten his name. Anyway, he was hired, and he was a good guy, I'd say. Then they hired a staff. They wanted to get into an area of lasers and as a matter of fact, in order to do that they had approached me about going to join that staff, but the way things stood, it was pretty clear to me that if I left, then I felt it would be an affront. I felt a real loyalty to Steve MacNeille. And I felt that if I left it could be the beginning of his demise as far as any influence in the company and maybe even his position in the company.

Bromberg:

Now was this Laser, Incorporated?

Snitzer:

That's what was established, yes.

Bromberg:

That's a wholly owned...

Snitzer:

A wholly owned subsidiary. And the thought was, you get a few bright people with an independent charter and then they go out and establish whatever their program is. And they hired this guy, and he was attempting to address the various needs of the corporation. I think technically he was quite competent. He was a reasonable guy all around. But the fit was a hard one, and there was built in from the very beginning a conflict situation between him and Steve. We all perceived that, and I perceived it in particular, and I felt a certain loyalty to Steve, and I was approached and asked to join them at Briar Cliff Manor. I wonder what would have happened had I done so. But anyway, I didn't. I opted not to. And then later on, the decision was made to put Briar Cliff Manor under Steve MacNeille.

Bromberg:

So that it would physically be there but administratively...

Snitzer:

Administered by Steve so that there would be one director of research.

Bromberg:

Now how many were there in this research department up here, just to get an idea. How big was the research department and how much of it was going into laser work?

Snitzer:

The research department then was maybe 160 people and the laser work maybe half.

Bromberg:

And that would include the sun-power lasers and the classified and the unclassified, the whole business.

Snitzer:

That's right. Making the devices as well as making materials and glasses. We were still socking in a lot of company money.

Bromberg:

And the Briar Cliff operation?

Snitzer:

How many people were there? 20 or 30 maybe. Maybe 40, but I don't think any more than that. They tended to emphasize more medical applications of lasers, because what happened subsequently was that a decision had to be made about which direction it ought to go, because it had been a little bit diffuse before Steve was put in charge of it, and I remember once he asked me(we had a lot of meetings with Steve about which direction to go in(and I said "If it's in lasers, there's no point of it going into glass lasers because they don't have available any of the things that we have here in Southbridge. But there are other areas to lasers that are more instrument-related and maybe they ought to get involved in that and that would be things like gas lasers, helium-neon lasers, CO2 lasers." I don't think we knew about CO2 lasers, then, but we did know about helium-neon lasers. Tape two, side two

Bromberg:

In Southbridge? Or down in Briar Cliff?

Snitzer:

No, it was in Briar Cliff, because Tom [Polanyi] had been with GTE in their Bayside, I think it's called Bayside(laboratory, not totally happy with the policies that GTE had with regard to their laser program. We knew about this because his brother Mike Polanyi had been working on medical instruments at AO in research in Southbridge, and Mike I considered as my closest friend in Southbridge.

Bromberg:

Now why do I think of Tom as a physician?

Snitzer:

He became very heavily involved with medical instruments. That was a CO2 laser application. That's where that came from. So he came on board first and worked on putting together a helium-neon laser, which he did quite successfully. He was involved in various laser things, involved with gas lasers. And then very quickly gravitated toward medical applications, something he was interested in, his brother was interested in it, and the whole thing just made a lot of sense, really. And then eventually the Briar Cliff facility was closed, but that was after '67. It's a big bridge in time, but at least you have that part of the story continuous. Warner-Lambert had then taken over American Optical.

Bromberg:

From Schumacher?

Snitzer:

Yes. Warner-Lambert bought out American Optical in '67. Then the laboratory was opened up in Framingham, because there was going to be a big emphasis on medical instrumentation. It was viewed that AO would be the medical instrumentation arm of Warner-Lambert which was in pharmaceuticals as well as in consumer products.

Bromberg:

Okay, well let's go back then before that, which sounds like a real turning point, anyway, and go before the takeover by Warner-Lambert. One of the things that just struck me, and I'd like to ask a little bit about it, is the very large amount of patents you hold, which is, I think, more than many of the people I talk to. Well, there are two things I wanted to ask about. Was that just company policy to do a lot of patenting, in other words, did they like to do patenting for whatever reasons? That's one question. And another question is on your own relation to that kind of inventive activity, whether you just enjoy making inventions or whether a particular project that you'd be engaged in would lead you to make inventions connected with the project, or what.

Snitzer:

As far as the company was concerned, different companies view patents from different points of view. Some companies view it basically as part of their marketing activity. For example, you ask about patent activities being different, say, at AO versus United Technologies. United Technologies doesn't attempt to make money from patents, that is, not directly, but they use that as part of their marketing, if they're going to sell to the government, and make the case that this is proprietary technology and get a sole-source contract. And that's a big part of their motivation. Not their sole motivation, but a part of it. Occasionally technologies come along which bear directly on the main line of their business, like jet engines, in which case they want patents on that. AO's posture was a little bit different. AO looked in terms of building a patent position in an area so they could dominate that area and then could license to make money from that. That's essentially what they had done in the fiber optics area. They had acquired all the significant patents. So if a company wanted to get into the business of fiber optics like imaging fiber optics, they only had to be licensed by one company, and they would be in business. And the thought was that if AO could build a similar kind of position with respect to glass lasers and glass laser devices, it would have a similar kind of commanding position and that could be a source of income. So they would license companies. And everybody who was making laser glass was licensed by AO. There was at one time a consideration given by some of the managers, the non-technical managers, to not grant any licenses at all but to make everybody who wanted laser glass buy it from AO. I recall expressing my opinion at that time that I thought it was a bad idea, because laser glass was still evolving, and you'd be in a funny situation where someone might have a glass that in one respect or another for a given application may be superior to what we would offer, and what do you do then? You end up with a blocking patent situation, and you're really impeding development. I would say the policy was mainly based on the idea that this constituted a way of commercializing on the inventions. The other feeling was that we're not a systems house. You know this distinction is often made that there are systems houses and components houses. The systems houses don't make any of the components, they just assemble and make the whole system function. That's a little bit of an exaggeration of what they do, because they all have to get involved in making some special things, even if it's just interfacing the components you have. This is distinct from the component houses, which are primarily concerned with making some of the basic elements that go into systems. The feeling was that most lasers had clearly taken a turn where there was some commercial activity going on, but most of the money that was being spent was military hardware, the funding was DOD. We had the ability to make a lot of this stuff, so the feeling was that a major way in which we would commercialize our activity, the way in which we would make money, would be to sell to the government or to sell to people who had contracts with the government. And as part of the program to make sure that we protected that capability, they had me in a situation where I was separate from a lot of this contract work, and where the stuff I was responsible for could be patented, and it was outside the contract work, so the company would have control of the patent. So a system developed, partially because I didn't have government security clearance at the time(where the feeling was to let me just pursue the work that I wanted to do and was interested in doing, and if any patents came out of that, they would belong to American Optical. And that would enhance our patent position without it being diluted by being partially owned by the government, although there were a lot of patents that did end up being owned by the government, because there was a lot of patent work that was generated by government contracts.

Bromberg:

And when you personally do this, did you sort of have in mind that whatever you could see of a laser system, you really ought to write up a disclosure? Or are you one of these people who enjoys inventing devices as a recreation?

Snitzer:

Well, I think both apply. Most of the disclosures that we had were only after the fact. If there was a new laser system, we would not propose it unless we made it work. I wasn't that heavily involved in government contracts, and if you were about to get a government contract, you want to file first and then you submit the proposal for the contract.

Bromberg:

Oh, I didn't realize that. But I see the logic behind it.

Snitzer:

In order to get the earlier dates.

Bromberg:

Sure.

Snitzer:

And then you argue with the government as to whether a constructive reduction to practice or real reduction to practice constitutes enough to control the thing. So usually what happens is you write up the patent, submit it to the patent office, that's a constructive reduction to practice, that is, you show how to do it, but you haven't actually done it. Although when you write it up, you write it up in a way that sounds like you did it. Then you write up the proposal and send it in to the contracting office. If the contracting office looks at it and says, "Yeah, looks pretty good; you can count it, but we've got to wait until the money's available," etc., and then when you know it's real, you can go back to the lab and you really scramble to try to make the thing actually work, because if you can now make it work before you get the contract, then there's no question whatever that the thing belonging to the company. That was the typical sequence of how people at AO scrambled.

Bromberg:

I see. That explains some things I didn't know about before.

Snitzer:

So companies were able to protect what they considered their own proprietary positions and at the same time participate in government contract programs.

Bromberg:

But you of course weren't under that pressure.

Snitzer:

No, because I didn't participate in that many government contracts. I was involved in kind of an internal consulting basis, but otherwise I was allowed to do pretty much what I wanted to do, in fact, I was encouraged to do as much as I was prepared to do on these things.

Bromberg:

Now there's something else that I find a little bit confusing that I'd just like to ask about. On the one hand, AO's stress on doing contract work in this period makes it sound as if they fully realized it was going to take a while for lasers to be developed into something useful. But on the other hand, they're setting up this Laser, Incorporated, which makes it sound as if they're trying to get in on the ground floor of this sort of commercial laser boom.

Snitzer:

Yes, that's right. They were pursuing both.

Bromberg:

I see.

Snitzer:

The various management policies were confused on that score. We had a difficulty, I would say, in that our immediate management like Steve MacNeille understood the process, but there were people above him who(I'd say this applies to Weldon Schumacher to some extent and it applied a lot to Vic Niss. Now Vic Niss was the second in command. Vic Niss' attitude was that what really counted was the ophthalmic business in terms of making money for AO. And someone once jokingly said that the major benefit of lasers to Vic Niss was that(he was a member of this expensive men's club, apparently(he could go into this men's club and say to his fellow members "We have the brightest laser in existence," which we did at the time. We had one of these oscillator-amplifier combinations that produced a very intense burst of light for a short period of time. So it was more of a prestige item for him with his associates. But he did not see it as a particularly financially remunerative effort. We had a laser retina L coagulator based on ruby, which involved a lot of complex optics in addition to the laser, and later it was used by Spectra-Physics or Coherent who came out with the retina L coagulator later based on an argonion laser.

Bromberg:

Bell had a lot to do with that. Eugene Gordon. Or is that quite a different thing?

Snitzer:

He did the argon-ion lasers, but I'm talking about the retina L coagulator. There was a commercial retina coagulator that was either produced by Spectra-Physics or Coherent. I think it was Spectra-Physics at the time. It should have been produced by American Optical, because the people at AO and the physicians that we were working with all were aware of the fact that it was necessary to shift the wavelength into the green, because you can do a lot more things with it. And we were just prevented from spending the money to develop that because of the reaction against the fact that here we'd spent so much money on the ruby retina L coagulator and we were just barely breaking even, so why spend more money in that area? And what basically had happened was that Steve MacNeille still understood what was going on in research, and still could project what was likely to be a money maker and where the resources should be directed. Above him were people like Bill Peck and Vic Niss, Weldon Schumacher, these were people who did not see it in those same terms, who were eager to exploit whatever financial benefits you could exploit now, and were unwilling to spend the money or see the money spent. They just didn't have the vision to comprehend how one proceeds in these fields.

Bromberg:

One thinks of the 60s as a period in which management had a lot of this kind of vision, that management was really willing to let a lot of science get done as opposed to the 70s which one thinks of as a period when you get more contracts.

Snitzer:

Some managers were willing to do it but I think the other bigger factor was you had a lot of government money pouring in, so this stuff that was related to basic research or advanced development got done, but it got done with government funds. There were a few selected companies or people who worked at those companies who appreciated how long it took and what the nature of the research costs was. So Steve MacNeille came from an Eastman Kodak environment where they were used to spending money over a long period of time to do something. It's the same thing at Polaroid here. I see the contrast between Polaroid and United Technologies. Here, they're used to spending money for a long period of time. They know that any significant new development is just going to take a few years and a few million dollars. They don't even think twice about it. At United Technologies, if the contract ends, the activity ends, unless you can show that if you spend some money, you'll get some return. Now that's not in all areas, I shouldn't be unfair to them. But I think that's a prevailing sentiment that exists there, and I think it's the mentality of the corporation, which is internally funded for its own product line, versus a corporation which to a very large extent is funded by outside government sources. And it's in response to their own special pressures that they encounter. So what happened was that there were a number of promising potentials that came up in that period. That was the time when some of these big interferometric controlled tables that were used by the then fledgling semiconductor industry were being designed and built. AO could have had a very good position in that, because there were some very sound ideas. People had begun using lasers and interferometers very early, not as a result of my inputs but as a result of other people's who were technically capable of doing these things. But it wasn't responded to because it wasn't understood, and there was a little bit of, for want of a better word, stodginess on the part of the top management, where they were very focused on short-term profits and wanted to cut down the scope of the research activity and have it more directed towards the immediate product line. [Looks at question list] On these various items which you mentioned as far as other glass dopers, yes certainly. What we were looking for were glass bases that would give efficient operation of neodymium, and efficient operation meant basically that the internal quantum efficiency for conversion of a pump photon to a laser photon be high, and the pump absorption bands be reasonably broad, so that the entire spectrum would be available for pumping. We did some fooling with sensitizing the fluorescence, because if you add another constituent like cerium, maybe cerium absorbs energy and transfers it to neodymium.

Bromberg:

Was that a big theme?

Snitzer:

We spent a lot of time on it, but nothing really came out of it. The most promising one was chromium, but chromium we worked on it for a while at low temperature, because the quantum efficiency for fluorescence in chromium in glass is not terribly high, so we had a thermal quenching of the chromium fluorescence. And then it looked like uranium for a while.

Bromberg:

As a sensitizer?

Snitzer:

Yes, but then the uranium was giving absorption from the excited states. So for a whole bunch of different reasons, none of those sensitized systems really worked, and it was just neodymium that worked.

Bromberg:

I see.

Snitzer:

I did some experiments with small fibers. There was always a lingering thought that there may be some interesting things you could do when you look at modes and stuff, but we didn't emphasize that. Mostly it was larger size rods, or convenient-size rods for various things, because people were thinking in terms of things like small range finders, which might be a few millimeter diameter by three or four inches long, but the big emphasis by far was the big, big rods.

Bromberg:

So what was going on there?

Snitzer:

There the intent was initially just to get efficient systems, and then there was the damage problem, and then there was the athermal problem, which had to do with the fact that if you pump the laser rod, you're producing a temperature variation from center to edge. Now the two different kind of problems that one confronts there, one has to do with the whole process of pumping. Excuse me, they both have to do with pumping. One has to do with a high repetition rate. For example, if you have a ten pulses per second then your loading can be expressed in a watts/centimeter basis, and it doesn't matter whether you have a somewhat higher pumping per pulse with a given repetition rate as against a somewhat lower pumping per pulse but with a higher repetition rate. It's the total wattage put in per centimeter that really counts. Under those conditions, the whole process of converting a pump photon into a lasing photon, even with 100% quantum efficiency, is going to produce heat in the material, and you've got to extract the heat, and if you extract the heat, you're going to have a temperature gradient from the center to the edge, where the temperature in the center will be higher than at the edge. Depending on how you're pumping, in a single flash, you could have just the reverse temperature gradient; the temperature could be higher near the edge and then lower at the center. But either way, you get a temperature gradient which you have to worry about, and that temperature gradient can distort the wave front, and in distorting the wave front, for small effects, it would lead to a divergence of the beam which is no good, so you no longer have a nice collimated beam coming out, and under extreme conditions, the rays that manage to get through along the laser rod would be such as to miss a fair volume of the material in the laser rod itself, so the net result is the efficiency goes down. So things can go to pot if you look at what happens as you begin to load the laser rod more heavily, by loading I mean in this case increasing the pumping per pulse or increasing the pulse repetition rate, or a combination of both, actually.

Bromberg:

Now how was this reflected in research?

Snitzer:

So we decided to go into athermal glasses. I had written a proposal, but it was a badly written proposal. It was just that kind of seat-of-the-pants calculation saying, "Yes, you should be able to balance a couple of things, the d/n/dT, the index of refraction change with temperature, you should be able to balance by the stress optic coefficient associated with the material. And the proposal was sent in and actually a patent application was begun to be assembled on that.

Bromberg:

When was this about?

Snitzer:

I don't remember. Probably '64, but maybe '63. I remember what happened was I talked about it with people internally but we decided we wanted to get the proposal in right away. This was the point at which I was working somewhat more with classified projects. I guess it came to my attention, and the claim was made that there was a very important problem to deal with and would I spend some time on it? I've forgotten how I got involved with it.

Bromberg:

So that proposal went down to Huntsville?

Snitzer:

No, I think it went to ONR, because Fred Quelle got involved with it later on. Then Fred came in and Fred said, "You know, you didn't really calculate that correctly." And I said, "Yes, I know that, and as soon as we get the contract, we will calculate it correctly." (Laughs.) In the meantime, Fred got very intrigued by the whole problem and so Fred went and wrote something up. He had suggested initially that we write a joint paper together on it. I was reluctant to do it. I felt that he had sort of taken advantage of his position of being the contract manager. See, if you're dealing with a contract monitor, you tell him things that you wouldn't tell somebody else who's competition, but then for the contract monitor to turn around and become competition by writing a paper, but then he said, "Well, let's do a joint paper." And in retrospect, it may have been wiser for me to have gone ahead and done it, but anyway, I didn't. And then he published by himself. And then subsequently in a review article I did publish that and extended it a little further to cover disks as well as rods. Anyway, the patent on athermal glass was in my name, and that predated any of the publications on any of that stuff.

Bromberg:

And then the damage, were you also heavily involved in that?

Snitzer:

Yes, I got involved with the damage and that was somewhat later. Well, the way that that came up was very interesting. We knew there was damage. We also knew there was need to get good optical quality. So we had made contact with various companies to produce glass, and one of the companies we made contact with was Parra Montoire. Now I don't know if Parra Montoire was part of St. Gebain. I think it was part of St. Gebain, which was a French company, and they had the reputation in the eyes of the glass community at least that they had good optical glass, being as good as Schott, particularly before World War II, but they didn't have the reputation of Schott. They didn't have a jazzy catalog, and a lot of things, but they did have a good reputation of solid glass chnology. And we had made contact through someone at American Optical handling foreign sales, and I recall at the Paris meeting of the Quantum Electronics conference a meeting with some people from Parra Montoire to go over what was needed for laser glass. And then we had a program where they made us some laser glass.

Bromberg:

Was this common for you not just to rely on AO, but to go outside?

Snitzer:

Well, the feeling was at that time, we had not solved the problem of making large melts of good optical quality glass, and the thought was we could get that quickest by having somebody on a contract basis from the outside do it for us.

Bromberg:

And this was again in connection with these DOD projects which were interested in these large glasses.

Snitzer:

And we had this glass made for us and the stuff was beautiful. We had indicated to them that it was important not to have inclusions in it because it would blow up. I remember specifically that that issue came up, and I remember that it came up in the strangest place. We went to an entertainment one evening, and I was with someone and that person said that so-and-so had asked about whether we could tolerate any small inclusions, and I said, no we couldn't, it would blow it up. But nothing more was said about it, because I think they had already frozen in the process they were going to use. And I didn't feel I knew enough about the engineering aspects of the technology of making glass to question it, except to state what was needed, but not how we're getting it done. And then they went ahead and made the glass and they sent it to us, and the glass was beautiful from the point of view of optical quality; you could look through a long length like that and could see no striae, no ripple at all. When you make laser rods of it, and you start to lase it, you blow it up every time, not even when it was Q-switched. It just blew up. Apparently it was loaded with platinum, it was made in a platinum crucible and made under reasonable oxidizing conditions, and that's the conditions under which you do get a fair amount of small platinum particles in the glass. There are a couple of mechanisms, at least two mechanisms to get platinum into the glass, and apparently both of those were operating with a vengeance. It was just loaded with these little particles. Of course, you don't see them by looking at it, but the laser light sees it and it heats up the glass and it just blows up.

Bromberg:

So this stuff would come into the research department. Now at this point, you're director by now, aren't you?

Snitzer:

No, I headed up a laser group. No, Steve MacNeille was still director.

Bromberg:

I've got you down as director of basic research from '63.

Snitzer:

Okay, basic research, yes.

Bromberg:

I see, but MacNeille was still in charge.

Snitzer:

Yes, I had a group. I was in charge of all the people who were involved with the research stuff in lasers, plus a few other things, some things that were laser related, Faraday rotation material, etc.

Bromberg:

I see, I didn't understand that. I thought you just took over MacNeille's position at that point.

Snitzer:

No, MacNeille was still there, and I took over effectively as the director of corporate research not until a fair amount later, the early 70s. So I was basically the head of a department within research.

Bromberg:

So this glass would come in and you'd cut it up.

Snitzer:

And that time I was heavily involved still in the lab. There were a number of people who were working with me, and we had frequent meetings, but I had my own lab, and I had a couple of technicians working for me. And I would say, spent most of the time working in the lab. That was the case until about '66 or '67, or '69 when I moved my office from Southbridge to Framingham, and then in Framingham I really pretty much stuck to administration. I still came back to Southbridge and supervised some stuff, but not very much. I was pretty much out of there.

Bromberg:

Okay, so that didn't help you with studying damage. That was too much damage to study.

Snitzer:

It was clear that damage was a problem, and that we had to do something about it. There are a couple of different approaches that could be pursued. Both of them were proposed by people at AO. Tape three, side one

Snitzer:

Then there was a big hassle about how do you get high purity mullite. We finally managed to get that in the right size crucible because we were making 50 pounds of glass per melt. And then there was the business of how do you homogenize this glass, because you always had little filaments of striae, and that was finally solved. So we were producing good glass.

Bromberg:

So you were very directly involved in glassmaking at this point.

Snitzer:

Yes, although there were other people on board who did more of the glassmaking stuff than I. I provided inputs as to what could or could not be acceptable. I guess I read up on some things in glassmaking. We had a couple of very good consultants who were extremely helpful to us. We had Don Uhlmann here from MIT who did some very nice stuff on calculations and interpreting the results we got on platinum inclusion damage. Then occasionally we had Al Cooper from Case Western consulting with us. We also at that time had Emil Deeg who was in charge of the glass lab, and he's an accomplished glass technologist. He used to be director of research for Schott. So there was a considerable focus on the problem of making good laser glass, good laser glass being the kind that wouldn't break. Well, we had other requirements that came up. It began to be recognized that for a number of applications, it would be nice to have glass for range finders. But in order to use the range finder, you'd like the gain coefficient to be relatively high. We were one of the earliest ones to say we should go after phosphate glasses. We didn't do very much with phosphate glasses. We did go after phosphate glasses because we noticed that the line widths and lifetimes were such that it looked to be the most favorable glass for getting a large gain cross section. And as a matter of fact, I remember in a summer study that was funded by IDA that was held quite early(it might have been very late 60s or early 70s, like '69 or '70, maybe '71(I was asked to come and talk to them. They had their meeting at La Jolla at the time, and I remember that one of the things I thought ought to be funded would be a program to investigate phosphate glasses, because phosphate glasses have what appeared to be a favorable gain coefficient, a large gain coefficient. It ended up that we did not pursue that. We did a little bit of pursuit of it ourselves, but not very much. We did do it in connection with erbium glass, and the erbium glass came up, well, that's another story. Let me just finish up with this other thing.

Bromberg:

Okay.

Snitzer:

There were some problems in phosphate glasses that we never really came to grips with properly and the net result was that we were not successful in having good phosphate glass on a consistent basis. Every so often, we'd get a good glass, just enough to tantalize us with the fact that it was good stuff. But we didn't understand what was going on. There was some speculation about what may have been going on, but we didn't understand it. We now know what the case was, and that was that there was a lot of water in the phosphate glass, and if you've got a lot of water in it, it's going to quench the fluorescence, and so what you really have to do is get the water out. And if you get the water out, you get good glass. And that was something that was arrived at conclusively by other people, principally Owens Illinois which was doing work on phosphate glasses.

Bromberg:

Well, we were going to talk about erbium, but I also want to ask about the world of glass lasers at some point.

Snitzer:

There was some other self-Q-switching glass which was kind of interesting and the erbium glass, but let me talk about both of those. Which do you want to talk about first?

Bromberg:

I don't care.

Snitzer:

Well, the way the erbium was going...it was recognized that you could make big systems reasonably efficient with neodymium. And people wanted to increase the efficiency by sensitizing, etc., but then there was also a concern, well, are we missing the boat? Maybe we should be looking at other materials. And we were doing the same thing in crystals.

Bromberg:

Now where was this concern? You said, "There was a concern. Are we missing the boat?"

Snitzer:

Well, we had a concern.

Bromberg:

You mean when you would talk in your meetings, and so on?

Snitzer:

Yes. Should we look at other materials? Should we try to lase others? And so we decided fairly early that we would look at the various fluorescent materials that could be made to lase, particularly those that you could do at room temperature, and get it to lase with a flash-lamp. Because if you had to lase it by pumping it with another laser, the efficiency of the overall system would be so low that it would not be all that attractive. Well, you could argue that it's good for producing another wavelength, but on the other hand, there are so many ways in which you could produce other wavelengths already in crystals and gas lasers that the motivation to do that was not all that great. So unless one could see some benefit by way of either an efficient system pumped with a flashlamp, or a system which gave you a wavelength that was unique in some way.

Bromberg:

That's interesting, because I think of the earlier period as a sort of mad dash for wavelengths without too much interest in whether they're useful or not.

Snitzer:

I guess what I'm distinguishing is that.

Bromberg:

I mean this isn't happening here, from what you're saying. There wasn't a mad dash for wavelengths.

Snitzer:

Well, there was some of that (laughs), but there was also another element too, and that was that here we had sort of staked out for ourselves going after glass fairly early, because there were a lot of other people doing crystals. Anything that you could do with a crystal, you probably could do with a glass, but then what's the big advantage to it, unless there was something else that impelled you? By that time, everybody was sophisticated enough to realize that glass and crystals were not that different as hosts. There were some characteristic differences, to be sure. There were some advantages to crystals and some advantages to glass. But like the alphabet soup laser, you know, there was thulium, holmium, erbium, ytterbium, well, anyway, I'm getting ahead of the story. Let me go back to erbium, because erbium was a significant new development that had taken place. It started with our interest in getting a Bloembergen quantum counter. If you look at the energy level diagram, the basis of a quantum counter is, say you want to detect the presence of the incident photon. And now you have a pump that goes up to here (level 2), and to the extent that you get an incident photon that populates this level, if you get an intense pump, it will kick you up to here (level 3). Then you can monitor a photon that goes over here, so this is output. So in effect what you've done is you've converted an incident photon into an output photon, now at a shorter wavelength where you can detect it more efficiently with visible detectors like a photo multiplier tube. This was called a Bloembergen quantum counter. Well, if you look at this, you can see that it really requires that it fluoresces from two different levels. There are not many rare earths in glasses that do that; probably the only host that does it to any extent is germanium glass as a host, and then it doesn't do it very efficiently either. So we thought that, if you look at the energy level diagram that was commonly used that was based on rare earths in La Clz by Dieke I think, it shows that if you've got ytterbium and erbium, it shows an energy level scheme where the ytterbium is just lower, so you get an energy level transfer like that. So now if you add a pump, and then you have the incident [photon], the incident [photon] would bring you up to here. But before you came back to here, the pump would kick you up to here. Then you would relax back to here, but if you had a high concentration of ytterbium, you could have a substantial probability for a non-radiative transition over to here, and then you could fluoresce here. So that relieves your requirement of a high quantum efficiency for fluorescence from this level. And erbium fluoresces well from here so the lifetime is okay here. This is a non-radiative transition back down to here, but we thought if you put enough ytterbium in the glass, that process could compete with this process. So we made it and it didn't work. It didn't even come close to working. Then when we looked carefully at it, we found that what was happening, basically, was this level is a little bit higher, and that's why it didn't work. The energy transfer goes from the Yb to the Er, rather than from the Er to Yb as would be required for the quantum counter.

Bromberg:

Does that mean that the energy level diagrams were wrong?

Snitzer:

The energy level diagram was for a particular crystal, and they were just shown as levels, and it wasn't very carefully done, and in the case of glass, an oxide glass, there was a little shift.

Bromberg:

You see, I'm getting a real new appreciation of one thing, or at least a new sense of one thing, which may or may not be right. I assumed that people just went to the library and looked up energy level diagrams, but now I'm assuming that there are lots of problems with energy level diagrams.

Snitzer:

Well there are, because it depends on what it's made for. You see, the energy level diagram very often is simplified somewhat and you get an energy level diagram for one material, but it may not apply in glass because you can shift up or down by a few hundred inverse centimeters, and that's enough to make a difference.

Bromberg:

So is it right to conclude that laser scientists in the 60s are going to treat energy level diagrams with a healthy suspicion in general?

Snitzer:

Yes, they're going to say "This is approximately where it is," but one has to make a measurement to establish precisely where it is. Now, what happened if this is actually higher, then it pumps to here, non-radiative to here (???).

Bromberg:

This is the ytterbium we're talking about.

Snitzer:

Yes. So now ytterbium is a sensitizer for erbium, and it does so with a vengeance because you put in 15 weight-percent of Yb2O3 and you only put in two-tenths weight-percent of Er2O3. The efficiency for the energy transfer is very high, and you really are able to make this stuff lase at room temperature, even though it's a three-level system.

Bromberg:

That is a cute story. So the erbium sort of came out of an accident.

Snitzer:

Right. It came out of an accident because of looking for a Bloembergen quantum counter.

Bromberg:

As I put in one of the questions, Bob Seidel pointed out that the erbium was less hazardous on the battlefield, and he wanted to know if that was an incentive.

Snitzer:

This was stuff that took place in '64, '65. Then the surgeon general decided that neodymium was a hazardous wavelength because of the transparency of the eye, so what you really wanted was something that could be absorbed by the front surface of the eye. But you don't want it to absorb in a very thin layer, but a few tenths of millimeters, which is the case here. This is 1.54-micron radiation for erbium, where it lases. So it was established that this does lase and then a big program was undertaken to get an efficient system that could make eye-safe lasers for range finders. The interest was entirely in range finders. At one time we thought with this we ought to be able to make the most efficient laser of all, because we could use neodymium now. It turns out that the neodymium has an energy level diagram like this, and this is the broad band here, and this level is up here, so you pump anywhere in neodymium, energy transfer to ytterbium, and then from ytterbium to erbium, and then erbium would lase. So all this stuff here from neodymium. The difficulty is if you get neodymium, ytterbium and erbium present, from here you get a back quenching so you can't put in very much neodymium, otherwise the neodymium quenches the erbium.

Bromberg:

Is this something that you already discovered in calculations or did you have to set this up?

Snitzer:

No, you set it up and try it, because you never really know about that. Energy transfer stuff depends not only on overlap but it also depends on some oscillator strength considerations, and you don't really know. So you do it. Looking at energy level diagrams for this kind of stuff(at least for me and the others who were working at AO(was a kind of a guideline as to things that one would try.

Bromberg:

So you'd go through the energy level calculations first, but then you would (???).

Snitzer:

Do some key experiment that ought to tell you what's going on.

Bromberg:

Another thing I want to ask in terms of the story is whether (???).

Snitzer:

Incidentally, there's another story here about neodymium and ytterbium that should be told.

Bromberg:

Okay, then just briefly, was the safety idea something that was coming in strongly after the middle of the 1960s, or was it always there as a steady thing?

Snitzer:

I think people were always concerned about the question of safety and worried about the question of safety. I think there's a great consciousness about it, which probably is a good thing, because there were very few accidents where people were hurt.

Bromberg:

Okay, let's go on to this other story.

Snitzer:

Well, this is an interesting case here because Ginther and Gandy had lased ytterbium at 1.015mm at liquid nitrogen [temperatures]. Then they combined neodymium with ytterbium and then, when they cooled it down to liquid nitrogen, it would lase at 1.015mm; when they had it at room temperature, it would lase at 1.06mm. So their conclusion was that neodymium would lase at room temperature, but ytterbium would lase at low temperatures. I looked at that and I was troubled by that. I thought, "How is that possible?" If you've got an energy transfer taking place, it's going to take place, and there's no way you'd expect the other way around. You would inhibit energy transfer as you go to lower temperature but the higher the temperature, the more likely is the energy transfer, where the more energetic processes for interaction between ions would take place. So I started to do some work, and then I published a paper on it where I confirmed that it did lase at 1.015mm at low temperature and at 1.06mm at room temperature. But when it was lasing at 1.06 at room temperature, there was also ytterbium lasing there too, because it could lase from both. And there was an experiment to set up what the inversion of the population was in neodymium and how that changed, and I think I published that in '66.1 So we had a fair amount of understanding of neodymium and ytterbium, and a fair amount of understanding separately of ytterbium and erbium on what could be done to sensitize the system. And then finally the system which was the one that worked well was you dope a core with ytterbium and erbium to get this energy transfer. And then you dope the cladding with neodymium and ytterbium to get this energy transfer. And then it turns out there's an overlap between the fluorescence and absorption bands of Yb, so when ytterbium fluoresces, that gets absorbed by the core, and that turns out to be quite an efficient system for coupling energy into the core.

Bromberg:

So this business of putting in a sensitizer does work very well.

Snitzer:

For erbium and ytterbium it worked superbly well.

Bromberg:

It just didn't work as a very general idea.

Snitzer:

It didn't work in the sense of improving neodymium. There wasn't anything that significantly improved neodymium. But it did work very well for both erbium and ytterbium. And for holmium it does work, in a crystal at least.

Bromberg:

This also touched on another question I had, and that was what the world of glass lasers was looking like as you get into the 60s. Your company is pretty much pioneering in it, but then a lot of people are publishing and I had a whole bunch of names here. And I wanted to know whether you had a lot of contact with these people, or whether you were all just thoroughly immersed in the whole laser community. I wanted in some way to get a sense of the professional world.

Snitzer:

With some of these people we had contact. There were several people at AO. Let me indicate what the nature of the contacts were. We had lot of contact with the people at NRL, Gandy and Ginther in particular, because they had done this work on gadolinium, and that was a short wavelength at .31 microns. I think it was subsequently established as a result of a lot of work in various places that it probably had not lased, but they had also done the work on ytterbium lasing at 1.015mm. They did some nice stuff, and were definitely, so to speak, "Members of the club," in terms of people who were active participants in laser glasswork. Pearson and Peterson I think made a significant contribution in the mechanism for energy transfer and quenching when you increase the concentration of neodymium in a host, which I think applied to crystals and glass. De Shazer had done a lot of work in making some careful measurements on gain coefficients as well as later on looking at non-linear properties. I don't remember what Hirakama's noteworthy contribution was.

Bromberg:

It's not so much what they did that was good as who you were in contact with.

Snitzer:

I had correspondence probably with all of them, but the people I saw mostly at meetings were Ginther and Gandy, Pearson and Peterson, and De Shazer. But Hirakawa I don't remember. And certainly, Bob Maurer at Corning and Mauer at Eastman Kodak, at Corning did some work on laser glasses. They were one of the companies that were sort of in and out of laser glasses at different times. I don't think they would put a big effort on it, but they pursued it as a research activity. Some government contracts supported their work. Mauer at Eastman Kodak adapted one of their compositions, lanthanum borate, which they could readily make that was used for making good quality lenses, because it was a glass that was very fluid at not too high a temperature, so you could easily get good optical quality. They doped it with neodymium and that indicates that most glasses could be doped with neodymium and would lase. You had to be careful that you didn't get quenching — there were some constituents in the glass that would produce that(and the other thing was that you wanted to make sure that the materials were of reasonably high purity, which means that you not have any transition metal ions or other rare earth ions that would quench the fluorescence. But those requirements are not that severe. However, the lanthanum borate turned out not to be a significant material, I think, because it got quenched too much by the presence of the boron. But in the early days, it was a glass that was available in good optical quality, because in the early days, the concern was to get a big chunk of glass of good optical quality so you could get a good wave front out of it. And we had contact with all of them; the largest contact was through meetings. There were scads of meetings. Optical Society meetings, Physical Society meetings, there were meetings coming up every few months. The results were coming out so rapidly that you always had new stuff to present at them; everybody was eagerly waiting for the latest development to be reported.

Bromberg:

You see the import of this question is a hunt for so-called "invisible colleges," for groups within the laser community that were in particularly tight communication and listened too each other a lot. I'm sort of looking for that in your case.

Snitzer:

Oh, okay. I'd say that the extent to which that happened at all was...You see, we began to get a fair amount of funding, and that funding kind of determined the direction in which things were going, and so we tended to have a lot of contact with the ONR people. But of the people who were actually doing the work, of this list of people here, we had contact with Gandy and Ginther, but not an unusually strong contact. I would not say that there was the emergence of a clique of some kind. I'm using a "clique" in the best sense of the word.

Bromberg:

What about people who aren't on the list? Is there anybody we ought to think of as people you were personally being exceptionally much in touch with? For example, to whom would you first communicate your results, besides the people here? Were there any such people? Whose results would you go out of your way to inquire after? That's another way of getting at this kind of thing.

Snitzer:

I don't know. I guess I would communicate it first internally to my boss, Steve MacNeille. By that time we had a fair amount of government contract support, and so we'd make it a point to let the various people in the ONR know, and those who were in the laser community that represented government activities. As far as other people in the laser area, it was mostly going to meetings. We had very cordial meetings; everybody was just excited and interested in what was going on.

Bromberg:

And there it was kind of everybody.

Snitzer:

And the meetings were held quite frequently. After a while, it settled down to an annual conference. There was always at least one major meeting annually, which the various laboratories were expected to participate in.

Bromberg:

And then there would be these smaller meetings, like NEREM or WESTCON?

Snitzer:

Yes, there would be things like that, although...

Bromberg:

There would be military-sponsored meetings?

Snitzer:

Yes. The Physical Society meetings or the Optical Society meetings, as distinct from special meetings that were sponsored by the Bureau of Standards on laser damage, for example. There was a series on that, and then there was an ad hoc committee on laser damage.

Bromberg:

Were IEEE meetings important for you?

Snitzer:

IEEE and OSA tended to have joint activities fairly early on.

Bromberg:

At one point I asked you whether the fact that you had an EE bachelors and you taught at Lowell Tech, whether this brought into your physics work any kind of electrical engineering knowledge or method that you're conscious of.

Snitzer:

Yes, probably so. Certainly with things like microwaves, because when I got a bachelors' degree in engineering, that was during the World War II when microwaves were the big new thing and wave-guides, stuff like that. That was the most advanced thing I had learned as an undergraduate at Tufts.

Bromberg:

When you taught at Lowell Technological Institute, did that skew you in any way towards engineering, or were you teaching physics there?

Snitzer:

No, I was in the electronics-engineering department, but there was no great distinction between electronics engineering, and physics, particularly at an undergraduate level. I guess that even though I was in physics, I probably always had more of an engineering physics outlook.

Bromberg:

I've heard from some people who told me they felt there was a real distinction, and the physicists, for example, really didn't understand oscillators; they really didn't have a good sense of oscillators, people have told me. On the other hand, I have the intuition from speaking to people, one's told me this directly(that the engineers weren't really interested in stuff like coherence studies. I don't know if any of this is true, but I'm wondering whether for somebody like you who had kind of a double background.

Snitzer:

I don't know what to say about that. My own background started out as electrical engineering and then it was in physics, but then when I began to be active professionally, I viewed myself as an optics person because I was at American Optical which is very heavily optics oriented, and physical optics was a very important part of what people were expected to know in the research department. And so my identification is very strongly with optics. The advanced stuff was physical optics. There were some coherence things, coherence in the more classical optics sense. I understood some of the things that Roy Glauber was doing in quantum coherence, but not enough to do any reasonable original work in it.

Bromberg:

Were you following, say, the Wolf-Mandel-Glauber stuff very closely?

Snitzer:

Yes, because Glauber was consulting for American Optical at the time. He was actually brought into this by Saul Bergmann. Saul Bergmann had approached him to be a consultant and then got arrangements for him to consult with AO. This was after Saul and I were no longer working together. So Saul was off by himself doing some things, and I guess he decided to pursue some theoretical work, and he approached Glauber to be a consultant with him. That was approved by the management of AO. Tape three, side two

Snitzer:

And it was presented to me by Steve as to whether we wanted to keep Roy Glauber on as a consultant. I remember hearing a talk that he gave, and was very impressed by the guy. I suggested to Steve that we continue to have him on. And so we did, and Glauber's contact with American Optical was continued.

Bromberg:

You see, I often wonder, on the one hand what the impact on the working laser physicist was of the coherence studies that were going on, and on the other hand, another thing I'd like to find out is what the impact of the Lamb papers were on the working laser physicists were, because those theoretical papers are often said to be landmark papers.

Snitzer:

You mean those early papers on non-linear(

Bromberg:

Of course that was on gas lasers.

Snitzer:

Yes. Gas lasers were a little bit different. You had to worry about coherence phenomena, that is, coherence from a quantum mechanical point of view, which meant that you had to worry about the relaxation times, tau1 and tau2, the off-diagonal elements in the density matrix, those kind of things. It turned out after Glauber was in his consulting relationship with AO when I had contact with him; we were formulating a little program of taking the erbium laser and looking for photon echoes, because you need a three-level system for that. And so he went through some theoretical description, he said that he felt there was an easier way to present photon echo phenomena. And that was going to become the basis of a Ph.D. thesis by one of the people at AO, a man named Herman Swope at Worchester Polytech, but he ran into experimental problems and the last I heard it didn't actually come to fruition.

Bromberg:

It sounds as if it was very unconnected with AO's interest.

Snitzer:

That's right, it wasn't, because it was just a straight academic piece of research.

Bromberg:

Did that happen often that a straight academic piece of research would be (???).

Snitzer:

Yes. It's very interesting: the attitude used to be taken, even by some people who were very doctrinaire in their attitudes about where money is spent, if it's something big and it's significant and it would bring a lot of glory to the company, then the more that that's the case, the more far afield can it be from the company's financial motivations. Those things were tolerated. More than tolerated, they were even encouraged.

Bromberg:

One thing I don't have very much information on, and something I'd like to ask you about how it looked from your point of view, is the interrelation of American work with foreign work. There was a trip to Russia in '63 that I sort of pegged a question around.

Snitzer:

I didn't go to Russia in '63. There may have been an invitation or there may have been a conference there, but I didn't go there until '75. I know I got an invitation for an earlier conference which for one reason or another — I've forgotten exactly why(I wasn't able to attend.

Bromberg:

Well, I'd like to get some feeling for whether you would be reading stuff by foreign scientists.

Snitzer:

Yes, we did. We were very much aware of and concerned with what was going on in France and Germany and England, because they all had competent groups there, there's Germany with Schott, France with Parra Montoire, England with Chance, and they all made laser glass of one sort or another, and we were interested in what they were doing. A number of these companies got licensed by American Optical, and there was a concern about just in general what they were doing. Oh, the Japanese, too. Hoya. We had contact with all of those companies from those respective countries. And then I got a visit from a Russian named Dianov, and later on there was another visit from a man named Sherbakov, and they were both in Prokhorov's group. And so finally that led to an invitation to go to Russia, and I went there in '75 with my wife. We went there as guests of the Soviet Academy of Sciences. There's an interesting little story. My wife wanted to go, and arranged for her to go there, in part because of our past interest in Russia because of being involved with left-wing stuff, and so we inquired, from various people who had been to Russia: how can we manage to get her in on the trip? I was thinking of writing them and asking if it was all right if my wife came along. I was told, "Don't do that, because if you do that, they'll say no." So the thing to do is, when they send you an invitation, you accept. You say, "I'm delighted to accept your invitation. My wife will accompany me." That's the way it's done. So I did that, and the invitation was forthcoming for both of us. With the invitation being for both for us, the company (AO) then picked up the tab. That was the significance of that.

Bromberg:

Did that inaugurate a closer relation with Russian scientists?

Snitzer:

It did right then and there, yes. We got invited to their homes, and then when they made visits here, we entertained them. And that persisted(let's see, that was in '75(until detente was over and then there was a sudden freeze on relations that was very striking in how suddenly it came about.

Bromberg:

Are there any influences from abroad that were important in the direction of your research?

Snitzer:

Well, there are some interesting things that happened as a result of that Parra Montoire business in France. There was a big French system that was being built that was purchased by NRL(I've forgotten the name of it, but it's the big French company that's involved in making large laser systems. At any rate, they designated the glass by a certain glass number, and we could never figure out what the story was. Then they in France. There was a big French system that was being built that was purchased by NRL — I've forgotten the name of it (CGE), but it's the big French company that's involved in making large laser systems. At any rate, they designated the glass by a certain glass number, and we could never figure out what the story was. Then they wouldn't say anything about what was in it or anything, and it was kind of a big mystery. They were making a lot of claims about how well it worked, etc. It turned out that number was an AO designation number of what we asked Parra Montoire to make. That we only found out about two years later. Within the Navy, there was the attitude taken that AO doesn't know how to do it, but the French do. And in fact, it's the usual kind of thing that people tend to be more skeptical of that which they know in detail and more trusting of claims made by people outside. The Russians picked up this whole business of athermal glasses very strenuously and that was what Dianov did. They wanted it for very high loading for welding and cutting, so they wanted a high repetition rate. They made it in a rectangular cross section slab, a slotted configuration. I used to get a lot of correspondence from various places. I had a number of visits from Izumitami, the director of research for Hoya. They have had a very active program in laser glass.

Bromberg:

But it sounds as if it was mostly a matter of their finding out of what you were doing, not their contributing new ideas or techniques to your research program when they came over.

Snitzer:

Well, I think what had happened was, it was well known what you had to do for the composition of the glass, and then an awful lot of the emphasis became building the system and how you configure that, and that would be done by a customer who would buy American Optical glass. So I think the work had evolved in such a way where it's like Lawrence Livermore Labs wants to make a big system. How would they go about doing it? They would make a bunch of calculations and stuff and then decide on a basic design. It turns out that their basic design was quite a bit different from what the Russians had decided on. And then just based on performance characteristics, decisions would be made by a third party that might want to build the system. And for the most part, they tended to build it like what Lawrence Livermore Labs had done rather than what the Russians had done.

Bromberg:

By the way, you were just in the middle of saying something, and somehow we never completed it. You said that you began to think of yourself as an optics scientist, and then I was going to ask, would mean in a sense that you would professionally feel most identified with OSA?

Snitzer:

I identify with OSA, yes, more than perhaps with IEEE or with the Physical Society.

Bromberg:

Now this is a period in which optics is really being transformed.

Snitzer:

And it's part of the reasons why I felt comfortable with that association.

Bromberg:

Because of the incorporation of electronics and so on.

Snitzer:

Right, because of the incorporation of quantum electronics.

Bromberg:

That's a tricky story, which I don't know how to unravel. (PAUSE) Then we want to go next to this question about how the research department got somewhat reorganized in the mid-60s to spin off a separate group that was interested in product lines.

Snitzer:

Well, let me just say a word about before the mid-60s. There were two separate groups in the research department, one which had clearance which did mostly laser systems, and was not concerned that much with laser development from the point of view of materials, the development of new lasers, or the emphasis on materials to develop improved neodymium lasers. The bulk of the contract work, virtually all of the contract work, was handled by the systems group because(well, I take that back. There were two categories of contract. One had to do with improved glasses. That was done in the glass lab. That was separate from this systems group that emphasized putting together these lasers and looking at their performance. A range of different things were done there. These included an amplified spontaneous emission output, where you get a very intense blast just as a result of the spontaneous emission coming from one end of the system and propagating through the total system; it would increase in amplitude and the pulse duration would narrow, and so emerging at the far end would be just amplified spontaneous emission, it was not a laser cavity at all. And then ways in which you could gang the output of a number of lasers together by nesting the beams. Of course, mode-lock Q-switch activity was done by them. Oscillator-amplifier combinations, second-harmonic generation, third-harmonic generation, all of those things. And these were built into systems. And then the decision was made in the mid-60s to have this as a separate commercial department.

Bromberg:

And that's the systems group.

Snitzer:

The bulk of this commercial activity was the systems group, but they also structured it in a manner that was more of what I would call the traditional business organization, in which they had a marketing person, had some salesmen, set up a little showroom internally, had their brochures out, were into getting a product line, and had quotes, so it was run like a regular business. But the bulk of that business still continued to be contracts which they got for these big laser systems.

Bromberg:

I see.

Snitzer:

At least in the early phase the bulk of it was that.

Bromberg:

Now you were director of basic research at this point. Does that mean that most of the rest of the research group was now yours?

Snitzer:

Yes, I had responsibility for the bulk of it having to do with lasers. There were contracts that we had for improved laser glasses that I had only, say, a consultative function to perform because people in the glass laboratory had the responsibility of making these glasses. So their job was to produce whatever the laser glass was that we had contracted to make.

Bromberg:

Does that include the samarium cladding that was something you were working with and consulting on them?

Snitzer:

When I say consultative relationship, I didn't have any direct contract responsibility, but I would meet regularly with the people. Just because the place wasn't that big, whatever problems came up, we all discussed it together.

Bromberg:

This is about the point where Warner-Lambert takes over.

Snitzer:

Warner-Lambert took over a little later, yes.

Bromberg:

Does that make a difference in what you were doing?

Snitzer:

Yes. Warner-Lambert took over and then they very quickly liked the idea of the lasers because they saw that as a glamorous item. There used to be jokes about the fact that the major contribution that the laser program had made to American Optical was to help be responsible for a fat price which Warner-Lambert paid for American Optical. After Warner-Lambert came on the scene, they weren't perhaps as enamored with the laser project, because it was not the kind of thing that was a moneymaker. I don't think anybody was making money at that time in lasers.

Bromberg:

Maybe Spectra-Physics.

Snitzer:

Yes, Spectra-Physics was doing well. But there weren't that many commercial products that people had and certainly not for solid optically-pumped lasers. There were some devices for welding, cutting, and machining in general that could have been a basis, because we had indications that people were interested in what could be done with glass lasers for that. There were certainly indications from Europe that this was being pursued quite strenuously. The French and the Russians in particular were using glass lasers for that. A lot of machining. And a significant business was built later by people who spun off and established a company called Lasers, Inc. using glass lasers for welding and cutting.

Bromberg:

Lasers, Inc. ceased to be an AO company and went off by itself?

Snitzer:

No, what had happened was they had decided to close down the Briar Cliff Manor operation which was called Lasers, Inc. and they consolidated everything to either Southbridge or Framingham. Later when they decided to get out of the laser business, a group that was within this laser commercial activity, which consisted of one of the guys who was doing marketing and who had previously been a technician, and someone who was basically a design engineer for laser systems, set up a company. And then they thought that, inasmuch as American Optical was getting out of the business, they would negotiate it as part of the package that they purchase from American Optical, along with laser glass and various other things of that sort, they also purchased the name and became Lasers, Inc. And they became a successful activity. Basically what they do is they make laser systems, relatively small laser systems as laser systems go, but quite large compared to range finders. Their system is capable of delivering a few hundred joules at a shot, and capable of relatively high rep rates. And they were used for welding and cutting in industry.

Bromberg:

So if at AO you weren't interested in machining, then the products you were chiefly interested in were first of all the medical products and then the range finders, things of that sort. Did you produce many range finders?

Snitzer:

No, we didn't get very much involved in that. I think we tended to respond mostly to the potential for big contracts which had to do with big systems, big oscillator-amplifier combinations. We were still getting contract support from departments of DOD to make these big systems and then test them out to second-harmonic, third-harmonic, even fourth-harmonic generation.

Bromberg:

I keep bringing up Warner-Lambert. Should we talk about that, in the sense that it was going to make a difference to laser activity once they take over? Is that something that's worth pursuing in terms of understanding AO?

Snitzer:

I can say in general terms that when Warner-Lambert took over, they were looking for ways in which they could get some commercial products out of the research department, and they reached the conclusion that the laser activity was not likely to be successful and there were lots of pressures to cut back on the laser activity. I recently had an occasion to talk to a physicist who's currently become involved almost full-time in the application of lasers in medicine, and his contention was that AO played a pioneering role there in the laser retina coagulator and hair removal, but most important was the CO2 laser work that was started by Tom Polanyi. There was just an unwillingness on the part of Warner-Lambert to sustain the activity long enough for it to become viable, although the people who were running it(Tom Polanyi and the people he had with him(were competent enough as physicists and knowledgeable enough about how one translates an instrument idea into a viable commercial instrument by working with physicians. They knew how to do it; they had an intelligent approach. This person expressed the opinion that it was unfortunate that AO had pioneered in this, and then when it began to pay off, they were no longer interested in funding it. There was a reaction against lasers to the point where even when it was clear that there were about to be very large procurements for laser glass, AO had gotten in effect out of the field, and they sold paid up licenses for a fixed amount, a royalty payment from companies that wanted to be licensed from AO.

Bromberg:

Well, this must have really cut into your budget. That would be my guess.

Snitzer:

We didn't think in terms of budgets then, but we had to have approval for whatever program we had to undertake, and it just wasn't approved. That is, we had to in effect stop all the laser activity. At first, we stopped everything related to neodymium glass. This was in the 70s already, though.

Bromberg:

About middle 70s?

Snitzer:

No, we were out of the laser business entirely by maybe '74.

Bromberg:

And Warner-Lambert, what's the date they came in?

Snitzer:

'67. And for four years we continued the laser work pretty much unaffected. Then there was a considerable decrease in laser activity.

Bromberg:

And you start shifting into other kinds of research.

Snitzer:

Yes, right.

Bromberg:

Is that one of the reasons you decide to go to United Technologies?

Snitzer:

Well, it was related, but only indirectly related. What happened was that I had a falling out. They (Warner-Lambert top management) had one guy who was brought in, Pete Paugh, who was first brought in to replace Weldon Schumacher. He was someone who had been brought in from Gillette, he was a Gillette executive. And at that time, the model they had of American Optical was like a consumer products business. Warner-Lambert had a difficulty making an assessment of what kind of business AO was and how should they run it. And Pete Paugh was actually a very decent guy, trying very hard to do a good job of running it, but he had certain notions about what ought to be done, mostly along the lines that he saw it as a consumer product operation. He made a number of changes. The worst one that he made as far as I was concerned was that he replaced Steve MacNeille; he removed him from the position as director of research, which was very demoralizing for everybody. And a number of projects were started that were unrelated to lasers. And I had decided at that time that I was going to stay at AO probably. I had had an offer in 1965 or '66 at University of Rochester, which I accepted, but then University of Rochester equivocated about the offer because of the question on the part of the U. of R. president on government security clearance. Then subsequently, it was either a year or two years later(I had an offer again and at that time I turned it down because we had just moved to Wellesley and it was in '67, the same year that Warner-Lambert bought American Optical, and that was just so hectic for the family. It just didn't make sense to make a change that would be so disruptive. So I decided that I would stay at AO and at that time they were still involved with lasers, but then after the laser activity began to decrease, I thought well, there are a bunch of other things that AO is concerned with: they're involved with progressive power lenses for ophthalmic applications, there was the beginning of contact lens activity, there were still a number of instruments that they were pursuing, and they were involved with holography. I had more of a management function when we got to these things, but it seemed to me enough. And there were still some laser things going on, we continued to have a program on the erbium laser, which was an eye-safe laser, and we saw it as a cloud height measuring device. There was enough for me to see myself being gainfully employed and enjoying it. And then Warner-Lambert had a change of management, another one, when they replaced Paugh by a fellow by the name of Glen Hastings. Now Hastings, you may have recently heard the name. He was one of the guys who was in the Atari games thing. He was one of the principal players in a lot of wheeling and dealing. He had a view of the company as being like the proprietary pharmaceuticals, that is, he saw it in terms of a product like Listerine, for example, which is what Warner-Lambert was built on initially, and he had no better grip on what American Optical was all about than Paugh did. He made a number of changes, and the major one of these changes had to do with the laser program; in effect, to cut out the laser activity. The only thing that survived was the erbium eye-safe laser. I remember one thing that happened. We worked real hard to get a contract. We first had a program to build one erbium eye-safe ceilometer, which is a cloud height measuring device, and then we were lined up for a contract to build four, because that one worked quite well and everybody was happy with it. The contract agency was NOAA, and we had the contract. We got it sole source, because NOAA was so eager for it because it worked so well. And then it only required his (Hastings) signature to accept it. We knew how much it would cost to build, because we had built one, we were clear that we were going to make some money on it, but he just was emphatic: we were getting out of the laser business, close that up, that's finished. When that happened, from that point on, we had such a disagreement about things that we never really had a constructive conversation about anything.

Bromberg:

So that from that point on you were really bound to leave.

Snitzer:

Yes, it was clear that I would eventually end up leaving the company. As part of the various rewards along the way, I had a deferred payment kind of thing, that and other perks I would get if I were officially laid off rather than quitting. Usually these things were done right at the end of the year. So I had a discussion with him, told him what my feelings were, that there didn't seem to be any place in the company the way it was structured for me. He suggested a staff function having to do with getting FDA approval for medical instruments. (Laughs.) It was something that I was totally unequipped to do. I had no talent, I knew nothing about it. But it was something that had to be done, and he was suggesting, well you could do that. It's strange; on a personal level, I think we got along reasonably well.

Bromberg:

He doesn't sound as if he was really a good manager. Did he build the place up nicely?

Snitzer:

Oh, he left it in a shambles (laughs). Paugh didn't help it much, but Hastings really drove it into the ground.

Bromberg:

That's an interesting story. I think some business historians are going to be very interested into looking into this.

Snitzer:

Oh, he really did a job on it. I mean, in all kinds of ways. It was the real downturn of the profitability of American Optical. Before then American Optical was a reasonably profitable company.

Bromberg:

It sounds a little bit like one of these classic textbook cases of where a conglomerate takes over and doesn't know what the industry is about.

Snitzer:

It's like United Fruit and bananas, except that that was Eli Black, in this case, it was Hastings. So then at the beginning of '77, we had an understanding. I was formally laid off, I had a certain amount of severance pay. We negotiated a severance package. It was an amiable departure. And then I went to work at United Technologies. At United Technologies they wanted me primarily for fiber optics. Tape four, side one

Snitzer:

And that was a period of time in fiber optics when fiber sensors was one of the active research areas which would be consistent with the resources and talents we had there. We had a reasonably good program. We did some worthwhile things that led to publications and some patents. There was not a great deal of laser work. We did some work on the erbium laser, and from time to time issues would come up in regard to what were in some of the foreign publications. We were asked to comment on neodymium. But mostly I was an internal consultant, no direct work on it. There was some work done by some of the divisions, but not by me. I did use the knowledge of laser glasses as a basis of a temperature sensor, where by redistribution of populations in various energy levels we changed the absorption coefficient at selected wavelengths for rare earths doped in a glass. And that was the basic temperature sensor.

Bromberg:

Why don't we go back to the color center work now that we have kind of a general framework?

Snitzer:

The color center work was very interesting. We had made a lot of different glasses and we would make them into rods. And then these rods would be either studied to determine their properties, or in some cases, they were just used as standard rods by people who had other things in mind. And there was one rod that had a sort of unknown history. We didn't know where it came from precisely, but that one of the guys in the lab, Chuck Koester, was using, so we used to call it Chuck's rod. Chuck's rod had this strange property that when you pumped it at a low level, this was a clad rod of terrible optical quality(it gave kind of a broad beam, maybe 6 degrees output, but quite a bit of beef coming out of it. And when you pumped it at a low level, it had the standard damped oscillation. There's a transient response where you have the non-linear oscillation, which you no doubt are already familiar with. You go like this and gradually it damps out like this. And it had this typically damped oscillation when you pumped it a little bit above threshold. But as you pumped it harder and harder, it would convert over into limit cycles, and this would just stay like this.

Bromberg:

Limit cycles just means they're evenly spaced cycles.

Snitzer:

A limit cycle is a non-linear oscillation process. Any non-linear oscillation is a limit cycle, really.

Bromberg:

Okay, I don't want to hold us up on that.

Snitzer:

The key point about this is that this is an incipient form of self Q-switching. And if you drive this harder, what happens is the spacing between these gets bigger until finally you can get to the point where you get only two pulses and that's it. And that was the basis of one of the publications on self q-switching. Well, for the longest time we didn't understand what was going on with Chuck's rod. We didn't particularly pay much attention to it, but were just fooling around with it. And then Bill Shiner, who was working as a technician he was one of the guys who set up Lasers, Inc.(made the observation that made it possible to conclude that it was a color center being formed.

Bromberg:

What was that?

Snitzer:

I don't remember what the observation was. Maybe it was that if you pumped it harder, you got this limit cycle behavior, and when you pumped it easier, you did not. Or maybe it was that he put a tube around the outside of it, and then you could pump it hard without getting limit cycles. I've forgotten what it was. Maybe it was that if you put this inside of a pyrex tube, and then went through the whole business of pumping the laser rod, it did not show limit cycle behavior. So that meant that it was not a characteristic of how hard the rod was being driven, but rather that there was ultraviolet light. That's probably what it was, that ultraviolet light was essential to their formation. So then the question was: what does ultraviolet light do? It produces color centers. So then we had a systematic program of varying the amount of exposed part of the rod, and putting sleeves of pyrex over the rod so you could expose different lengths of the rod to ultraviolet but pump the whole rod with pumping light.

Bromberg:

Now I had the impression that even apart from this, the effect of ultraviolet light on laser glass was of great interest. Is that true?

Snitzer:

Yes, that's correct. It was.

Bromberg:

Was that in the glass laboratory, or in your group, or in both?

Snitzer:

Well, everyone was concerned with it because there was the fear that it would produce absorption that would basically lead to inefficiency of operation of the laser. The other thing was that it produced a lot of heat, because to the extent to which you absorb the ultraviolet photons you don't get that much output. When you're absorbing down to 3 microns and you're emitting at 1 micron, you've got two thirds of the energy converted into heat.

Bromberg:

So in a sense it was that these two expertises were coexisting on the one hand, an expertise in solarization (is that the right word for ultraviolet?)

Snitzer:

Solarization, yes.

Bromberg:

And on the other hand, an expertise in the crazy idiosyncrasies of Koester's rod.

Snitzer:

Yes, that's right. So there was Bill Shiner who made the observation, I think that's what it was(that when you put that tube on the outside, it had something to do with changing the ultraviolet light present, which changed the property of limit cycle behavior versus no limit cycle behavior. So then, with that as the clue, we then went back to look at the rod more carefully, and then checked a whole bunch of things. We concluded that we must have taken that rod from a certain melt, and that melt was characterized by certain things. So we went back and deliberately made glass of that composition. You normally put antimony into a glass. There are various things you can put in. You can put either antimony, cerium or arsenic as what they call a fining agent that gets rid of the small bubbles that are in the glass, because it changes the valence state and the bubbles are often associated with oxygen which leads to the oxygen being soluble in the glass. And if we left that out, then that led to a situation where these color centers could be formed. So anyway, that's what we did. And sure enough, we could systematically produce this anytime we wanted. So we could make this self Q-switching glass quite readily and then a bunch of compositions were made of the self Q-switching glass. The other property that these fining agents had, and antimony in particular, was that it influenced the near-ultraviolet absorption. So it was clear that near-ultraviolet absorption was required. So you limited the near ultraviolet by the glass itself, and had the absorption take place instead by the antimony that's present. Or maybe the antimony acted as traps; there were a number of unanswered questions still associated with that.

Bromberg:

Now how did this work as a Q-switching device?

Snitzer:

Pretty good.

Bromberg:

Did it become one of the standard Q-switching devices?

Snitzer:

No.

Bromberg:

What happened to it?

Snitzer:

It's an interesting kind of Q-switch. It's totally different; it's like a mode-lock. I have to sketch this.

Bromberg:

Was this something that was developed as a product at AO?

Snitzer:

What happened was this was the basis of some contract support that we got? No one was particularly interested in doing this, because people either wanted Q-switching where they could control the Q-switching at one pulse out which they would do externally by whatever the Q-switch is that they're controlling it with, or they wanted a mode lock in which case they didn't want it uniformly distributed. There was one place where there was an interest in this and that was for a battlefield illuminator, they called it a battlefield illuminator, and that is where they could just illuminate a scene with a series of saturation pulses and where the beam spread wasn't particularly significant. It was much easier to do this if you could have a wider beam spread, a few degrees. So it turned out the combination of all those factors was such that there was only one place where it looked like it was applicable, and they wanted a high rep rate. We could get a high rep rate, 6 to 8 per second, with as much as 100 joules output per pulse. But it didn't go anywhere. That was the only interest in it.

Bromberg:

You did a pretty thorough job of working on the physics of this?

Snitzer:

Yes, we looked at the nature of the color centers that were present. We had a very competent consultant working with us, Prof. Ralph Bartram from University of Connecticut in Storrs, CT.

Bromberg:

Was that something which you did because of this contract or did you just get interested in it?

Snitzer:

We got interested in it and we also had a consultant at that time, a very competent guy, Ralph Bartram, who was at the University of Connecticut, and we had one of his students actually we had a couple of his students, but one of them was working on this, Bob Landry. It looked like this could fit a model for color centers based on weak polaron theory. So that led to publication involving Landry, Bartram, and myself, where we explored this experimentally and theoretically.2 Later on, when I visited Russia and talked with some of the people who were working on these things there, one of the people, a fellow named Tolstoy, who's not a relative of Leo Tolstoy; Tolstoy is apparently a name as common Smith is in the United States, who was the grandson of Alexei Tolstoy, a favorite of Stalin's. And he contended that it required a small presence of iron in the glass in order for this self Q-switching action to take place. There's a weak absorption that takes place in the near ultraviolet, and that gave the electrons and hole traps that were short-lived, and we speculated based on the properties that weak polaron theory would explain it. And his (Tolstoy) contention was that it would furthermore require the presence of iron, a small amount of iron. Exactly how the iron acted in the material was not entirely clear, but they did demonstrate that if you had very clean glass free of iron you wouldn't get it.

Bromberg:

At that point of course you had done this some years ago, but it wasn't completely understood. Okay, I have some questions about whether you were involved in one or another applications which I don't think you were. You were never really involved in fusion at all, were you?

Snitzer:

No.

Bromberg:

There was this paper you worked on with Bernd Ross when you were in Pasadena for a while. Was that an interesting episode?3

Snitzer:

Yes, it was. He was on the West Coast, but that was actually done at Southbridge. He brought some equipment to Southbridge and we set the experiment up in Southbridge. He brought the light source, one of the semiconductors which emitted at 1.06mm, and the drive circuits, etc., and then we set up the pump for the fiber amplifier. And there was a related experiment that was also done on that as an amplifier that used one of the helium-neon lines which emits at 1.0621mm.

Bromberg:

How did he get interested in doing this with you?

Snitzer:

Well, he was interested in semiconductors. He had approached me. There was some interest and concern at that time, which was based on a proposal I had written which never got funded, but the idea was that if you had only one diffraction-limited point you wanted to look at, the noise that you want to see a signal against is based on the fact that...In a maser, you use the maser as a pre-amplifier. So the question was, could you use a fiber laser as a pre-amplifier? And if you've got the fiber like this and light comes in over here, that's the signal. But you have the equivalent of injected noise associated with the spontaneous emission. In fact, I had written a paper describing that but it wasn't in the regular journals. It was the first all-European conference.

Bromberg:

Is it on your bibliography?

Snitzer:

I doubt it, because I looked at it as kind of a semi-commercial conference. This was the paper and you probably don't have it there. "Fiber Lasers and Dispersion in Fibers," reprinted from the Proceedings of the First European Conference.4

Bromberg:

Yes, it's in there. Item #25.

Snitzer:

In that is a description of fiber lasers that starts here. Actually, the first part of the part of the paper is not really a very good piece of work because we neglect the materials. We handle the materials part separately from the structure part of the fiber. And as a result, there are some cross terms that are neglected and that makes the results wrong. So one shouldn't look at the first part. But the fiber lasers, there are not many places this is described. That is, what is the spontaneous emission, which you get into a fiber? Because the spontaneous emission represents injected noise, basically. And I do present that, and it's a sensible presentation. It also goes into the business of gain saturation in a fiber. And this was actually done well. This was very simple, but a very worthwhile, very nice presentation of the topic, I would say. The basic conclusion of this is that the noise which is injected in terms of the number of photons per second equals the number of modes times the line width of the laser. And if you had a narrow light source and could filter the output here so this is made narrower, you could make this sensitive to as much as only one photon of signal.

Bromberg:

So just for the tape's sake, if you can make the frequency interval sufficiently small.

Snitzer:

If you make the frequency interval narrow by using a filter here and have just a single mode. The single mode is what was very interesting, because the line width is typically quite broad for, say, a glass laser. But if you had a light source that was generated, say, from a crystal or from a gaseous light source, that could be made quite narrow here, and then if you have a single mode, you'd have really a very low noise pre-amplifier.

Bromberg:

Now is that connected with the thing you did with Ross?

Snitzer:

Yes.

Bromberg:

This came first or did this come later?

Snitzer:

This came first, because I think I was trying to peddle this and then somehow or other it connected with Ross. When was the Ross reference?

Bromberg:

Well, the Ross reference is 1970 in the Journal of Quantum Electronics, and this reference gives a date as 1972. But that doesn't mean you didn't do it first.

Snitzer:

This came first because this was a case where we took the neon emission at 1.0621 and amplified it with a glass-fiber amplifier. And then we did the Ross stuff. So this was kind of a summary of older work that had been done. Yes, I remember now. We were trying to promote the idea that for a detector to have an array of single-mode fibers, all of them pumped in a rod, and to image the scene that had been illuminated with a narrow-band wavelength source, you get the backscatter, and then each fiber amplifies only that light which is in that one image point so that you can use the single-mode idea.

Bromberg:

Did you particularly want a low noise in that?

Snitzer:

Yes. The idea was, say we are looking to image a scene. By a scene I mean some distant object which doesn't have too many image points on it, maybe 100 x 100 or 200 x 200, whatever it might be, say 100 x 100. Well, maybe that gets to be a little bit too much. Say it's 10 x 10 (laughs). Then you would illuminate that scene with a narrow-band pulse of light, say, as gotten from, for example, neodymium in YAG, or an amplified 1.0621 neon emission, so that's a narrow-band emission. Then after you illuminate that you take and you image that scene onto the end of a fiber bundle consisting of a bunch of laser fibers. And then these fibers are all pumped, and because each fiber is a single-mode fiber except for two polarizations so this would be equal to two, and then your line width is very narrow, so this is a small number. That, then, constitutes your noise, so the number of signal photons you're able to see could be quite small, and your signal-to-noise would be better than you could get with any other light source. And it's still a sensible idea.

Bromberg:

But it hasn't been done.

Snitzer:

No, because nobody's motivated to do it yet, because there's no need for it now as a practical device. It's a specific device that you're motivated to do only if you've got a problem that requires it.

Bromberg:

And so this would be kind of an example of the kind of thing that's worked out, invented, but it really doesn't fit in anywhere.

Snitzer:

There's no need for it now at the moment. But somewhere along the line it will probably get picked up.

Bromberg:

I guess what I'm really asking is would be correct for me to see the research activity going on here as a certain amount of this kind of invention which isn't taken up particularly but just a nice idea.

Snitzer:

You mean does that continue here?

Bromberg:

I don't mean here, I mean looking at American Optical in the early 60s and early 70s.

Snitzer:

Yes. The justification at the time was that here's an interesting way of getting a much better signal-to-noise for low signal strengths for imaging a distant object. Tape four, side two

Bromberg:

Well, my final question had to do with how you characterize your style of doing science, and I put in a lot of sub-questions whether you preferred to work out theory first. But you sort of answered that one by saying that you do a certain amount of calculations, but then you get into the laboratory and see how things work out in the actual.

Snitzer:

The theory sort of acts as a guide to see whether one is anywhere near the right place, but then as soon as one is able to get into the laboratory and begin to get some results of some things. But that's not always the case. It depends on what it is. There are some things, for example, that are involved with cross-talk in fibers, where I felt very strongly that that was a completely calculable thing, and one ought to calculate that first because there were no uncertainties there. If you got a peculiar result that meant that something in the experimental set-up was wrong. In a lot of this laser work, there are various things that can go wrong; after you've made the calculation as to what should lase, it won't lase, for example. And because of that, it's so valuable to get some laboratory equipment and some results indicating where one is and what the problems might be. So at least one is not addressing it as a totally abstract question but rather where there is some indication of some kind of experimental performance of the system or material or whatever.

Bromberg:

Then in a way we somewhat handled this, but I wondered if within the whole group of theories, there are certain theoretical frameworks you most customarily employ; I wondered what your most important theoretical tools were. Clearly you did a lot of work with electromagnetic theory.

Snitzer:

Yes.

Bromberg:

What theoretical developments did you really employ most? We spoke of Lamb's work being really directed towards gas lasers. Is there other work that was being done that fed into your own research that we ought to (???)

Snitzer:

Well, there was an awful lot of materials behavior that was important. This had to do with index of refraction changes with temperature, and stress distributions, because when you lase a rod, you get a temperature gradient. You get what I call more or less classical physics having to do with what's generally considered as part of the elasticity theory of how the stresses are present in a rod, and the kind of heat exchanges that are produced from them. I guess basically classical electromagnetic theory and elasticity theory would describe the whole thing. It didn't involve that much quantum mechanics. Now there were some quantum notions, to be sure. The idea of energy levels, there's no way of avoiding it. The full-blown quantum mechanical description of a solid laser you really didn't have to go into, because except for the photon echo experiment which I didn't end up doing; it was somebody else in the lab attempting to get a Ph.D. thesis(you didn't have to take into account the off-diagonal elements of the density matrix, and if you deal with diagonal elements alone, all you need deal with is populations, and there are no other complications necessary to cope with.

Bromberg:

Then another thing I wanted to ask is I'm getting the impression that you did both theory and experiment equally. For example, there are some people who will say to me, "I'm a great idea man, but really I'm kind of a klutz in the lab, so I mostly had the ideas, and my collaborators mostly did the actual experimental work. But from our conversation, I've gotten the idea so far that you equally were doing the theoretical work and the laboratory work with not a real preference for one or the other.

Snitzer:

I don't know if I'd call this theoretical work, because I pursued some theoretical work, there is no question of that, but for the most part I used the theoretical work as a kind of guide for what would go on experimentally. I was involved in the laboratory but I was often involved in the laboratory when other people were doing the lab experiments and I would collaborate with them about it. I'd have the glasses made and I'd make decisions about configurations, or how to put the thing together as I worked with technicians. And later on, I tended to work more with senior people where I discussed what I thought made sense to proceed with and then they would sometimes do it and sometimes they wouldn't, depending on whether it would work out. They had latitude to pursue their own ideas, but it was an interactive process.

Bromberg:

You might call it an administrative experimentalist that you evolved into. You sort of were partly doing these as a kind of (???).

Snitzer:

Yes, I was sort of the administrative boss, so to speak. They did report to me; on the other hand, they had the freedom to pursue whatever they thought made sense. And very often what happened in those situations was pretty soon it evolved in a direction where most people I've found really do want to talk to others about what they've doing. They get a result, and they don't know what it is, and they want to talk to others about it. I pretty much adopted the attitude that anybody who felt they had a good consultant that they could interact with effectively and who was knowledgeable in the work that they were doing, then by all means take them on. Because if it's a good consultant and the interaction's a good one, and the amount that one pays a consultant is so little and the returns that one can get from them are immense. I encouraged wherever possible for consultants to be taken on.

Bromberg:

My final question is: have we missed something we should have talked about?

Snitzer:

It seems to me on this fifth page [of questions] that there were some things, but let me just read it quickly. The stuff with Bernd Ross is a whole area that I think is interesting. I still do think it's interesting. And that was one of, as I say, several experiments that were done having to do with the fiber laser pre-amplifier.

Bromberg:

This one was published in collaboration with Ross. The one that you just described to me was in the summer of '72.

Snitzer:

There are a couple of presentations made at I guess CLEOS or CLEA. I forget what the name of the conference was at that point. But that's these two, the one with Holst, Wallace and I on the high coherence-high power laser system at 1.062. That's a reference in this publication. Do you have that there?

Bromberg:

No.

Snitzer:

Okay, this was an abstract only, and that was an abstract only, too.

Bromberg:

In fact, I think I looked for those and didn't find them. Yes, I did look for those.

Snitzer:

There was Holst and I and then there was Holst, Wallace and I.

Bromberg:

I just didn't find it.

Snitzer:

And then we were going to write this thing up as a publication but then that was kind of late and there were a lot of changes. See the one in '69. Warner-Lambert was raising hell with us.

Bromberg:

They weren't interested in optical communications.

Snitzer:

No, that was furthest from their minds.

Bromberg:

Am I right that this was connected with optical communication applications?

Snitzer:

It was a device for doing high-resolution, high-sensitivity imaging. It does have applicability to optical communications, but it was not primarily that.