Isaac D. Abella

Abella:

I took my qualifying examination at Columbia University in December 1958, and was ready for a thesis. I was especially interested in the maser, and I went to Townes and asked him to be my thesis adviser. Townes said he was looking for a student, but not for microwave masers. He was interested by the fact that I had been at Toronto, where there is a focus on optical spectroscopy. It was for this reason that I had originally wanted to do microwave masers. After 3 years of optics, physical optics, geometrical optics, waves, I was tired of optics. Townes took only 2 weeks to look up my record and decide to take me on. He gave me a pre-print, in mimeographed form, of the Schawlow-Townes article. I had not previously been a member of his group I’d been a teaching assistant.

I had come to Columbia in the fall of 1957, and I had had some contact with Townes; I had taken quantum mechanics from him and sat in on one of his seminars. I also knew students at the Columbia Radiation Laboratory (CRL), but I had not worked with Townes. Herman Cummins was the senior student on the infra-red potassium laser project. Townes had a system whereby senior students would break in more junior students. Townes would come around to talk to us, and he was very open to our going in and consulting him. I started working with Cummins in February 1959, but I was still taking courses that semester. I started full-time work with him in summer, 1959. Cummins was having problems. He had 4 linear lamps surround the linear discharge tube, but they were weak. We could see that we weren’t getting much energy into the potassium. We were using flat mirrors. We made them by evaporating gold into plates; this gave us good reflection in the infra-red. I spent most of my time [in early fall 1959] trying to get fringes with a Hilger interferometer. But you would heat the tube and the fringes would disappear.

The potassium would permeate the glass and darken it, which was one of the unexpected problems. That caused us to do research into alkali-resistant glasses, and we chose sapphire because it is stable and resistant to potassium, but then bonding the sapphire to the glass or quartz tube probed to be a problem. We had to distill the alkali metals, and sometimes during the distillation everything would blow up because potassium is so reactive. We measured the fluorescence of the potassium and it was much worse than what our calculations indicated. We later found that, when distilled, a little bit of hydrogen entered the potassium, which quenched the state we were working with. It was Townes’ idea to use an elliptical cylinder and place our lamp at one focus and the tube at the other. We took a thin cylinder of aluminum and used an elliptical clamp to distort it into a cylinder. When Townes proposed this, Cummins jumped at it. I was suspicious, however. If you take an elliptical cylinder and cut a non-right section through it, the foci are not the same for rays at an angle. I thought that before spending the money machining the cylinder, we should find out if Townes’ idea was correct.

I worked out a long calculation in detail, showing that the skew rays would encounter ellipses of higher eccentricity. The foci would move out and the idea wouldn’t work. I brought this in to Townes who said, the calculations looked nice, but must be wrong because there was a physical argument showing the scheme would work. I couldn’t imagine how a physical argument could show it correct if a mathematical argument showed it to be wrong, but Townes drew a vector representation of the ray and resolved it into horizontal and vertical components. He showed that the horizontal component will arrive at the focus and the vertical one will merely be displaced vertically. I went home and thought about it and realized that I had, in fact, made a subtle mistake in my formulation of the mathematics.

Visitors from all over the world came to the lab. Some of them saw our elliptical lamp, and one of them published an article about it in the IRE journal a few months later. We were incensed at this pirating and publishing of our results, and Cummins wrote a letter to the editor about it. One day during the summer of 1959, I noticed a sign on the bulletin board announcing a meeting of all Townes’ students on the 8th floor. This was unusual. Townes never met all of us simultaneously; he’d meet separately with those in particular areas. (Incidentally, some of these meetings were structured around articles he’d been given to referee. He’d ask us to report on the papers and then meet with us to discuss them. He himself would make the final decision on the paper. Once he gave me a Javan paper for which I had to look up old German papers by Ladenburg and others. I learned a lot of physics that way.)

At the time Townes had students working on free radicals, solid state physics, parametric resonance, microwave spectroscopy, masers, radio astronomy... Although he met us in subgroups, there were enough of us so that we would have our own monthly seminars, without him. The meeting itself devastated me. He said he was going to Washington. Those well enough along could finish off their theses. The newer students might want to get another sponsor. I was helping Cummins at the time and hadn’t really been given a problem of my own. I went to him afterwards and asked: “What do you advise me to do?” Townes replied: “Stay with this project. I am very interested in it, and I will spend more time with you than with students working on other projects.” Later, he told Cummins and me privately that he’d asked for a visiting person to come and help us. He had read Oliver Heaven’s book on optical properties of thin films, and had arranged to have him come.

This was just before the Shawanga Lodge Conference (in Catskills) in the fall of 1959. Townes asked his students to be conference staff. We showed slides. When someone made a comment, we would go up to him/her and ask to have it written on a slip of paper; we would then go to the speaker and ask to have the reply written. These appeared in the book, Quantum Electronics, (Columbia Univ. Press, 1960). In the fall of 1959, I began to look into other pumping sources. The potassium lamp was inherently weak and I was very frustrated. For example, the 3888 line in helium discharges overlapped an absorption line in cesium. This was good because helium lamps were easy to operate. I also found, by looking at tables of alkali lines, that a violet line of potassium was close to one of mercury. A mercury lamp could pour in thousands of watts.

I read a Philips Laboratory catalog on mercury lamps and found that pressure broadening could give the needed wavelength. I bought a GE mercury lamp, pumped the potassium, and saw some weakening in the absorption spectrum. I went out to Murray Hill to get a better reading on Schawlow’s spectrograph. Schawlow had one of the first Jarrell-Ash spectrographs. (He had had them paint it the Toronto Univ. colors, royal blue and white.) We saw the absorption dip. Right away, he set up a photomultiplier cell to look for fluorescence, but the trace impurities were quenching the fluorescence. In the spring of 1960, Schawlow came as visiting professor. We’d gotten a Jarrell-Ash spectrometer of our own by then, not a vacuum machine like Schawlow’s, but another type. He taught me how to focus it, how to get good spectra and see hyperfine structures. We were working with cesium now. We had the same concerns as before, mirrors, distillations, and so forth, but now with cesium.

The physics were simple, but the technology was messy. I came into Schawlow’s office and said: “I remember you talked about ruby at Shawanga Lodge. It would be so nice if we could use a technologically simple substance like that.” I had a chunk of ruby I had obtained from Bill Rose. I had shone a mercury lamp onto it and it had fluoresced beautifully. I was getting depressed with cesium and it was of paramount importance to me to get a material that was easy to work on. We were spending allour time with technical preparations and I was concerned that we would never get a degree. Schawlow convinced me that we couldn’t get enough optical power into the ruby to depopulate the ground state.

(In answer to a question about his attitude towards the idea of a Fabry-Perot resonator). As an undergraduate I had done experiments with the Fabry-Perot as a passive element and I felt uncomfortable about putting material inside one. We were happy at the time with Schawlow’s argument that a reflecting cavity remained a resonator when one removed the walls. We didn’t worry too much about this problem. By the same token, the Fox-Li results didn’t make much impact. We were concerned with mundane stuff, seals, construction of tubes and so forth. Cummins and Jacobs at TRG were friends. Through that connection, I met Gordon Gould. Gould was supportive and suggested things to try. I was invited to TRG to give a talk. I developed close relations with Gould. When Heavens arrived, he got CRL to buy an evaporating plant. I learned about thin films from him. He helped us in making mirrors, monitoring reflections from the front surfaces, and so on. He didn’t help much with the laser itself. He knew a lot of people and spent a good part of the year travelling and giving talks.

He went to China Lake, for example, where he knew the Bennett’s, and to Fort Belvoir. It was a reasonable way to spend a year abroad. He’d talk about the optical maser while we were back at the bench, working. In the spring of 1960, he was out talking. The picture he gave was that we were about to make it work; he was optimistic and we were optimistic. He spoke in this vein at the Rochester Conference on Coherence, June 1960. (1 was not there; I believe Cummins went.) As I understand it from talking to others, Heavens reported that we were on the verge of a laser. Malcolm Stitch then telephoned his management and told them the Columbia group was about to publish, and Hughes then announced the Maiman laser.

No one was at CRL at the time. Townes was away. We had a public address system and the announcement came over it that there was a call for Townes. I took it and it was Walter Sullivan of the New York Times. He had just attended the Hughes press conference on the optical maser. Sullivan often called Townes to get elucidations of new physics. I told Sullivan: “It can’t work in ruby.” Sullivan said: “They used a flash lamp.” We had never considered anything but continuous-wave pumping. I had to concede a pulsed system might work.

Cummins and I shut everything down and went across to the West End and had more beers than I care to remember. We had thought that there was no competition. Javan’s group didn’t even know at the time if they had gain. In addition, he had a reputation at Columbia as a good idea man but a poor experimentalist. A week later, I got a phone call from Townes in Washington. He told me to pay special attention to Maiman’s result. Some general in the Pentagon wanted to see a ruby maser and was preparing to go to California. Townes told him he could see one at Columbia instead. He asked me: “Can you build one?” Rose had a chunk of ruby worth perhaps $1000 and he gave me some of it. I got flash lamps from photographic supply stores. I called Schawlow to find out more about how to do this. I spent about 1 and 1/2 months on it. Bell Telephone laboratories pink ruby laser was the second and mine was third. This constituted a change in the direction of my work from cesium to ruby. Cummins was still doing cesium; I had angst on Cummins behalf. I could now see a thesis coming whereas he couldn’t. (He did wonderful work after his degree in Raman scattering and other areas.)

Cummins and I conferred on how to reconcile Maiman’s result with what we were doing. We reasoned that we could be the first with the continuous-wave laser. Cummins didn’t throw in the towel until Javan’s HeNe laser. (At Columbia, we called them optical masers until Townes secured his Nobel prize in 1964.) My paper on thermal tuning of the ruby maser was a fluke. I had expected sharp spectral lines but they were all over the map. The thermal tuning paper made the cover of the Journal of Applied Physics and I got numerous invitations to speak, at WESCON and other events.

One feature of our work was the quarterly or semiannual reviews conducted by a group from the Joint Services. They were called TAC meetings, for the Technical Advisory Committee. Irving Rowe, Paul Johnson, and other DoD monitors would come. All the students prepared 30 minute talks. At midday, we’d have a fancy lunch. The table conversation, the question period, the socializing was all important. When Townes went to MIT, I was working on the 2-photon absorption experiment. I would come up to Boston to do calculations. I met Szoke, Garmire, Chiao, and Jaseja on these visits. I served as a kind of consultant to help them build their laser program. Townes asked me to come up to MIT to finish my work, but I feared I’d lose a year, and thought the 2-photon work was enough for a degree. Novick was director of CRL at that time. I talked to him and he wrote to Townes: “We think he’s got enough. Why not invite him up as a postdoc?’ Townes didn’t bite on that one.

I spent fall 1962 working on my thesis. Novick brought me into his office to talk about the echo experiment. Sven Hartmann was brought in 3 days later. Novick hired me as his postdoc to work on the echo experiment. I collaborated with Hartmann, but I was not paid by him, nor responsible to him. I think I should have shared the OSA prize with Hartmann. He brought in interesting knowledge from resonance physics to the research, but he didn’t know about ruby and lasers; that was a contribution I made. In this period, I made all my lasers myself. Raytheon, Trion and others were selling some early commercial varieties, but this didn’t affect me at this time.

I used my original ruby laser in the cesium 2-photon experiment. The cesium cell experiment also stood me in good stead; for example, I became a skilled glass blower. The laser used with the photon echo experiment was also home built by Norm Kurnit and me.

Postscript: I can tell you need more about the photon echo experiment and its origins. There was not enough time to do so at the initial interview.