Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of George Pimentel by Joan Bromberg on 1984 May 18, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/31409-2
For multiple citations, "AIP" is the preferred abbreviation for the location.
This interview is tightly focused on Pimentel’s creation of the first chemical lasers in the mid 1960s. Relation to his Air Force funders. Roles of his graduate students and post-does in the research. Experimental techniques and instrumentation. Place of chemical lasers in his over-all research program.
Today is Friday the 18th of May, 1984, and Professor Pimentel and I are going to continue our conversation.
It’s probably worthwhile making a remark which seems in retrospect to be part of the explanation for why we were able to discover the first chemically pumped lasers, and in fact, more or less stayed ahead of the field for a few years. The popular and more or less obvious way to look for a chemically pumped laser that attracted almost everyone’s attention was the expectation that the proper kind of chemical system to use was a continuous reaction, that is chemiluminescence. Chemiluminescent reactions are normally known because they’re chemiluminescent in the visible spectral region so people looked for chemical reactions which were known to emit light in the course of the reaction and in the visible region. This was done in more or less conventional techniques and in fact, we thought in terms of conventional techniques with throwing to reactors together to cause the reaction to occur in a zone which would be the zone of interest for the laser to operate on, and then pump it out rapidly, so you replace continuously. That means that it was done in a continuous way, and the significance of using the visible region is that electronic excitation of the reaction products is what’s implied. What we did that was different was to look in the infra-red spectral region. Where the significance of the spectral region is that one’s looking at vibrational excitation instead of electronic excitation. The other difference between our work and this more conventional approach was that we decided we had to look in a short time scale.
Well, now what —
— a micro-second time scale.
How did these things come about? How was it that you decided to look in the infra-red? What was the progress of that?
The reason that we were out in the infra-red at all, of course, was because of my preoccupation with determining molecular structure, and if you get the molecules more complicated than triatomics, electronic spectrum is very very difficult to interpret. That’s why physicists left the field of spectroscopy, essentially; after they got to diatomics and a few triatomics it became so difficult to interpret electronic spectra that they more or less left the field. Chemists, being interested in molecular structure, picked it up and did the best they could. The most rewarding field, spectral region, was the infra-red, with reference to determining molecular structure, and that was the spectral region of interest to us.
Now, from what you say, chemists would generally, you’d think, be looking in the infra-red for lasers.
Lots and lots of people were doing infra-red spectroscopy, you see, but not with the interest in the possibility of working with emission. Now, I should remark, you will be aware, that the infra-red spectral region is what’s commonly called the heat end of the spectrum, you see, and so everything emits infra-red light just by reason of thermal excitation, and what one is trying to do here is to get more light than this background light. We decided that one of the important problems to deal with was the relaxation phenomenon, that the reaction may indeed produce population inversions, but collisional redistribution of the energy very quickly redistributes that. So what we felt was one of the important dimensions of the problem was to produce this population inversion and detect it in a time short compared to relaxation processes.
Now, when are these thoughts going on? Is this already back in ‘62, in the very beginning?
Yes, when we were first getting any kinds of infra-red spectra by rapid scan spectrometer, and began to think more deliberately about, what can we expect to see? Then we began to realize that if our molecules were born hot, vibrationally excited, then we will, instead of seeing an absorption spectrum, see an emission spectrum. Now, back to setting this in the context. You will realize that in this volume, I think it’s a correct statement that essentially the only person who drew attention to the possibility of vibrational population inversions was Polanyi.
I see.
And he was doing chemiluminescence in the conventional flow system, in which he produced hydrogen atoms, for instance, and brought it together with fluorine, in a zone in which he tried to collect enough light to see the emission. He, of course, discovered and more or less pointed the way that if you want this to mean anything, you have to do this on a time scale short compared to relaxation times. And the way he finally decided to cope with this was to have the whole chemistry take place in a cryogenic environment. So as fast as the products were formed, they were frozen in the walls.
I should tell the tape that the volume Professor Pimentel referred to was THE APPLIED OPTICS REPORT of the La Jolla meeting in ‘64.
Now, one more detail about what caused us to move into this particular spectral region with this technique, and perhaps caused other people who were more deliberate and more acquainted with laser behavior not to do these experiments. One of the popular approaches at the time was to try to calculate the gain that you would have for a given population inversion. This involves a transition probability, and of course, the higher the transition probability, the higher will be the gain of a laser for a given population inversion. Infra-red transitions have traditionally, everybody knows, extremely low transition probabilities, so this looked like an extremely unfavorable system with which to work.
Now, you knew that. What did you say to each other about it?
Well, I was used to using infra-red transition probabilities and getting absorption spectra. The way I tried to cope with this, you see, was to make the assumption, we will have to have extremely efficient light collection because the light will be at such a low level, but under any circumstances, we were not intimidated by the transition probabilities because we were used to dealing with them. Now, one of the factors that was called to Kasper’s attention at La Jolla by some of the physicists, that they cited as evidence that it was probably not laser action but some artifact, was that the first laser, that photo dissociation iodine laser, has an extremely low transition probability, and they said, “No way is the gain going to be high enough.” What we, in our preoccupation with other aspects of the problem, stumbled on, if you like, was that by doing our experiments in pulsed waves we got extremely high population inversions because we didn’t have relaxation to worry about. There wasn’t time for that. And if the chemical reaction produced a high population inversion, that more than made up for the low transition probability.
This is all a very interesting story about what happens when you come from a different perspective.
Exactly.
Now, the pulsed aspect, as opposed to the generally continuous aspect of other people’s research, had to do with the fact that you were concentrating on transients and interested in the rapid scan?
Right. We regarded the vibrationally excited molecule as a transient species, and felt that this is just the way to look for such transient behavior as the production through a chemical reaction of the population inversion, and to demonstrate that population inversion before there was time for it to relax. Plainly this is extremely important from the chemistry point of view, because one of the areas about which we knew very little, up until this time, was how the energy in exothermic reactions is distributed among all the possible degrees of freedom. And having no other assumption at our disposal, it was always more or less expected that it would presumably go in a chaotic way, that is to say, in a statistical way among all the possible degrees of freedom: and that had the implication, which I as a freshman student learned, an exothermic reaction takes place, the fragments fly apart with very high velocity, because translational degrees of freedom have the greatest statistical probability, as they’re getting all the energy. And then of course the other degrees of freedom could carry some energy too. Well, it turns out, this is nowhere near correct. In this reaction, the two reactions that represented the first real chemical lasers, H plus CL2 and F plus H2, amounts of energy going into vibration are of the order of 50 to 60 percent of the total energy. That is by far the favored degree of freedom. That’s where the population inversion comes from.
Did you have any inkling of this before you actually started?
The only inkling we had was, we had a direct inkling in the chemiluminescent work of Polanyi. He was seeing emission in the infra-red and deducing that its intensity and distribution among the transitions he observed must imply a heavy favoritism for vibrational degrees of freedom. We had another line of argument that we used, and to some extent still use, that accounts for how we happened to discover the iodine laser. We made the relatively familiar assumption that as a chemical reaction takes place, there will be what we call Franck-Condon like influence; the meaning of that expression, Franck and Condon, came up because of the idea that in an electronic excitation, the molecule doesn’t have time to change shape, and what you see in the excitation spectrum will be associated with the attempt of the molecule to absorb light while maintaining shape. Even if the upper state has an electronically different equilibrium configuration. Now, that is not obviously so in a chemical reaction, because the time scale is different, for a chemical reaction, compared to absorption of light. Still, it was a good zero order assumption. To be specific, for instance, in CF3I, methane like molecule, the tetrahedral structure, the CF3 is in an umbrella or pyramidal shape. We were breaking the iodine away, leaving the CF3 alone, and we expected that that molecule would become plainer, so the assumption we made was that that molecule (by this Franck-Condon argument), has to be excited very vigorously in this degree of freedom — the umbrella closing degree of freedom — in order to be born from a state which is pyramidal, because its unexcited state would be plainer. So we assumed it would be born with lots of energy in that umbrella motion. That is what I call a Franck-Condon-like guiding principle. So that was a reason for expecting vibrational population inversion, you see. Now, the amusing and ironic thing about this is that we were also looking in absorption CF3, and what we discovered as we managed to detect it and get some of its spectral features and other people, other kinds of experiments, detected the same thing – We discovered to our surprise that CF3 is not plainer. It’s pyramidal. So the reason we went looking for it was swept out from under the experiment. But in the meantime, we were seeing light, and it turned out to be the iodine. So I think that is in that SCIENTIFIC AMERICAN.
That’s in the SCIENTIFIC AMERICAN, though not in that detail.
So we did have a working hypothesis that led us to think that CF3 I was a good likelihood to produce a vibrationally excited molecule. It turned out, to everyone’s surprise, that the guiding principle was misapplied here, because the molecule did not change shape. But to everyone’s astonishment, we got 100 percent population inversion in the iodine, from CF3I.
Which in a way sounds a little disappointing. You weren’t really getting chemical information.
No. As far as we were concerned, that was a gravy outcome. It was one of these fortuitous outcomes that everybody was interested in, but it was not what we originally intended to try to learn. But the other side of the coin is that just by its discovery, it clarified for us what we were actually seeing in our experiment; and that opened up the hydrogen chlorine explosion. So in a way, it was a direct step to what we really wanted.
Is this a good time to put in some of the material on the slides which illustrates how you tried to maximize the effect you were going to get?
That carries me back to talking first about the rapid scan spectrometer — which was evolving at the same time that we were trying to get geared up to do that emission work. In our first applications, we tried to get onto this 10 nanosecond time scale by rotating the grating as rapidly as possible. We thought that something around 10,000 rpm was about as fast as we could rotate a large grating (large meaning about 5 or 6 centimeters in diameter) and that more or less limited our time scale. But it turned out that that made us a reasonable rate of spectral scan, and it got into the time scale where we were pressing the time response of the detector; so that was OK. Our first spectra were merely of atmospheric gases under normal conditions, like carbon dioxide or ammonia or water in the gas phase, and we got some very nice spectra. As soon as we started to try to produce a transient, then we ran into two big problems. One was that we were intending to produce a transient with an intense flash of light, so we assembled great big capacitor bank, charged it up to 10 to 20 thousand volts, and then discharged this more or less conventional flash spectroscopy technique through a discharge tube, in times of the order initially 20 or 30 microseconds. This produced an enormous amount of electronic noise, that essentially caused our detector to be flooded for a millisecond or so. Ken Herr used to say that the detector was off the air so long that he could stand there tapping his foot, waiting for it to come on, which is what he regarded as a millisecond. So we had a big problem with getting rid of that noise. That was ultimately solved by moving the flash bank into the next room, cutting a hole through the wall to bring the optical path through on the expectation that it was both an acoustic and electronic aspect to this noise and the metal lathe in the wall and the rigidity in the wall would eliminate both of these. There’s a little amusing anecdote, or to us an amusing anecdote, about this. When we decided that maybe this was going to be the only way that we would solve these noise problems, we came up with the thought that if only we had the capacitor bank in the next room, then that would permit us to escape the noise problem. But then the problem was, how to bring the optical path the distance to the spectrometer. We thought about bringing it by mirrors, through the door, and this was obviously impractical. This was one evening, so we decided, we’ve just got to have a hole in the wall, and I pointed out that this had to be done by Buildings and Ground; and Buildings and Ground traditionally took about two or three months to do that. And as I left I said, “It really is too bad, isn’t it, that it’s going to take so long to get this hole in the wall,” and I went home, and the next morning there was a hole in the wall. At just the right place.
And nobody ever found out how? When this was going on, you worked with Herr and you worked with Kasper, or was it really that everybody’s working together to some extent?
Actually, the earliest parts were before Kasper was actually working on this problem, and I’m talking now about working between Ken Herr and myself. In any event, to continue on this, that did solve our noise problem, and then the next problem that I referred to, two big problems, after getting rid of the noise, then the question was raising our sensitivity to the point that we could see these small concentrations of transients. And for that purpose, we decided we had to press every dimension of sensitivity. One of these was to try to have a long path length, and so we went multiple refraction cells, first cell was essentially a meter optical path, by reflecting back and forth, and we finally ended up with a 40 meter optical path by reason of going to physically larger cells. The other dimension that we tried to improve upon was the intensity of our light source, doing absorption spectroscopy, just to raise the signal to noise and permit us to narrow our slits so we’d get better resolution, and that caused us to move from conventional infra-red sources, which at this time were Nernst glowers or Globars, heated to around 1500 degrees Kelvin, to what we discovered with a little bit of research was the most intense black body source we could find, this being a drive-in movie carbon light. Fortuitously for us drive-in movies at that very time were beginning to move from carbon arc technology to huge xenon lamps which gave nice white 1ight but didn’t have the problem of carbon monoxide generation that the carbon lamp generates. And consequently we found that there was an active market in second hand carbon arcs, large carbon arcs, and we got one I think for about $300 from a drive-in movie, second hand. This required 10 or 15 amperes of current, and of course generated a large amount of carbon monoxide. We had to arrange ventilation. But it gave us a black body temperature up around 3000 degrees Kelvin, very very much more light, and it turned out to be quite a significant help. Somewhere along in here was where Jerry began building what we called a large White cell, a multiple reflection cell designed in the initial use by this physicist White who was used to Raman sources. Speaking about Raman spectroscopy, that the very low intensity of Raman emission required the development of highly efficient collecting optics, and that’s what a White cell was. Ours was about a meter long, and I think in its final design, it gave us something like a 40 meter effective path line, and we put our flash tube right inside the Raman cell, and it turned out then to have highly efficient collection optics, for what we expected would be a weakly emitting sample, weak because of these low transition probabilities that I talked about. That then represented the setting in which the first chemical laser emission was observed. We were actually rotating the grating, trying to get a complete spectrum during a flash to see whether we had population inversion. One of the systems we tried was a hydrogen and chlorine explosion. We were attracted to explosions because in our absorption work, we were finding out that the problem was, how much transient molecule molecular species we could produce at a given time, and an explosion helps you. For every photon that starts a chain, the chain produces several additional fragments. And so it had a multiplying effect, so the hydrogen chlorine explosion looked like a good system, particularly in the light of Polanyi’s work. So that caused us to work with the hydrogen chlorine explosion, and one of the most confusing aspects of these months immediately preceding the recognition of the chemical laser action was that we were getting extremely bright flashes of light at spectral regions that we could not associate with what we were looking for. Whether it was CF3 or HCL. We got extremely blinding flashes of light, as far as the detector was concerned, but not at the right angle of the prism. What finally developed was that Jerry recognized that this might be reflection off the corners, the edges of the grating, flashing light onto the detector without dispersion, somehow or other, and so some spurious collection of light on the detector, an extremely sensitive detector, and he stopped the grating rotation. And when he stopped the grating rotation, then he discovered that the light was not at all in the spectral region we thought it was, so it couldn’t be spurious HCL light coming off of the edge of a grating, it was actually in a different spectral region. And he finally was able to identify it as new infra-red emission, and that permitted the recognition of this iodine atom transition, which we hadn’t been looking for at all.
I’m assuming this is all happening during 1964?
That’s right. And that’s more or less very crudely documented in my log book, where we’re talking about the troubles and we’re going down this long list of chemical compounds that might be used to produce vibrationally excited population inversions, and I remember looking at that list, maybe two-thirds of them were tried, the ones that were circled, and so it was a very frustrating time. In any event, once he stopped the mirror, then we realized that we can’t look for this laser emission the way we do our absorption spectroscopy. We have to have the grating sitting still, looking at one frequency at a time, and find where the spectral region is where the light’s coming out. And that was the sense in which the recognition of the iodine laser emission then permitted quick unraveling of the similar spurious light we had seen from HCl. Then it became possible to recognize that what was happening was that if it lased, the laser emission was so intense that light would get to the detector no matter what the grating position was. But it would lase when it was ready to. It didn’t ask the grating if it was at the right angle. And so we ultimately found that the grating angle had nothing to do with when we saw laser action, it was when the population inversion took place that had to do with this, going from a very, very weak source to a very, very bright source. Incidentally, you may be interested, as an aside, that many years later, in 1968 I guess, I tried to get into the science astronaut corps. This was the second time that they decided that they would put scientists into the astronaut program, and this time they really meant it. They wanted to get scientists, not jet pilots who could be called scientists. They looked for just scientists, and then they would have to make them jet pilots later. So I went through this procedure, my application, and finally was not accepted because of a scar on one of my retina. They had a battery of ophthalmologists, the very best that they could find in the country, look at it and try to decide what it was. And it has only been very recently that with the analysis of Karl Kumpa who was the person who discovered the HF laser, went back to Germany and became “Mr. Chemical Laser” in Europe, and he is the one who build up the iodine laser to potential application for fusion application, and so together at my dinner table about a year and a half ago, we started talking about how much power you needed to get eye damage, and we were able to decide that we did have enough to do eye damage back in ‘64, and I have laser burns on my retina from focusing the thing. See, Jerry and I would get down and focus it. Infra-red light, you can’t see, and you wouldn’t know when it was coming out. The eye people had never seen a laser burn, and didn’t recognize it.
Curious.
This is all my own analysis, understand. I haven’t had any person look at it recently and say, “Oh yes, it does look like a laser burn.” But you know, it’s such an obvious thing. In any event, so I related how this spurious light we were seeing mainly from the iodine laser and from the HCl system was finally recognized as being due to laser emission, and we went then from moving grating to stationary grating. Then, after the La Jolla meeting, we went into conventional optics, and from then on used conventional optics.
At what point was this little jocular experiment done by Kasper, of putting an index card instead of a mirror?
Well, it had to be within that time period between our first recognition of the iodine laser, which I think was identified early in September, and when we mailed off our manuscript, which you saw the covering letter here, it was in October. It was between those times, October 27, and after the meeting at La Jolla. It’s one of the things I told Jerry after we had proved that we could get this light emission in the conventional laser optics, that one of the most important things to be able to say something about is the gain, and that got us onto the gain, and it was one of the experiments that he performed. He came back to my office and told me that he had now tried to measure the gain and had a gain which he said, if I recall correctly, he said was above 30 decibels. I was unfamiliar with decibels, so I said, “I have to calculate what that means,…” In any event, I calculated what this amounted to in terms I could understand, that if one photon entered one end and we had a meter path length how many photons would come out the other end? And the answer was so enormous that I said, “This is ridiculous, it couldn’t possibly be this high. It would lase without mirrors, if the gain were that high. You wouldn’t have to go back and forth.” Jerry was both very smart and very confident, and he said he was sure he was right. I said, “You’ll have to prove it to me.” He came in the next day and showed me a record of laser emission that he had obtained, putting an index card in front of one of the mirrors, and of course thereby immediately confirmed in a qualitative sense that the gain was as high as he contended. I say, I recollect him initially saying it was above 30 decibels. We actually said in our first publication, our measurement was 106 decibels. As I remarked, this was an extremely fortuitous thing that we ran into this population inversion. It was completely unanticipated. There is today no real strong theoretical understanding of why this particular molecule gives that population inversion. We finally convinced ourselves that it was 100 percent. We tried, in addition to that, methyl iodide, the normal methane iodine compound — instead of having CF3I, it’s CH3I. It gives laser emission but only 60 or 70 percent population inversion. So we just stumbled on the perfect molecule, is what it amounted to.
Now, you gave me something I’d like to know about. You got the iodine and the HCl and what kind of research program did you now begin to?
The first thing that we wanted to do was show that we could get kinetic information (information about chemical reactions), from the hydrogen and chlorine system. As far as the iodine system is concerned, we went into a program in which we tried to find what other iodine compounds would give iodine emission.
What was the motivation behind doing that?
Well, because it was such an unanticipated outcome. That photolysis and bond rupture would preferentially give you the electronically excited species. We were groping around for some basis for understanding that. I should add, that we were interested in the iodine laser as a means of adding to everyone’s understanding of why a low transition probability transition would give such intense laser emission. We too could write down the gain equation and see that the gain factor was working against us, insofar as the transition probability was concerned. And the other aspects of that reaction that turn out to be just perfect for laser emission are that the fate of iodine atoms, one of the predominant fates of iodine atoms, is recombination, going back to I2, I plus I. But that is a two body reaction. Then when two bodies come together and form a biatomic, the molecule is born energetically excited, so it can fall apart again. And so that’s an extremely slow reaction as a bimolecular reaction. The only way you can get two atoms to recombine is to have a coincidental collision with a third body, three body collision, to take away some of the energy so now it’s trapped. That’s improbable. So one of the loss mechanisms by which this highly reactive species would be lost was extremely slow. And that’s a kinetic question too. In any event, the hydrogen chlorine looked like, system looked like the one where we could most immediately advance in showing we could learn something about chemistry here, and so I had the next graduate student, a student named Paul Corneil, go to the D2Cl2 reaction, and HD Cl2 with the idea that we would look at isotope effects on the reaction rates, and see how the energy distribution went with deuterium. It turned out that, and looking backward one can see, that the hydrogen chlorine explosion is not a very favorable system. You lose control of an explosion, in the first instance, and it has the additional problem that the gain is not very high because it’s a chain reaction in which one chain gives you a population inversion, one reaction in the chain, and the second reaction gives the same product but cold. See, the second reaction in the chain is an H plus Cl2, pardon me, Cl plus H2, in which you break the H-H bond and form the Cl bond, and that’s a thermal neutral reaction, so it forms HCl, the product we want to emit light, but completely cold.
Now, you could predict this in advance, or this is something that came out of the —
— no, no, no, that was predictable in advance, you see, and the reason we ignored that as a big problem was because Polanyi was seeing emission from the system, and so it indicated whatever is going on, that isn’t killing it, but in any event, in retrospect it was a bad feature, and added to this was that when we went to deuterium, all the level spacings were smaller, because the deuterium mass effect, isotope effect on the spacings caused the —
Now, you knew that was going to happen too?
Yes, but we weren’t really sensitive to how much it was going to hurt the laser gain.
I see.
What it amounts to is smaller spacings, and hurts you in two ways. It dilutes the population of any single level because there are more levels to share, and it speeds up relaxation. So both of these things made the deuterium chlorine system a very hard system to work on. But Paul did manage to get on top of it, and we got some interesting results out of it, but it was slow going. And in the meantime, not much was happening, because we were the ones who were doing this and we were having a tough time, so our second publication was on D2Cl2. At that time, I had Karl Kompa come; and I decided that the next system to try was an OH laser, using ozone. When Karl arrived, he came with magnificent references, outstanding references, and he lived up to every bit of them. When he arrived, he discovered that he had one hand that was badly disfigured, and it turned out that he had had an ozone explosion and wrecked his hand. So when I saw him I said, “Gee, Karl, you probably don’t want to work with ozone, maybe you should find another system.” Then he was extremely able and extremely confident and he said, “No. I know how to control ozone now,” and he went ahead and worked with it for six months, trying to get an OH laser. Again, in retrospect, that turned out to be a way in which you can generate a laser, but we were unsuccessful. We were not able to find conditions in which we could get that OH laser to go, and so, we became frustrated with it and decided to go for an HF laser.
Now, what put you on the trail of the HF? What were the reasons that you thought HF would be —?
Well, one of the obvious reasons was that it was a system which was sufficiently similar to the hydrogen chlorine system that you could more or less predict that it was going to work, if you could control it. Chlorine is very easy, from the chemist’s point of view. Chlorine is a very easy chemical to work with. Fluorine is very difficult chemical to work with, and we had long thought that the logical system, the H2 plus fluorine, F2, but decided that’s really too tough a system to work with. So what Karl did was look for photochemical sources of fluorine atoms, and he was and is a good chemist and he looked without any inhibition. And of all things, came up with a possible fluorine atom source — uranium hexafluoride and uranium hexafluoride is sort of a weird chemical to think of in the first place. You don’t normally think of a molecule with a heavy atom as being gaseous, but of course it is gaseous. It’s the basis of one of our gas diffusion isotope separation techniques.
Yes, I was wondering if you had been in touch with that chemical in the war, when you did —?
— I knew that it was being used, but —
— you weren’t personally —
— I wasn’t personally involved with it. I had no experience with it. But I knew it was a gas, and when Karl found that it had the right photochemical properties, that was the logical thing from that point of view.
Now, was this a world in which lots of people were looking for hydrogen fluoride?
No. See, we were still more or less alone in the infra-red, so (crosstalk)
— I see, let me ask about Deutsch who was working with arcs on that —
Deutsch was putting discharges, Tom Deutsch was putting discharges through gases, and seeing rotational laser emission, and we tended to, as I indicated to you already, we tended to regard this kind of an experiment as heavy-fisted type of experiment in which it would be very difficult to interpret. So we never did — never ever — I don’t think I’ve ever looked for laser emission through discharge. But in any event, of course, Polanyi was doing a variety of things, and I’m not sure where he started working on the F plus H2 reaction, but he may have been working.
Is this a situation where you were in contact by letters a lot or by phone a lot or just seeing each other at meetings or what?
I just watched for his literature. Because I knew that what he was doing was of great interest to me, but no, I really didn’t see him much at meetings at that time. In any event, we decided that the way to handle fluorine was to get away from P2 and to get away from the explosion. And that led us to photochemical sources. Karl discovered that the photolysis of UF6 has a high quantum yield, and in the spectral region was suitable to our flashes, when he tried it, and he got laser emission relatively quickly. We were pretty good at building optical cavities by now, and detection was in our pocket and so on. So he brought that to fruition rather quickly. Now things began to blossom, because now we had a system to study, and it was an intense system and we had control of it, in contrast to the explosion, because now the number of fluorine atoms was determined by our flash, and we knew a lot now about relaxation and so on. So I suggested to —
— a real turning point in your —
— yes, right. One of our students, one of my students, Jack Barker to, that perhaps fluorine atoms would abstract hydrogen from methane, and Jack tried UF6 with Karl and found that EF atom abstracted hydrogen from methane and gave HF and gave laser emission. Then, of course, the question was, what else works? And he went down the shelf and everything worked. Somewhere along in that year I got this very brilliant student Mike Barry, and talking to another friend of mine Eric Whittle from Cardiff, this fellow being the first post-doc I ever had, now a prof at Cardiff, he suggested maybe we could get a reaction between CH3 and CF3 to form ethane, trifluoral ethane, with all the bond energy of that carbon—carbon bond, which would make that molecule extremely hot, and it would have enough energy, he felt, to eliminate HF. Maybe we’d get a laser from that. Mike Barry tried that. It worked just beautifully.
I really want to interrupt for just a minute. How did the rest of the world react to the HF laser? I put in a question here, whether the OSR, the Air Force people were particularly interested, because I know that the HF laser became a prime military laser and I’m wondering how they reacted or whether they offered any suggestions or whether any laser manufacturing firms or what your colleagues said?
There wasn’t any sense on our part of recognition of potential applicability in those eaz4 days with the HF laser, though plainly it was a much easier laser to operate than the HCl laser. The Air Force was just very encouraging because they could see that this was new and fruitful kinds of research, and were quite happy to see us go ahead and move ahead, and I honestly cannot tell you when they began thinking in terms of possible applications for weaponry.
I notice there’s a group at Aerospace, W.F. Gross, T.A. Jacobs and Cohen, were they?
They were sort of early seventies, weren’t they, I believe?
These were in the May ‘69 second conference, so they were — ‘68?
McDonald Douglas, wasn’t that the one?
Wasn’t it St. Louis?
St. Louis, yes. McDonald Douglas was another place.
Won’t you on the program? … I see … didn’t know about that.
That we had — and again, I think, Air Force money. And they had quite a center going there. Well, that dates it, then, somewhere along in ‘68, I guess, or so, people must have started realizing maybe that there was a potential for weaponry and money may have started moving into that. As you can tell by my answers, I have never ever had any money to work on a classified chemical laser project. They’ve always supported me without any classification; and everything we’ve done has always been published; and I’ve never been privy to any secret information about lasers. Though of course they’re very pleased when I go to meetings where various laboratory people are present to exchange information with people. I always exchange. I exchange everything I know and they tell me what they can tell me.
I really was just trying to get at whether there was an interaction with these groups which was of any interest for the way in which you were proceeding trying to see whether there was any —
No, I didn’t — I interacted with these people at meetings, just as I did with all other scientists, just that we were all trying to understand these things, but plainly they were being supported because there was thought in somebody’s mind of the possibility that there would be applications. But we never went into applications.
Well, one of the things I wanted to know was, do you think it’s worth talking about this second conference on chemical and molecular lasers a little bit? Was that an interesting thing? You were on the program committee. There was a meeting. I wonder if you have any comments that would indicate, say, in 1969 how things looked, or during the period you were organizing this? Any impressions of the conference?
I can only say that it was, if I’m thinking of the right meeting, it was — although I must not be thinking of the right meeting. It seems to me there was a later one in St. Louis, about three years later, wasn’t there?
I don’t know.
Because the meeting I’m remembering was one that was sort of an Old Home Week for me, because it brought together quite a number of my former students, all of whom had worked on lasers, and I have lots of nice pictures of all these people that I’d been working with so much that came to this conference. But it wasn’t in ‘69.
Polanyi spoke on general chemical rules for lasing, and then the Aerospace people spoke on HF lasers that are initiated by detonations, and you gave four papers, then there was Chen and Moore from Berkeley.
Can you be more explicit about, who were my co-authors on this?
OK, you gave one with Tablas, one with Barry, one with Parker, and one by yourself on the chemical uses of lasers.
Now, that one with Tablas, that was our first proposal of the tandem technique.
Yes, that’s —
Tablas actually worked on it for about a year. All we could do there was say that we think it’s a very good idea but we’re having trouble with it. And the problem turned out to be one that you can only feel very sorry about in retrospect. We had fort let’s say what at the time seemed good and sufficient reason, decided to keep our optical paths all at a certain elevation above the floor, so that they could share a detector. Another system was on a table that was already in use that defined the elevation above the floor of the detector. And he was on a different bench, and it turned out that that required that he have rather high posts to his optical mirror. We later began to realize that there was just too much flexibility in his mirror mounts, and we couldn’t keep those cavities in focus. When we dropped them down closer to what was our mirror table, we didn’t have optical tables, we had steel I beams and mount our optics on those to give rigidity. When we got it down closer to the table, the system began to work. But it was entirely that mechanical problem of having too much vibrations brought on by the acoustic … (off tape)
All right, we were talking about these acoustic shocks that were causing the mounts to vibrate, but —
Yes, home made mounts, you might say, too.
Perhaps we should throw in here a little bit where you got this idea for the tandem?
Well, the idea was more or less born out of what you might call frustration and necessity. We were trying to learn about population inversions, and nation populations, primitive populations that you get in a chemical reaction, through laser emission. Needless to say, for every time you flash that flash bank, and cause for a chemical reaction to occur, for every time that that gave laser emission, there might be 20 times it didn’t, and we’d learn nothing. And the question was, isn’t there some way we can carry out these experiments so that, every experiment we get some information, even if we don’t have a population inversion? So we came up with this thought that, we’re studying HF lasers a lot of the time, why don’t we have one HF laser to give us laser emission every flash, and then put another system in the same cavity and perturb it, and that will let us see when we’re getting close to laser emission, see.
Now, who is “we” here, who was in on all this?
Well, it was mainly Tablas, Koompa and Parker and then Barry arrived and Mario Malina, and so, it was during that period. And so I asked Tablas to try this in cavity, what we called tandem technique. Malina, who was an interesting little personal situation there — Tablas is from Madrid, Malina turned out to be from Mexico City, so we had these two Spanish speaking people working together and they got on very beautifully, and Mario was able to pick up, and then Tablas had to go back to Madrid, and carry this technique forward and demonstrated it magnificently; and in essence developed what we call the zero gain technique. The zero gain technique, you see, was associated with varying the temperature of what we called the slave, the one that was perturbing the driver. The driver we would do in as reproducible fashion every experiment, on a single transition, just giving us a reproducible laser emission. Then we’d perturb it with the slave, varying its temperature, and hunt for that temperature at which we lost perturbation. Lower temperature, we would raise the gain, higher temperatures, we’d lower the gain. And as we passed through the zero gain situation, then we had a temperature control on the chemical reaction. This is the point at which the population inversion has disappeared. And that gave us a quantitative measurement. So that was the whole principle of this technique, and allowed us to measure both positive and negative gains, so we were getting a complete picture now of the population, not just, when the population was sufficient to give laser emission.
Were there any surprises in the course of this, or it just got developed in a very smooth way?
Well, it led to some very interesting thoughts and additional reactions. One of them, for instance, was, -in the hands of one of my students, Bob Coomb. We decided that it would be interesting not only to vary the temperature, but to see if we couldn’t with that zero gain technique learn something about the rotational degrees of freedom — whether that was a factor in the energy distribution of the products. That happened to be a question we could address because hydrogen has ortho and para forms, and one form only has even numbered rotational degrees of freedom occupied, and the other one has only odd, and by using pure ortho or para, you could have a different intrinsic rotational distribution, and we did find some effects associated with that. That was an unanticipated kind of experiment that we carried on.
Well, it’s now about 17 minutes to 12 … I’ve got about three things I’d like to briefly ask you, so, I’ll end the tape right here.