Oral-History:Nicolaas Bloembergen
About Nicolaas Bloembergen
Nicolaas Bloembergen received the IEEE Medal of Honor in 1983, "For pioneering contributions to Quantum Electronics including the invention of the three-level maser." He shared the 1981 Nobel Prize in Physics with Arthur L. Schawlow and Kai Siegbahn for their contributions to the development of laser spectroscopy. At Harvard, Bloembergen became an associate professor of applied physics in 1951, the Rumford Professor of Physics in 1974, and the Gerhard Gade University Professor Emeritus in 1990. Bloembergen also held numerous visiting appointments at international universities. This interview covers Bloembergen's education; his magnetic resonance, maser, and laser research; and his service to the Strategic Defense Initiative.
Born in the Netherlands, Nicolaas Bloembergen received his bachelor's degree in 1941 from the University of Utrecht. He passed his doctoral qualifying exam in 1943, only three weeks before German forces occupying Holland closed the university. Students who had not yet passed exams were required to either sign a "declaration of loyalty" or be transported to Germany as forced labor. Spared, Bloembergen joined the university staff. During the final winter of WWII, however, he had to go into hiding.
After the war, Holland was in ruins. Bloembergen wrote to the United States, hoping to do research for his Ph.D. thesis at an American university. In 1946, he was accepted at Harvard, where he worked with Dr. Edward M. Purcell, Dr. Robert V. Pound, and Dr. Henry C. Torrey, who had just discovered magnetic resonance in paraffin. While unknown at the time, this discovery resulted in a significant medical application: Magnetic Resonance Imaging. Bloembergen's thesis work on nuclear magnetic relaxation with Purcell would impact the later development of the laser by Maiman.
Bloembergen, whose interest remained in magnetic resonance until the early 1960s, invented continuously operating masers. He assigned non-exclusive rights to Bell Labs, and in return they took out the patent. Although he stayed out of laser development, by 1960, he had begun thinking about optics, eventually leading to his work in nonlinear optics.
Over the years, Bloembergen often interacted with both the corporate world and government as a consultant. In fact, he co-chaired a study on directed energy weapons that attracted significant attention because it addressed part of Ronald Reagan's Strategic Defense Initiative. Pushing aside political questions, this committee attempted to determine whether it was feasible to design a system of strategic defense by deploying directed energy weapons in space. In retrospect, Bloembergen contends that the final submitted report appears increasingly accurate.
About the Interview
NICOLAAS BLOEMBERGEN: An Interview Conducted by Andrew Goldstein, Center for the History of Electrical Engineering, May 15, 1995
Interview #256 for the Center for the History of Electrical Engineering, The Institute of Electrical and Electronics Engineers, Inc.
Copyright Statement
This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Staff Director of IEEE History Center.
Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, IEEE History Center, 445 Hoes Lane, Piscataway, NJ 08854 USA, or email ieee-history@ieee.org. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.
It is recommended that this oral history be cited as follows:
Nicolaas Bloembergen, an oral history conducted in 1995 by Andrew Goldstein, IEEE History Center, Piscataway, NJ, USA.
Interview
INTERVIEW: Nicolaas Bloembergen
INTERVIEWER: Andrew Goldstein
PLACE: Cambridge, Massachusetts
DATE: May 15, 1995
Education, University of Utrecht
Goldstein:
I would like to spend some time trying to reinterpret your career from the perspective of electrical engineering. I would like to focus on the sorts of equipment you’ve designed, and how the availability of equipment influenced what you’ve been able to accomplish as a scientist. Can we start at the beginning of your career, maybe your early education?
Bloembergen:
I’m a native of The Netherlands. I went to the university at Utrecht. This city is centrally located in the Netherlands. I went to the Gymnasium, which you would translate as the Latin School, in Utrecht. Then I entered the university and majored in physics. I received the equivalent of a bachelor’s degree in 1941. I passed my doctoral qualifying exam in 1943, just about three weeks before the German forces that occupied Holland closed the university. I was very lucky in that respect because as soon as you had passed that exam, you were not considered a student anymore, and all the other students who hadn’t passed such an exam faced a difficult situation. They either had to sign a so-called "declaration of loyalty," or they were transported as forced labor directly to Germany. I didn’t have that situation. I was very fortunate. Since I was not a student, I became part of the staff of the university. I had belonged to the fire brigade, so in the event of a bombing I was supposed to help put out the fires. I had a little freedom for another year. The final winter was terrible and everybody was in hiding because any man between seventeen and fifty-five could be picked up. I was in hiding until the end of the war, which was exactly fifty years ago. They just celebrated the fiftieth anniversary of VE Day.
Goldstein:
What was your education in physics like? Was it highly theoretical or more practical?
Bloembergen:
It was a combination. I considered myself an experimentalist. You had to pass oral examinations, four of them in various main topics of physics, E&M, mechanics, quantum mechanics and statistical mechanics. Those were the four required main fields. I passed those in 1943.
Ph.D. studies, Harvard
Bloembergen:
After the war, I didn’t know what to do. Holland was in ruins. Most of the rest of Europe was in ruins. I always had the idea of doing my research for my Ph.D. thesis somewhere else to get a broader outlook on things, and this was reinforced by the deplorable conditions. My older brother suggested I write to the United States. I wrote to three universities in the United States. The University of Chicago never answered. The University of California, Berkeley, wrote a letter that really caught me by surprise. At the end of July I received their reply, and they said that they would love to have me, but they couldn’t admit foreign students until the war was over. Only then did I realize that the war wasn’t over in the Pacific. That was so remote. The people who had been occupied for five years in Europe didn’t even think that the war was still going on. Two weeks later the atomic bomb was dropped on Hiroshima. Harvard wrote, “Please send some more letters of recommendation,” and I was admitted as a graduate student. I arrived on a liberty ship. There was no regular passenger line schedule. There were no airlines either at the end of 1945. I came early in 1946. I had the good fortune to look around and meet a Professor Edward M. Purcell, who had just discovered with his two associates, Professor Robert V. Pound and Professor Henry C. Torrey, nuclear magnetic resonance in condensed matter, in fact, in paraffin. They were very anxious to get a graduate student to do the work in the lab because they were still writing the famous MIT Radiation Lab Series that is twenty-five volumes, and each of them wrote one or one and a half volumes of that series. They were writing full-time till the spring of 1946. So here I was, and I became familiar with the problem. I had heard about the problem because in Holland there was a Professor C.J. Gorter at Leiden, and he had tried to do something similar, but had negative results during the war. I knew it was an important topic, and because it was so new it was very exciting.
Magnetic resonance applications
Bloembergen:
I have written on this for the Encyclopedia of Magnetic Resonance, which Wiley is going to publish this summer. It is the fiftieth anniversary of the discovery of magnetic resonance. Professor Purcell, even four years later, in 1950, said there would be essentially no practical applications for this esoteric academic field. He is very pleased, and so am I, that there is this big medical application, MRI, Magnetic Resonance Imaging. In 1950, we had no idea that it would have this application. Some people had the foresight to know that magnetic resonance would be very useful in chemistry and chemical analysis. As NMR, spectrometers are used in most chemical and biological labs, so that is a very large application, but the real big scale technology is MRI.
Goldstein:
It would be interesting if we could reconstruct the story of how this esoteric research moved over into the applied field. Do you know much about that?
Bloembergen:
Not much, because the field is so vast and there have been thousands of workers. I remained interested in magnetic resonance until about the early 1960s, when clearly with the advent of lasers my interest shifted to optics. Through my work in magnetic resonance, I invented continuously operating masers, so called three-level or multi-level masers. They are based on my knowledge of magnetic relaxation phenomena, both in nuclear resonance and in electronic paramagnetic resonance, so-called EPR. There is a very direct line between nuclear magnetic resonance and masers. There was another independent line made by Charles H. Townes through micro spectroscopy and the ammonia beam molecular resonance. The ammonia beam maser was the first maser.
Goldstein:
Were they pulsed?
Bloembergen:
No, they were not pulsed, but the beam never became a useful instrument. It isn't really very good as a frequency standard, although the hydrogen maser is. There is a direct line through the beams to the hydrogen maser, which, of course, Professor Norman F. Ramsey, a colleague at Harvard, was instrumental in. For the condensed matter maser, continuously operating stuff, it was the solid-state maser, and that same pumping is used in essentially all lasers. It’s historically remarkable that the first operating laser, or optical masers as they were then called, was the ruby laser of T. H. Maiman in 1960. They used the same crystals that we used in solid-state masers. You have a set of optical energy levels that operate the laser, and it’s the same speed of pumping at one electromagnetic frequency and getting amplification at another. There is complete parallel between the optical case and the microwave case.
Goldstein:
Can you tell me something about the direct line from the magnetic resonance work to the pumping scheme? Is there any continuity in equipment or techniques?
Bloembergen:
Yes. Professor Pound worked on it. There had already been done double resonance on two frequencies of a lithium crystal, but you had four coupling levels with unequal spacing. Pound simultaneously irradiated with two frequencies in the radio frequency region. Then there were the better known experiments of Overhauser, the famous Overhauser effect, which simultaneously used a microwave resonance to operate on the electron spins in metal, and at the same time a radio frequency to observe the nuclear spin signal. He achieved enormous enhancements. I was familiar with those with a background similar to my own, and that is why I hit on the scheme of saturating two non-adjacent energy levels. It really becomes very straightforward once you think of it. You get maser action to a transition at the intermediate level.
Goldstein:
Was it really that straightforward?
Bloembergen:
It’s just amazing if you look back in history that there is a very straight line from my thesis work on nuclear magnetic relaxation with Purcell to the development of the laser by Maiman.
Goldstein:
Was it all in the idea, or did it involve some new experimentation?
Bloembergen:
It was in the idea, and my stumbling block was one that people usually forget about, and it’s almost trivial. I always thought of magnetic resonance levels as having equal spacing. Even if they don’t, you still have a selection rule that you can only make a transition, so-called D m, equal to plus or minus one, which is one spin flip. For the pumping scheme, you have to break that selection rule, which means that you have states that are a superposition of m-states. Everybody now knows that you get those mixed states, superposition states, automatically when you have a crystalline field with an axis not parallel to the magnetic field. All you have to do is to put the magnetic field at an angle to the crystal field axis. But, people in spectroscopy always avoided that. They’ve always wanted the simple case that the magnetic quantum number “N” was a good quantum number. That’s how you always arranged your experiments.
Goldstein:
So people steered away from it?
Bloembergen:
Exactly. I had to deliberately say, "No, we put the magnetic field at an angle to the axis of the crystal, the crystal field of the chromium ions. And then this happens, then you can make transitions between arbitrary levels.” They’re all allowed then.
Goldstein:
I wonder if there were any cases where somebody was working with a sample where the fields weren’t parallel and they got output that they couldn’t understand?
Bloembergen:
No, they understood that already. They did a lot of that work at Oxford, but there they never used two frequencies at the same time. It’s important to have the combination of double pump irradiation, a pump field and a signal field, two different frequencies. You study two transitions at the same time, but also have them between states that are a super position of Zeeman quantum numbers.
Goldstein:
When this occurred to you, were you surprised by the results?
Bloembergen:
- Audio File
- MP3 Audio
(256 - bloembergen - clip 1.mp3)
I expected them. I wasn’t surprised. I said, “This has got to work at one frequency or another.” And then the question was just a matter of finding the ion. Since I thought that theoretically it’s easiest to do it for three-levels, I solved those equations. I said, “Let’s state system with spin one that just has three levels,” because the degeneracy is 2J + 1, and if J is one then you have three levels. The nickel ion has three levels in the microwave region. That’s what we tried first. Now that was a very unfortunate choice because relaxation times in nickel salts are inappropriate. So then we tried chromium, which has four levels, and I picked any three of them. We put chromium in a crystal of potassium cobalt cyanide, because that’s the crystal we could grow ourselves from an aqueous solution, and we could add just the right chromium amount. We grew those crystals and we got the signal. We weren’t the first; the experiment was immediately confirmed at Bell Labs within a few months. We were still the first to get amplification at the twenty-one centimeter wavelengths, which is very important for radio astronomy, because that’s where the hydrogen hyperfine structure line is in astronomy. That was another example as to why it took us almost a year longer. We first had to grow our crystal, but more importantly, it was much harder to get the amplification at a relatively low frequency, twenty-one centimeters. You pumped the K-band and got amplification at the X-band. That was the first one done at Bell Labs. Of course, they also had a much larger group. Another ion, Gadolinium, had been explicitly suggested and has a spin of seven halves at eight energy levels. I had suggested Gadolinium, nickel, and chromium and I knew it could hold for any of the transition at lines.
Goldstein:
You said the nickel had the bad relaxation time. Did you understand it at the time?
Bloembergen:
Yes, we understood that. Also, the line broadens too much, and when the lines are too broad they overlap, and any overlap between these resonances kills the effect. That’s why you have to have the effect in dilute concentrations. People who work with 100 percent chromium salt would never get it. You have to have slightly colored rubies.
Bell Labs, patents
Goldstein:
Now you said at Bell they were getting amplification at X-band. They were interested, I guess, in communication systems?
Bloembergen:
They were interested in communication, and they had Rudy Kompfner telling the people, “What we need is a continuous wave operation in condensed matter, or even as a beam, but we need something that will operate continuously.” There had been schemes, which were well known, of giving a 180 degree process of inversion in magnetic resonance, and many groups played with that, but that clearly wasn’t considered practical. That was the case that Professor Strandberg at MIT called versitron because it reversed magnetization in one pulse. I heard a colloquium by him at MIT, and I said, "Why are you people so excited about that? I mean, everybody can flip a spin with a 180-degree pulse. Purcell did that at Harvard in 1951. But why do you want this thing?” Well, he said, "We want a low noise microwave receiver." And then I said, "Yes, that is important. That is an important application." And so then I started thinking of getting CW operation on a low noise microwave amplifier.
Goldstein:
But you were thinking of the twenty-one centimeter band for radio astronomy?
Bloembergen:
I had a reason to do it. Then I said, “We are going to build it for the twenty-one centimeter radio astronomy because they need it.”
Goldstein:
Was the group at MIT targeting a different band?
Bloembergen:
I don’t know what MIT did. The people at Lincoln Labs, Kingston and Lacks took over and they built microwave masers at the 21 cm. frequency. Everybody jumped in at that point very quickly. Bell Labs was first because they were close. What really happened, although it’s never appeared in print and maybe you should check it, was my former colleague, Les Hogan, who was at Harvard from 1954 to about 1958, told me, "Niko, this is an important idea, and you better write it up in a notebook and have it witnessed." I responded, "Well, you will witness it. I understand it." He was still in contact with Bell Labs, where he had invented, in the early 1950s, the Faraday Isolator, widely used now in microwaves as a one directional device. He apparently dropped a hint to his colleagues at Bell Labs, his former colleagues at Bell Labs, and said, "Bloembergen has an idea about CW operation, and it’s so simple you will kick yourself in your pants." That hint is all they needed for Scovil to probably independently hit on the idea. Later they sent emissaries out to find out exactly what I had, and I remember that Rudy Kompfner himself came over one day here at Harvard and we had a chat, and he was convinced. I showed him the notebook and told him what I had. So he invited me down to Bell Labs, and then Bell Labs, in the patent department, wrote the basic patent on three-level, multi-level pumping.
Goldstein:
For you?
Bloembergen:
For me. But they had a license, a non-exclusive license. Then Phillips became interested and they took out patents in twelve different countries. The remaining rights, apart from the non-exclusive rights of Bell Labs, I later transferred to Perkin-Elmer Co., where I became a consultant in the early 1960s.
Goldstein:
So you assigned the rights to Bell Labs?
Bloembergen:
I assigned non-exclusive rights to Bell Labs, and in return they took out the patent for me. They also paid me a nominal fee. The main thing was that the patent was written up through Bell.
Goldstein:
Was that to take advantage of their expertise in patents?
Bloembergen:
Yes. I didn’t know how to go about it. I was a babe in the woods.
Goldstein:
I’ve never heard of people doing that. Did you know anyone else who did the same thing?
Bloembergen:
The real problem is, as I learned the hard way, either you set up your own company and you hire your own patent lawyer or you have someone write up the patent for you. I was never interested in becoming an entrepreneur. You have a very hard time getting any financial compensation for your inventions because what mostly happens is that they cross-license each other between companies. I had no company, so I had nothing to cross-license. That was the real problem. I had the idea, and they acknowledged that I was there first, and so they acknowledged I was the inventor of the continuously operating solid-state maser.
Goldstein:
When you were working with the staff at Bell Labs did you feel that their interest in it matched your own?
Bloembergen:
Yes.
Goldstein:
Everybody had the same view of the potential?
Bloembergen:
Yes. They saw much more potential. They have this vast group of excellent people. They had a big group in magnetic resonance, and even in nuclear resonance, and I had been interacting with them at meetings, not directly at Bell Labs, but there were always magnetic resonance meetings and we read about each others' work.
Magnetic resonance development
Goldstein:
You talked about working with Pound and Purcell, and I wonder if they have had to take time off to work on the Rad Lab books? Do you know if that experience, working in the Rad Lab and working on the books, played a role in the development?
Bloembergen:
Yes. Everybody knew about magnetic resonance and molecular beams. They also knew that there was the possibility of doing magnetic resonance condensed matter, and they knew of unsuccessful attempts by this Dutchman, Gorter. This was in their free time when they weren’t worrying about radar warfare. They talked about this and that’s how Pound and Torrey and Purcell got together. In their spare time on weekends they hooked up a thing in the building next door just before Christmas, 1945. The second time they tried it, I was already here, and I had to prepare a sample that had both protons and fluoride nuclei. So I made a mud pie of calcium fluoride powder and mineral oil, and we put that in the microwave cavity. They were so influenced by microwave techniques that they hadn’t built a simple LC circuit -- a simple self-inductor and a separate capacitor. They had a big box of about a liter volume, and it was a microwave cavity.
Goldstein:
They didn’t use normal radio engineering techniques?
Bloembergen:
No, they had the loop, it was a cylindrical cavity that was the central post, and then capacitively loaded. It was very heavily loaded to get the frequency low enough. It was a sort of microwave cavity design operating at thirty megahertz. My first task was to build an LC circuit, reducing the volume by a factor of a thousand.
Goldstein:
Could you get good resonance from such an apparatus?
Bloembergen:
Yes.
Goldstein:
Tell me a little bit more about the apparatus that you were using.
Bloembergen:
I built an RF bridge based with an LC circuit, and since I wasn’t really that familiar with radio engineering, I built a very primitive RF bridge with a second identical arm. They were then hooked up in bridge form so as to give a “no” result, except that the one arm was in the magnetic field and would get unbalanced because of the magnetic resonance. I built the bridge with two identical LCs. It was a sort of a home built RF bridge.
Goldstein:
Was there anyone around who you could talk to for advice?
Bloembergen:
Yes. I knew about phase sensitive detection. We had used a so-called AC Galvanometer for that purpose, and Purcell was very excited, so we used phase sensitive detection. We had a Dicke lock-in amplifier. We had to build according to Dicke’s design, a lock-in amplifier, but with all the RCA components I quickly learned. Purcell said, “Talk to some graduate students around the lab.” One was very helpful, and he introduced me to the stockroom and to the RCA Handbook.
Goldstein:
Did you get any interesting ideas from that kind of work, or was that just a technique you needed to learn?
Bloembergen:
I have to be familiar with what is available, and as soon as I have the minimum to do my experiment, then I’m satisfied. I don’t go for big product engineering improvement. I never did.
Goldstein:
Did you need high precision?
Bloembergen:
Well, in the first experiments we didn’t. Of course, then you try to get better signal to noise and so on and make those analyses. Yes, that comes second.
Pumping scheme patent and publication
Goldstein:
When did the patent for the pumping scheme come out?
Bloembergen:
I’ve forgotten the year. We applied in 1956, and I think it came out in 1959. It was awarded in 1959.
Goldstein:
What happened after that?
Bloembergen:
Well there was another thing. I was worried about classification because it would be a low noise detector, and in fact, it was used in the DEW-Line, the early warning system. They installed masers for a while along the Arctic. I heard that, if you made a submission to the patent office, the submission was automatically reviewed by the security people. Before I applied for the patent I submitted an article to the Physical Review and it took about four months from submission to publication.
Goldstein:
So the cat was out of the bag?
Bloembergen:
I put in the patent application just before the printed journal appeared. I don’t know whether that had any influence or not, but at that time I was concerned they would want to keep the information classified.
Maser research
Goldstein:
So tell me what happened after 1956. Did you continue with this kind of research?
Bloembergen:
Yes. We worked on the details of the maser. We wanted to understand under what circumstances does it really work, and we did a lot of cross-relaxation. If there is a lot of relaxation between different resonances, you are in trouble. We quantitatively analyzed at what concentration of paramagnetic ions the thing would quit working. We really wanted to understand the details of cross-relaxation. It is very basic for the operation. Many people jumped into the research, so the traveling wave masers were developed and we followed that.
Goldstein:
It seems to me that in your career you eventually changed from working on masers or lasers to using them for other research?
Bloembergen:
Until 1959 or 1960 we worked on masers. I realized that the same principle would apply to light frequencies, but I never tried to build a laser because I thought it would be too difficult for our situation at Harvard. We had experience in magnetic resonance. We had nothing in optics here, and we didn’t know how to polish crystals. I knew that with such a race going on elsewhere, that rightly or wrongly, I stayed out of laser development. But, I realized as soon as Maiman’s optical maser, the ruby laser, was a reality, that you could do a lot of new physics with it.
Lasers and nonlinear optics
Bloembergen:
So right in 1960 we started thinking more about optics, and that eventually led to nonlinear optics. I never developed any kind of laser. I never improved on any of them. I just used whatever was available to study the optical properties of matter at high light intensities, which is basically the field of nonlinear optics. It turned out to be very interesting. I never did any engineering on lasers.
Goldstein:
Can you tell me a bit more about what difficulties you thought would come with working on optics?
Bloembergen:
Look at what really happened. Practically all types of lasers were first realized in the industrial research labs in this country. It required a massive organization that could focus on the problem of combining radio electronic techniques with optical techniques. The first laser by Maiman was made at a research lab. Ali Javan at Bell Labs built the second laser, the gas laser, by using helium neon. Then came the semiconductor lasers, almost simultaneously in three places, General Electric, IBM, and Lincoln Lab. These were all organizations that had enormous expertise in either gas discharges or semiconductors and so on. Then came the dye laser at IBM in 1966 and then the CO2 laser by Patel at Bell Labs. All the important types of lasers were first realized in industrial research organizations in the 1960s, which were booming.
Goldstein:
Yes, they had the staff.
Bloembergen:
They had the size. They also had the administrators that said, "This is what this organization is going to focus on." At Harvard we have no technicians to speak of, because they have to be supported by soft money and we were always very, very cagey in having technical support, because you couldn’t guarantee these people a life-long position. I am sure that if I had joined the race to develop the laser I would have lost it. I know that. When you work with students you have to train them. Even in nonlinear optics that was an important consideration. We could do theory very well and could compete there, but in experimentation, to compete in those days with Bell Labs, RCA, General Electric, IBM, Hughes was almost impossible. So what I chose in nonlinear optics were unpopular topics. What was an unpopular topic? We did the first measurements of nonlinearities in absorbing media. Of course, everybody used practical devices of clear transparent crystals, so they were already ahead of us experimentally there. We concentrated on nonlinearities of properties of metals and semiconductors because they weren’t useful. You could observe them in reflection and develop very interesting theories of complex susceptibilities and nonlinear susceptibilities in those materials.
Comparison of corporate and academic labs
Goldstein:
I want to understand better what expertise those big labs had, like IBM or Bell Labs. I mean, what did they know? What specific subjects?
Bloembergen:
Well, they knew magnetic resonance. Anybody with a Ph.D. in magnetic resonance got hired in those days. They had expertise, and students who had done microwave spectroscopy at Columbia or here in chemistry. They had the experience, and they used it.
Goldstein:
When you say that they could do what you couldn’t, I want to know what exactly you mean?
Bloembergen:
Their researchers joined a large organization. I mean, I could do it if I had twelve post-docs and a big budget and so on. But, we had a very small organization. At Harvard, everybody has to get his own contract, and not everyone has the entrepreneurial capabilities of running a large contract and hiring twelve post-docs, and additional technical staff. It was really a small individual operation. My maximum size was four post-docs and twelve graduate students, and I only had that for a limited number of years, because that’s really too much to pay personal attention to.
Research collaboration
Goldstein:
So if your early work in magnetic resonance led to masers and lasers, can you tell me what else it lead to, where else did it branch off?
Bloembergen:
It branched off into chemistry and chemical analysis. There followed the high-resolution spectroscopy of the magnetic resonance line and these very fine magnetometers and the atomic clocks.
Goldstein:
I wanted to learn more about the connection between what you were doing and what Ramsey was doing.
Bloembergen:
We really didn’t talk that much. I mean, we said, “hello” and we kept abreast of the developments. I knew he was working on the hydrogen maser, but we really didn’t interact on a daily basis.
[End of tape one, side a]
Bloembergen:
I’m not exactly a loner, but I’m also not very gregarious and I don’t always want to interact with colleagues. I want to do my own thing, and I like to do things from A to Z, so as not to be a cog in a much bigger organization. I always liked that these were small-scale experiments, mostly NMR or the nonlinear optics. They fit on a single tabletop, and you can really understand the details of what each student is involved with. It’s not a grand industrial picture. On the other hand, I always enjoyed my contacts with industrial organizations. For NMR, I was a consultant for Schlumberger Oil Surveying Corporation. It’s really an oil service company. It’s on the big board. It’s been traded on the New York Stock Exchange since the 1950s. They wanted to use magnetic resonance probes to search for oil. It’s a fantastic development. They use photon counters and gamma ray counters operating at 200 to 300 degrees centigrade down in these bore holes.
Goldstein:
So they had that idea and they needed your expertise?
Bloembergen:
They wanted to probe. What they really wanted was to measure relaxation times, so they could distinguish between water and oil. My thesis had been on nuclear magnetic relaxation, so there was a direct relationship.
Goldstein:
Did you help them design the probes?
Bloembergen:
No, they designed them. They had very good electromagnetic engineers. I just listened. I was a sort of sounding board and made suggestions, but not on the engineering aspect. It was very impressive what they managed to do.
Goldstein:
They contacted you?
Bloembergen:
They contacted me. I lectured them on the basic principals of nuclear magnetic relaxation, and I listened to them and made suggestions. The same thing happened in nonlinear optics. I was consulting with Perkin-Elmer in the 1960s. At the time, I was on several government committees in Washington on the possible application of lasers in national defense.
Goldstein:
What use did Perkin-Elmer have for non-linear optics? What were they trying to do?
Bloembergen:
They didn’t know, but they wanted to be in on lasers. They finally beat us out and built some helium neon lasers. In the 1960s, all the electronics or optical companies wanted to be involved in research. Those were the golden times. Now they’ve all cut back on their research, but not on their development.
Goldstein:
So how did you help them out?
Bloembergen:
Well, they were interested in building lasers for some of their optical instruments. They thought a laser might be handy in the spectrometers. So again, I lectured on the basics. Everybody in those days was trying to build electro-optic modulators, which took a much longer time to develop. It only came to fruition decades later.
Electro-optic modulation
Goldstein:
Did you ever get involved with that problem?
Bloembergen:
I have a basic patent on electro-optic modulation and on what is now called quasi phase matching, which has been made practical in thin films. But, the basic idea of quasi phase matching we had in our first long theoretical paper on the theory of nonlinear optics. I had taken a patent out, but of course, it’s only for seventeen years, and only now, thirty years later, are there the first rumblings of a possible technology for these devices.
Goldstein:
Are you interested in the reduction to practice process?
Bloembergen:
I’m interested in following it from a distance, not in doing it myself. I’m not personally involved. If I were, I probably would have started my own little company. I was never interested in the detailed engineering.
Goldstein:
Too many details?
Bloembergen:
It’s probably my background. I started as a physicist and I’m still mostly a physicist.
Goldstein:
Would it aid your understanding?
Bloembergen:
It doesn’t help the understanding of the basics. I always found my interaction with the corporate world fascinating and mutually beneficial. They learned some basic principals; I learned some new techniques. But, I never really improved on those new techniques.
Corporate consulting
Goldstein:
When you were working as a consultant for companies, did your interest ever diverge from theirs?
Bloembergen:
Yes. I often did it on the side. Consulting, as a former colleague said, is “like wine.” It should be taken in moderation. The beneficial policy is twenty percent of your time or one day a week. What it usually amounted to was that for several months I would do no consulting at all, and then I went for a visit of several days in the summer. I loved the interaction. New viewpoints. New problems. You broaden out and you hear many different things. They used me as a sounding board. Often it was a show and tell at the industrial site. It stimulated their workers because they thought, "Here’s a basic scientist interested in what we are doing." So it’s mutual.
Goldstein:
Can you think of any examples where you got some interesting science out of a project that a client showed you?
Bloembergen:
Can I come up with one right now? No. I know that in many cases it was very stimulating for me to come back from those discussions. I think the high power, CO2 laser, the gas dynamic laser, I became interested in because of the interaction with government committees. When there was a new type of laser coming out, I heard about it often at Perkin-Elmer, or at IBM.
Goldstein:
So you were able to use some of the lasers that were coming out to do research?
Bloembergen:
Yes, and I was aware of their coming.
Goldstein:
Were the lasers an important tool for you in your nonlinear optic research?
Bloembergen:
Yes. I needed them to do my research on nonlinear optics.
Goldstein:
Were there any other important tools that came out to help you do these experiments?
Bloembergen:
The laser in the '60s was always the limiting component. The development of the dye laser was very important for spectroscopy as were tunable lasers, and the pulse type power CO2 lasers, the TEA lasers.
Goldstein:
Any other devices useful for data collection?
Bloembergen:
Yes. When you visit a lab or a colleague you see how things are done and you can combine things.
Directed energy weapons
Goldstein:
Tell me about your research. Are there any other new periods after you started working in nonlinear optics?
Bloembergen:
In later years, I got involved with things that are really sort of extracurricular. Because of my interest in lasers, I was asked to chair this study on directed energy weapons, which I did with Kumar Patel. We were co-chairmen of that study. It attracted a lot of attention, especially in Washington, because the Cold War was still going on, and this was part of Reagan’s Strategic Defense Initiative. Even during the Cold War on the basis of our study they cut back their plans on deploying the big weapons in space.
Goldstein:
Have you thought about why you were selected?
Bloembergen:
I had knowledge of lasers, but the main thing was that we had a committee of fourteen, and the first requirement was that people shouldn’t have any political preconceptions. That means that they should have never been in print anywhere favoring the Strategic Defense Initiative, so that they were at least publicly neutral. I know that on our committee there were some people on the right and there were people on the left, but we got a report out which was unanimous. There was no minority opinion. We often had long discussions, and then I said, "Look, here we are all engineers and scientists. Why can’t we agree on facts?" It was either one of two things. In some cases the question was a purely political one, and in those cases, we said, “That should not go in our report. That is not what we are here for." In other instances, the disagreements were on technical issues. The wording that was used in the draft was not neutral. So then we would go very carefully through it and reword it so that we achieved a unanimous report with people who had a very broad political spectrum of opinion.
Goldstein:
What were some of the political questions that you had to push aside?
Bloembergen:
Well, the obvious one was that Reagan was motivated by the Cold War strategy. It may have worked that it helped to bankrupt the Soviet Union. That was a typical question that we wouldn't address. What we were there for was to determine whether it was feasible to design a system of strategic defense by deploying directed energy weapons in space. There were two types of directed energy weapons. They were either high-powered lasers, with large mirrors and telescopes, or they were accelerated hydrogen particles.
Goldstein:
When you were thinking about the feasibility, did you have to project the rate of technological advance?
Bloembergen:
Yes.
Goldstein:
And how did you do that?
Bloembergen:
The important thing was that from the start we insisted that we have access to all classified information. But, we had to come out with an unclassified report. That was difficult. It caused some problems. We had the cooperation of the Strategic Defense Initiative Organization. They approved of the people on the panel and they were interested themselves in having a technical assessment of what they were doing. The actual workers inside SDIO realized that this would be needed because they were running off wildly in all kinds of directions. We visited various sites and had many meetings and received input from them. We analyzed it and came up with the report. As each year goes by, and it’s now ten years, the report is more correct. We were right on. The report was delayed by two security investigations. One was clearance by the SDIO, and then it went higher up and we also needed clearance from the Secretary of Defense, and that meant another four months study. The report was delayed by eight months.
Goldstein:
Do the original versions exist, the versions that were drafted before review by these boards?
Bloembergen:
There was very little change. There were rather trivial changes here and there. A number of items were classified and in fact were declassified so we could put them back in at the end. We knew what we could write. We were very knowledgeable, so we knew what we could not say.
Goldstein:
Do you think you were conservative in your guesses about how technology would advance?
Bloembergen:
No, we were right on.
Goldstein:
I guess what I’m asking is whether your accuracy was due to a conservative viewpoint or has technology grown just as rapidly as you expected?
Bloembergen:
Technology and the how of things hasn’t changed that much. They are now talking about reviving the so-called alpha laser, which was the most advanced and most powerful chemical laser. We knew that it existed, two million watts is the figure we could quote for continuous wave operation. One such thing might operate in space, but not a whole system, and it wouldn’t be capable of tracking the hundreds of missiles that would come simultaneously in a strategic defense situation. The problem now, of course, is very different. Now it’s how to defend against an occasional rogue missile, so the problem has to be reanalyzed.
Goldstein:
When you were projecting what technology would exist in the future, did you only consider projects on the drawing board, or did you look at fundamental research and think how that might evolve?
Bloembergen:
Well, I don’t know what you call x-ray lasers. You call that fundamental, or on the drawing board?
Goldstein:
It’s a tough call in each case.
Bloembergen:
We had to say something about that, although it was very highly classified. We realized from what we heard that it wouldn’t fly, and then Congress, years later, decided to abandon the project after spending one billion or more on it. The technological issues are really too formidable.
Goldstein:
In addition to the high-powered lasers, there’s also the computational problem if you’re trying to develop a system.
Bloembergen:
Yes. We had systems experts on board.
Goldstein:
How long did you spend on that committee?
Bloembergen:
It was really two and a half years.
Goldstein:
The other members, did they seem to be scientists or engineers?
Bloembergen:
Both.
Goldstein:
Both worked in the same capacity?
Bloembergen:
Yes. It was very, very interesting.
Goldstein:
It sounds like the physicists really minimize the difference between the physicists and the engineers?
Bloembergen:
Yes.
Technological advancement
Goldstein:
Is there any other period of your research career that we should discuss?
Bloembergen:
No, I think we’ve covered it. The three main phases: magnetic resonance, masers, and then nonlinear optics. Looking back on my own life, it is a very continuous, historical development.
Goldstein:
Do you think that development is analogous to the development of the science and technology?
Bloembergen:
It’s always a mutual interaction. Sometimes technology is ahead. It happened after World War II with the microwave technology. Then people realized they could do a lot of new science with it, and the new science led to new technology, and so on. It was the same with lasers; there were these very interesting schemes to get very short pulses, pico second and now femto second. That was stimulated by the mode locking technique from microwave frequencies.
Goldstein:
Are there any examples where the technology got ahead and you personally thought, "Oh, I need to understand that?"
Bloembergen:
In the laser field, the technology got ahead, and the laser was a solution looking for a problem; it was the same with the short pulses. When the short pulse came along there was enormous new science you could do in studying chemical reactions while they occurred in real time.
Goldstein:
I see that as being a little bit different because there’s a technology that lets you do science in another field, but it’s not like you needed to understand the fundamental operation of the short pulses.
Bloembergen:
That is true. The fundamental understanding came from the microwave technology, mode locking, and saturable absorbers. The science came first. But, you are thinking of a technological development that wasn’t foreseen by the theorist.
Goldstein:
I thought you were saying that that sometimes happens?
Bloembergen:
Yes. As I say, a very good example is the microwave technology that came first. Now, clearly, the big example is computer, semiconductor technology, large-scale integration and then large-scale computation. I think that came first from the engineers. Although if you trace it far enough back, of course, it all started with the transistor, and the transistor was based on solid state in the 1930s, electrons and holes. So there’s always this give and take. I mean, before that the vacuum tubes. You can say that the first person there was Thompson who discovered the electron, and very quickly Lee DeForest built a vacuum tube, and that clearly enabled science to do an awful lot.
Devices as research tools
Goldstein:
In your career, it hasn’t happened too much that a new device comes out and it becomes available and you want to understand that device’s operation. It seems like you’ve used devices to understand other science.
Bloembergen:
Yes. I used the laser to do properties of atoms and matter at high light intensity. We’re now at very short times, femto-second time scale and so on.
Goldstein:
Can you tell me some of the results of that research?
Bloembergen:
It’s in a period of very rapid growth. People now talk about femto-chemistry, femto-biology, and I have used the interaction of femto-second pulses with semiconductors, gallium arsenide and germanium, and what really happens on that time scale.
Goldstein:
Has the output brought about some basic understanding?
Bloembergen:
Yes. We have a phase transition that occurs in the electronic structure of gallium arsenide on a time scale so short that the atoms, the gallium and arsenic atoms or ions, have had no time yet to respond. They are not yet moved from their tetrahedral position. But the electron gas has already lost its band structure, and so that’s a very rapid phase transition, only for the electrons and not yet for the atoms. That’s a new concept.
Goldstein:
Do you know if anybody’s working on that in an industrial context?
Bloembergen:
No, not in an industrial context. Of course, this is all based on laser annealing, which is used very heavily in semiconductor technology and metalworking. It’s known that lasers can on time scales of pico and nanoseconds anneal materials at the surface. As for the femto-second time scales, there are no specific industrial users. The big payoff would be femto-second x-ray laser pulses.
Goldstein:
Do your experiments still fit on a tabletop?
Bloembergen:
Yes, they still do. That’s an amazing thing.
Goldstein:
Have you still kept away from large research groups?
Bloembergen:
Yes. Right now, I don’t do any research of my own. All I’m good for is to give keynote speeches and historical reviews, like this one.
[End of interview]
