Interview: Paul W. Palmberg
1998 Gaede-Langmuir Award Recipient
November 1998
POWELL: Good afternoon. My name is Cedric Powell, National Institute of Standards and Technology. I’ve been a long-time member of the AVS. We are here today in the Baltimore Convention Center, Wednesday November 4th, 1998, on the occasion of Paul Palmberg being awarded the AVS Gaede-Langmuir Award. Paul is being cited for his innovative and revolutionary inventions that resulted in the development of practical energy analyzers for surface analysis by Auger electron spectroscopy and x-ray photoelectron spectroscopy. I’ve known Paul for about 30 years. Paul received his B.S. and M.S. and Ph.D. degrees from the University of Minnesota. He went as a post-doc to Cornell University, worked at North American Rockwell Science Center, and then with others formed the Physical Electronics Company. He’s had a distinguished scientific career. He has a number of patents and IR-100 awards, and he has run a very successful business. It’s a pleasure to have you here with us today, Paul, on this project.
PALMBERG: Thank you, Cedric. Thank you very much.
POWELL: You’re welcome. I’d like to ask you a number of questions. First, could you tell us how you got into surface science? You were there in the early days of surface science in the modern generation. Could you tell us about that?
PALMBERG: Certainly, yes. I had received my degree, as you indicated, from the University of Minnesota, and my undergraduate degree in 1959. At that time, there was a very active research program going on at the University of Minnesota for the development of low-threshold or infrared sensitive cathodes. In particular, the Defense Agency in the Army was interested in infrared sensitive detectors. So I joined in graduate school the research team led by Professor Peria in the development of such devices. I did fundamental studies, low-energy electron diffraction, work function, and secondary electron emission studies on germanium and sodium-covered germanium. So that’s how I came about to use a low-energy electron diffraction apparatus for the purpose of determining the energy distribution of secondary electrons from low energies all the way out to the elastic peak of their primary reflected electrons from solid surfaces.
POWELL: Those techniques had very good surface sensitivity, but they did not give information on surface composition. Could you tell us about the development of Auger electron spectroscopy, which does give this information, and how did you get involved in that particular topic at that particular time?
PALMBERG: Well, that of course, was one of the frustrations of surface science at that time, is that we had no direct method of determining the chemical composition. Everything was inferred from work function, and in my case, low-energy electron diffraction experiments. It happened that when I was at Cornell University, and I’d left the University of Minnesota, that Larry Harris at General Electric had begun a program to develop a practical Auger spectrometer using 127º sector analyzer. That work came about because of some experiments that Rey Whetten was carrying out, using such a device on graphite. It appeared to him that it would be quite a sensitive technique.
At the Physical Electronics Conference in Boston, I believe it was around 1968, Virgil Stout talked to Bill Peria about the work they were doing, and Bill Peria recognized that the apparatus I had constructed while at the University of Minnesota could be easily converted to do Auger spectroscopy by simply differentiating the retarding field characteristic twice, and they did that with a lock-in amplifier. As it happened, I had visited my friend, Roland Weber, on a weekend from Cornell, subsequent to the Physical Electronics Conference (I think it was around June of that year) when he was prepared to do these experiments. So I had the occasion to work with him, to obtain the first Auger Spectrum using a LEED apparatus on silicon. Then we put cesium on the surface and determined that, in fact, it was very, very surface-sensitive.
I, of course, was very excited by those results, and immediately went back to Cornell and converted the Varian LEED apparatus we had there at that time to do Auger spectroscopy. So that’s how I got started with the technique.
POWELL: You mentioned the subject of surface sensitivity, and I, for one, remember an important paper that you co-authored with Thor Rhodin when you were a post-doc at Cornell on measurement of surface sensitivity and Auger spectroscopy. Could you tell us about that early work and how it got started, and any details that you care to mention?
PALMBERG: Yes. Prior to converting the low-energy electron diffraction apparatus to an Auger spectrometer, I was using low-energy electron diffraction to determine the atomic structures of the overlayer structures. So I was already prepared to deposit very clean metal films, monolayer type thickness, onto a clean metal substrate. Using this technique, I deposited silver on gold, monolayer by monolayer, while monitoring the Auger peaks from the substrate and the overlayer, and was able to determine that the escape depth was in the range of 4 Angstroms at about 300 eV. up to about 8 Angstroms at somewhat higher energies. So it was very soon determined that Auger spectroscopy was an extremely surface sensitive technique.
POWELL: Okay. That’s very interesting, and just a comment, that this is still an active subject of inquiry today, because,still, people need to know values of this parameter. Probably the single most important development for the growth of Auger spectroscopy has been the development of the cylindrical mirror analyzer. Could you tell us about how this came to be developed, and particularly your role in it?
PALMBERG: Yes. Subsequent to my post-doctoral fellowship at Cornell University, I went to the Rockwell Science Center at Thousand Oaks, California. While I was there, I came upon an article by Sare in the Review of Scientific Instruments in which he described the CMA. It was immediately obvious that this would be a rather ideal instrument for doing Auger spectroscopy, in that the transmission of the device and the signal-to-noise ratio should inherently offer greatly improved sensitivity compared to either the 127o instrument used by Harris or the Auger LEED apparatus. So I immediately set out to build such an instrument, and in the early testing it became very apparent that this instrumentation had the possibility of revolutionizing the Auger technique.
POWELL: You were one of the founders of the Physical Electronics Company. Could you tell us something about how the company came to be founded, and who were the key people in the early days?
PALMBERG: Yes. Roland Weber, who was a colleague of mine as a graduate student at the University of Minnesota, after his pioneering Auger work, which he conducted for his thesis, went on to become a professor in the Electrical Engineering Department at the University of Minnesota. He continued his surface science activities and found that there was a great deal of interest in the LEED Auger instrumentation. He and Bill Peria, as a favor to some of their colleagues in industry, built such LEED-AES devices. It seemed to them that this would be a viable commercial technique to start a new company. Roland Weber tried to get me to join at that time. I was a little reluctant because I felt that the market for LEED-AES may not be large enough to sustain a company. But then as I got into the use of the CMA, I changed my mind; it seemed to me that the CMA had a possibility of being a very viable tool. I was aware at that time that the quadrupole mass spectrometer, which mounted onto a single flange, became a very important adjunct to any kind of surface science activity. It seemed to me that a CMA, particularly if we could put an electron gun inside it, so as to make it a very compact device, could be used in a very similar way to allow every surface science investigator a simple means of determining the chemical composition of the surfaces under investigation. That’s how I happened to finally decide to leave my career at Rockwell and join Roland and others at the University of Minnesota in the starting of Physical Electronics.
POWELL: Just for the historical record, what was the year that that happened?
PALMBERG: The company was actually incorporated in April of 1969. I joined the company on a full-time basis in February of 1970.
POWELL: Could you tell us a little about some of the early developments of the company, and how you interacted with your market, or created a market, in fact, and some of the main products in those early days?
PALMBERG: Of course, they already had begun when I joined in February, 1970. They were already manufacturing the LEED-AES system. I immediately set out to incorporate the co-axial electron gun into the CMA. We introduced that product in the fall of 1970. It turned out to be an extremely successful product. The market was there. We really didn't have any sales force at all, but just simply with our own contacts in the research community, we ended up growing the market very rapidly, right from the outset.
POWELL: One of the earliest successes of the CMA, and in fact, one of the earliest applications, was for thin film analysis. And I believe you interacted a lot with Fred Wehner at that time. Could you tell us about that and about the development of that application?
PALMBERG: Yes. It was really Fred’s suggestion to use AES for composition depth profiling. As you know, Fred Wehner was, as he called himself, the mother, or rather father, of sputtering. He had suggested that maybe we should incorporate an ion gun into the Auger system so that we could controllably remove material from the surface, and in this way determine the depth composition going into the surface. We were a bit hesitant, or at least I was initially, because at that time we did not have a differentially-pumped ion gun, so we had to backfill the whole system to a pressure of about 10-5 Torr. We were concerned that the gas would affect the operation of the CMA analyzer. As it turned out, it really had no significant negative effect at that pressure, and we were able to simultaneously monitor the surface composition with the CMA while steadily removing material from the surface, and in this way, obtain an Auger depth profile. That had an immediate impact on the market; people in the semiconductor industry, in particular, were very interested in studying the inter-diffusion between films using such a technique. So this immediately expanded the market as a result.
POWELL: One of your other early products was a handbook on Auger spectroscopy. I’ve been amazed,…well, not amazed but surprised at the large number of citations that this handbook has been utilized in literature. Could you tell us a little bit about it, and how it came into being?
PALMBERG: Well, as soon as we started marketing the Auger spectrometer, it became obvious that a significant impediment was the interpretation of the data. While it was not so difficult to interpret the data if you already had experience, new users were rather overwhelmed to see all these peaks and try to understand which element they originated from. So we set out, myself and my technician, to obtain—and very quickly, just in a matter of a few weeks—spectra from all of the most important elements and quickly put that together into our original handbook, which was a tremendous aid. Subsequent to that, of course, the handbook was refined with much better data, more accurate data, better energy resolution and so forth.
POWELL: Okay. Physical Electronics is now a successful company, but it probably wasn’t easy in those early days. Could you tell us something about the early days of the company, and particularly how it fared as a business?
PALMBERG: Well, one of the strengths, and also weakness, of the company is that we had depth of scientific talent in that the four principals in the company, myself, Noel McDonald, Jerry Rich, Roland Weber, all had Ph.D. degrees in Electrical Engineering. However, we had nobody in the company with any background in business. We made a mistake by selling, in early 1970, the company to a holding company, Bayfield, which was also involved in many other start-ups, completely unrelated to our business. It was felt that that selling to the holding company was a way of averaging the risk. Well, it turned out that all of the other companies the holding company was involved with went bankrupt. So there we were, about to go down with the sinking whale. Fortunately, with the help of two venture capital groups, we were able to extract ourselves from Bayfield and make Physical Electronics independent. There was a time, though, while we were going through that restructuring phase, when we weren’t even able to pay our bills and we had to extend the payments to some of our creditors. However, surprisingly to the creditors, we paid off everybody and within a matter of about six months, and quickly became a healthy financial organization.
POWELL: Paul, you’ve been intimately involved in the development of instruments for x-ray photoelectron spectroscopy, or ESCA. Could you tell us something about your role in the development of those instruments?
PALMBERG: Yes. We were obviously very successful as being the first company to commercialize the CMA Auger instrument, and it naturally led us to think about how we could also enter the XPS market. Of course, there were already several companies in this field that marketed complete XPX systems, but there were no components available for doing XPS or UPS. To convert the CMA into an XPS spectrometer, we had to find some way of defining the source area. Of course, in Auger instruments, the focused electron beam did that for us, defining the source area. But in XPS, where you typically use a flooded x-ray source, we decided to add a second stage to the CMA with an aperture between the two stages to accomplish source definition. We had this idea and then around 1971, Gerry Lapeyre, who was a professor at Montana State University, stopped by Minnesota on his way to the synchrotron, in Stoughton, Wisconsin, and we talked about the possibility of using the CMA for his studies on the synchrotron. We decided to build him a prototype double-pass CMA device. We also had to put spherical retarding grids in front of the CMA to slow the electrons down in order to get better energy resolution in the CMA. And so, within a few months time, we had constructed such a device, and Gerry used it very successfully in his synchrotron experiments. That led to the development of a commercial device which we sold as a component to many researchers who were then working on synchrotrons, or with UV sources in their own labs. So the double-pass CMA very quickly became a very popular component. Over the next several years we sold more than 500 units to the research community. We also used the double-pass CMA as the basis of our ESCA systems for the next ten years or so.
POWELL: What about the more modern instrumentation, which is being developed and marketed by your company, the Quantum 2000? What could you tell us about it’s development and, in particular, your role in it?
PALMBERG: Yes. That was a fairly recent development, going back about six years ago, when I was the general manager of the Physical Electronics Division of Perkin-Elmer. Paul Larson, Dr. Paul Larson, was the technical expert in support of XPS instruments at PHI and he and I had many discussions on how we could improve the spatial resolution of ESCA. Most of the other ESCA instruments flooded the surface with x-rays and then fused analyzer optics to sense only a small region within that excited area. These techniques had the inherent disadvantage in that you were causing radiation damage to a large area, but only getting information from a small portion of that area. We decided to pursue the use of the elliptical monochromator, to see how small they could make the x-ray beam. And with some initial experimentation, we were encouraged. We had taken our standard commercial monochromator, which had 36 crystals, masked off all but one, and found we could get a beam size down to about 12 microns. So then we set out to make a more practical device in which we used the best available quartz crystals, found a way to adhere the crystals to the elliptical glass substrate without causing a lot of distortion and within a short time, we found we could readily get the X-ray beam diameter down under 10 microns, and in some cases, even down to 5 microns. We could scan this X-ray beam by simply scanning the electron beam on the anode. So this resulted, I think, in a major improvement to XPS, in that we could not only get smaller areas, but we could also obtain chemical images of the area of analysis by scanning the x-ray beam over the surface.
POWELL: You’ve been successful as a scientist with many publications in journals. You’ve been recognized for your innovation by the award of three IR-100 awards. You’ve also run a successful instrument company. What advice would you give young people today, who would try to emulate your success, any one of those three areas?
PALMBERG: Well, I suppose that one of the important factors is to be lucky—to be at the right place at the right time. In many ways, I feel very fortunate to have entered the surface science field at the time that it was just beginning to accelerate, lardely as a result of the availability of standardized stainless steel hardware. Such hardware made it easier to develop, surface science instrumentation.
Another factor, I think, is that when you see an opportunity, to move quickly, don’t hesitate. I feel that one of my characteristics is to try a lot of things without worrying about failure. For every success I have had, there were many failures. So don’t be afraid to fail. Give it a shot. If it doesn’t work, try something else.
Also, with respect to the starting of Physical Electronics, my advice is, if you have a dream pursue it, better to do it while you’re young. If Physical Electronics had failed, I was right back to where I started. I didn't have anything before and I wouldn’t have anything after. I would just have gone back to doing what I was already doing, which was research. But as it turned out, I think we were very fortunate and I feel I’ve had just a wonderful career, and have enjoyed every minute of it.
POWELL: That’s great. Surface analysis, we all know it has been used extensively in many technologies of interest to the American Vacuum Society, particularly in more modern advanced technology, as well as the older technologies. From your vantage point, what would you see for the future for surface analysis and it’s application?
PALMBERG: Well, clearly, where the market is going for surface analysis equipment is to more and more sophisticated applications requiring better and better depth resolution and, perhaps even more important, better spatial resolution. There’s only so much that you can do with depth resolution, other than use more gentle or less energetic ion beams and grazing detection angles. All these are certainly being pursued in modern instruments. So I think the most important area for improvement is in spatial resolution. Spatial resolution is determined in most cases by the size of the exciting beam, either x-rays or electrons, so a lot of effort is being made to form more intense beams, smaller diameter beams. But along with that, of course, goes more damage. So if we look down the road a bit, I think there’s got to be more attention paid to improving the overall sensitivity of the detection system. For electron spectroscopy, this would mean, for example, being able to analyze the energy of all of the emitted electrons at once. This could be accomplished, for example, with a pulse excitation beam, and a time of flight electron spectrometer. Up to this point, the electronics has not been fast enough to accomplish this with the precision needed, but as time goes on, the improvements available in commercial electronics may allow this to happen.
POWELL: Well, it’s been a pleasure to have you here with us today, and thank you very much.
PALMBERG: Thank you, Cedric.