| Charles S. Fadley - 2005 Medard W. Welch Award - Interview
Dr. Charles S. Fadley
2005 Welch Award Winner
PH - I am Paul Holloway, a member of the AVS History Committee. As part of the Society's Historical Archive series, today I will be talking with Dr. Chuck Fadley who is this year's Welch Award winner. It is November the 2nd, 2005 and we are at the 52nd Annual AVS symposium in Boston, Massachusetts. Let's read into the record the citation for the Welch Award that you received, Chuck. That is "for the development of novel techniques based on photoelectron spectroscopy and synchrotron radiation and their application to the study of the atomic, electronic and magnetic structure of surfaces and buried interfaces." That is quite a mouthful but it is a short synopsis of a long career of distinguished accomplishments. Congratulations on receiving the Welch Award.
CF - Thank you very much, I am very honored to have received it and to be on a list with a lot of people that I have always admired and that are very distinguished scientists. I humbly accept this award.
PH - It is well deserved and you certainly are quite an accomplished scientist. That is part of what we hope to accomplish by doing these recordings; to get some flavor for how you got to where you are, and what drove you and who influenced you in your lifetime. So could you give us a little bit about your background?
CF - Well, lets see. I grew up in a small town in Ohio of 10,000 people and went through the standard public school system in the 1950s. I would say that in high school I already decided that I liked technical and scientific things, maybe through a long-standing interest in building stuff and figuring out how things worked and so forth. Then what year was Sputnik? I should know that.
PH - I think it must have been '59 or '58. I don't remember exactly.
CF - Yeah, it may have been a little earlier, I don't know honestly. (Ed. note: Actually it was October 4, 1957) But that came along and suddenly it was really quite a national endeavor to boost science and technology. So you could both be interested in it and feel like you were involved in a grand, momentous, and patriotic effort. So that led to my heading off to MIT as an undergraduate and ultimately to getting this award in Boston, in fact staying in a hotel room with a window that looks out at the Great Courtyard at MIT that is quite a heady experience. It's a long way; a long circle that I feel I have completed.
PH - It probably looks a little bit different. You are higher and looking at a well-developed town, now.
CF - Yes, the developments in the Back Bay area in which we are now are incredible; back then, it was much more residential. To get back to MIT, I actually studied chemical engineering after trying a couple of other things. I was in biology for awhile. I always liked physics but there were too many physicists. I have never been a joiner, so to speak, and there were a lot of people going into physics. So finally chemical engineering was a nice major that provided a little more intimate professor/student contact and had some really great faculty who taught me thermodynamics and fluid mechanics and heat transfer and things I actually have used later to good benefit. Physics I always really liked so when I went to graduate school in '63, I picked UC Berkeley, a place where chemical engineering and at least physical chemistry were basically in the same college. Berkeley has this really odd structure left from G. N. Lewis in which both chemical engineering and chemistry are in the same college. In fact two departments make up a college. It is an incredible system for them. I was thus able to slide rather easily after a Masters degree in chemical engineering into what turned out to be a Doctorate in chemical physics, essentially, with David A. Shirley. So that is how I really got to Berkeley, plus the fact that I never really quite enjoyed Boston winters!
PH - You like to look at in the Fall but not in the Winter?
CF - Yes. We are at the Sheraton Boston for this interview and I lived on the Boston side and we had to walk across the Harvard bridge every day, twice at least if not four times, and in the winter it was deadly. At Berkeley, Dave was as much a physicist as a chemist always. He was doing Mossbauer spectroscopy and perturbed angular correlation measurements in nuclear decays on different compounds. And just at that time, a young Swedish post doc named Stig Hagstrom showed up from Uppsala and he was one of the first people finishing in the Siegbahn group in Uppsala doing electron photoelectron spectroscopy, or ESCA as they had begun to call it.
PH - What year was that?
CF - 1965! That is when I switched from chemical engineering to chemistry and Stig must have shown up in '66 or so because we wrote the first papers in '67 on x-ray photoelectron spectroscopy, or XPS as we then began to call it.
PH - Did Stig actually bring XPS over to you or did you already have an instrument?
CF - That was another just plain lucky thing. But I think looking at it particularly in the present context, it was also just one of those beautiful things about having a national laboratory. Because Berkeley had enough money to buy essentially a duplicate of an iron-free magnetic spectrometer that Siegbahn had designed for beta spectroscopy. They just shipped this thing across the sea and it sat in a custom-built non-magnetic wooden building because the spectrometer could not have any magnetic material around to disturb its field profile. They had tuned it up so that it was doing really nice beta spectroscopy. But right about then interest was rapidly decreasing in beta spectroscopy. So the nuclear people were going away and Stig came to Dave and said "all we need is an x-ray tube and a Geiger counter". I designed the x-ray tube. That was the first piece of equipment I ever built.
PH - Do you keep it in your library?
CF - No, the tube got lost along the way, but I have pictures of it. I do however still have the first detector, the first channeltron-based detector, another early design project. Anyway, the x-ray tube was my first big design project and like a lot of us, I have been designing gadgets ever since. I love it. So we outfitted this beta spectrometer for XPS and began to do chemical shift measurements first of all.
PH - So this was on solids or on gases?
CF - On solids, in the beginning, powders that were ground up then brushed on double-stick tape. That was the first standard holder. Then I designed a little stick that slid up and down inside an o-ring that gave you three samples for every loading of things. A big advance!
PH - What vacuum were you running with at that time?
CF - Around 10-6 torr, I think, it was pretty good.
PH - It was pretty good but not what you would call overly high.
CF - No, no, about 10,000 times worse than today's vacuums! So we did some measurements on a series of iodine compounds and europium compounds, things that were familiar to Dave from Mossbauer work. Iodine was especially of interest because it has this big range of formal oxidation states from minus 1 to plus 7, so you have chemically-shifted peaks all over the place. Europium also has a fairly broad oxidation-state range.
PH - It has plus 3 and plus 4 states as well.
CF - We tried to look at europium metal but we couldn't look at the metal because of oxidation; anyway, we saw some shifts. Dave said "I think you could interpret these probably in terms of some sort of Madelung model and energy cycle model." So, I went off and read Born and Huang, a classic book on ionic solids or something like that and figured out what all was involved in the Madelung sums, and thus was born, for us anyway, this so called potential model. Are you familiar with that one?
PH - I am not familiar with that one.
CF - It was a really simple idea where if you remove charge in forming a partially ionic chemical bond, you form an ion. The ion left behind you can describe quantum mechanically, and the other charges you create on neighboring atoms you just describe as a point charge array. So it just conserves charge and treats all the surrounding atoms as point charges.
PH - And you come up with the potential model?
CF - It is really important to conserve charge, because if you just take charge off an atom to infinity you get these huge chemical shifts and that is not what everyone was seeing. Of course you are not moving it that far. You are moving it, roughly speaking, to a little sphere of neighbors around a given site. The point charge correction, a Madelung sort of sum, has the opposite sign, and is smaller. So that was our first paper, in Science actually, on europium and iodine compounds and this potential model.
PH - Did you have a lot of communications with Siegbahn's group?
CF - We never wrote any papers together. We were of course watching what they were doing and they were watching us. Kai Siegbahn came on sabbatical to Berkeley while I was a student but after we published the first couple papers. Then they published the first of these books the ESCA-1; I forget what it was called. They published two books with a title of ESCA. That was in '68 when it actually appeared and then there was another one about '70 on gases. We certainly knew what they were doing, but after the beginning exploration, we went in sort of different directions. I got involved in magnetism. Dave was interested in magnetic materials so he felt there might be some magnetic effect in a magnetic atom.
PH - That must have come from his Mossbauer background? Is that true or not?
CF - Yes, I think it came from thinking of Mossbauer. In Mossbauer, you see various effects, but an important one is that you have a hyperfine field that is caused by the spatial separation of the up-spin down-spin electrons. The only way you can explain that theoretically is to do spin-polarized atomic-orbital calculations. So Art Freeman, who was another very important early mentor of mine, was doing these things, these wonderful electronic structure calculations for a long time. If you look into those from the point of view of how the orbitals split in space based on spin, and if you have a net spin up on a magnetic atom, the spin up orbitals can get you closer to the nucleus. Just the result of a nice net attractive exchange when the spin downs are a little further away. In addition, you then get these enormous fields that the Mossbauer effect can measure. But for us it was more what is the energy difference between the spin up and spin down electrons in any one of the core orbitals. That is what Dave had noticed in some theoretical calculations and it turned finally into one of the big chapters in my thesis. We were looking at multiplet energy splittings in manganese, and in fact my group is still looking at manganese compounds. Got in a real rut, but it's a nice rut1
PH - There are always new things to learn, new things to discover, just like you illustrated in your talk today.
CF - It took us a while to figure out just now to look at those splittings and we were really helped by Art Freeman, Jeff Morrow, one of his students, and Paul Bagus at IBM, who knew a lot about multiplet splittings since they were atomic people. So it all came together nicely .
PH - So you learned a lot about theoretical modeling in physics. Did you learn that from formal coursework or just pick it up by self-study?
CF - It was a mix, I took the senior quantum mechanics course in physics at Berkeley, then I took just one semester of the graduate quantum mechanics and whatever else I know about quantum mechanics is from using it, reading about it, or teaching it . Because as you know teaching is one of the best ways to learn something. I took group theory as well in the Physics Department at Berkeley, which was very helpful in all of this, and statistical mechanics and a good deal of the normal complement of other grad. courses. Also, Dave is a very smart guy and he knew a lot of theoretical physics.
In any case, the multiplets represented the second sort of study that I started with in Berkeley. A third study was just looking at valence levels. We ended up doing 15 metals that we studied with XPS, just looking at the valence bands with an aluminum K-alpha source. We showed there was really good correlation with whatever was known about the density of states in these materials, at least roughly speaking. Which is still the way people view that type of measurement: as a rather direct way to measure valence-band densities of states. You will get a kick out of the way we prepared the samples. Because they were all metal foils, and undoubtedly oxidized on the surface, I built a little cell for them with an aluminum window and a tiny orifice for the electrons to escape from. Then we heated the hell out of them while blowing hydrogen on them to reduce the native oxides. Who knows what quality the surfaces really were but the O 1s signal went away and they were all polycrystals which is fine for studying the density of states. Overall, I was in the Shirley group for five years and it was a wonderful period.
PH - You obviously had some appreciation of the importance of the surfaces at that point. I understand that Siegbahn, in the literature, says that he tried to maintain that ESCA was a bulk technique for quite a long time, then finally came to the realization that it was very surface sensitive. Is that accurate or is that not true in your opinion.
CF - I don't know that aspect of it. I remember that early on they did some measurements, I think that must have been in the 1968 ESCA-1 book. It was dated '67 but didn't come out until '68. In there they have data from some Langmuir-Blodgett films that were used to estimate inelastic attenuation lengths, I am quite sure. By growing these with different numbers of layers using standard methods they could estimate the attenuation length, and looked at with hindsight they found a somewhat misleadingly long sensing depth: 80 Angstroms is the number I seem to recall. We didn't have much data on it but we somehow knew enough to be really careful about it for the metals.
PH - You may have been able to tell that from the spectral distributions, too.
CF - I think that people knew. If you look back in the classic Mott and Massey atomic scattering books, the elastic and inelastic scattering cross sections for atoms in the gas phase were known and I think there was some data relevant to the solid state that would have put us on to the surface sensitivity. So anyway, that is a brief summary of my Berkeley experience, except for the socio-political side. As As I noted this morning in my talk, it was a remarkable period to be at Berkeley.
PH - It was quite a time.
CF - I came in '63 and left in '70. So all of what happened in the 60's, I watched it expand, explode, and then sort of collapse or whatever it did.
PH - Then you moved to Europe?
CF - I went to Sweden for a year on a post doc with Stig Hagstrom at Chalmers Technical University in Gothenburg and did one of the first two papers, I guess, on single-crystal angular distribution measurements in XPS. I had had that in mind for some time as an interesting measurement, although without any detailed notion of what electron scattering would do to these patterns. I just thought it looked interesting. So we did some measurements on gold and almost simultaneously, they published it a little before ours, Siegbahn's group did something on sodium chloride. Both were single crystal specimens. The gold was epitaxial on mica, a typical way to grow that. Both of us saw these bumps and wiggles in the angular distributions. Then I sort of went off on that one to try and see what it could be used for.
PH - So did you have a good theoretical description of the diffraction effects?
CF - In the early days, no. The Uppsala guys had correctly said this is sort of like the Kikuchi bands in high energy electron diffraction. In our first papers, we would just trace out the bands where we knew the planes would be and where the bands crossed you got a peak so you knew you were kind of on the right track. But it was another five years or so before the first theoretical papers on that came out. Ansgar Liebsch first published a theory paper for emission from localized valence levels at low excitation energies. And then, Phil Woodruff and his group did a LEED-based theory, again focusing on lower energies. I can't remember the name of the theorist he worked with on low energy photoelectron diffraction (Ed. note: It was Brian Holland). As you expect, they took a lot of the ingredients from LEED theory, like propagation and scattering factors and inelastic scattering and all that. With my group in '76-77, we wrote the first programs to do this in a single scattering way and realized that at high energy, by contrast to the prior stuff at low energy, there would be this forward scattering, and that for some cases, multiple scattering would not be so important. Forward scattering has turned out to be a really nice trick, really nice to get bond directions and low index directions. Then that program evolved into a multiple scattering program by the late 80's that made it more accurate and more quantitative. Of course multiple scattering would take much longer to calculate, but it is better to do it that way for some cases, like scattering along lines of atoms.
PH - Was there much cross fertilization when you were in Sweden with the Uppsala group?
CF - No not much. I know and appreciate these guys a lot, particularly Ulrik Gelius who was one of the key people there. But they were working pretty heavily on gas phase things and never really did much, at that time in that generation anyway, on density of states kind of things or multiplet splittings and magnetism, they just weren't doing that.
PH - You haven't done very much in gases in your career have you?
CF - No, we did atomic manganese photoemission with Manfred Krause; do you know his name? He was one of the leading guys. In fact he and Tom Carlson had a group in Oak Ridge that really produced the first gas phase x-ray photoemission. I think perhaps before the Siegbahn group. By the mid 70's things were really up and rolling.
PH - So you went from Sweden to?
CF - Actually, I went to Africa for a year.
PH - I noticed that you went to Tanzania. How did that come about?
CF - I felt that at that point in my life, once I came back and got into a long-term career commitment in the U.S. that would be it for quite a while. I felt like I really wanted to see another piece of the world, in particular a part of the third world, and maybe do some bit of good. I don't know; I guess I did a little bit of good down there.
PH - So a little bit of Berkeley rubbed off on you?
CF - A little bit of Berkeley rubbed off, exactly. Africa was very much in the limelight in the '60s. There were Mandela and other people. Even a guy that some people have seemed to slip in general memory, but there was a guy in charge of Tanzania named Julius Nyerere, have you ever heard that name? He was probably one of the wisest leaders Africa has had since independence, because he was at that time trying to build a more egalitarian, quasi-socialist state in Tanzania. He was somewhat of a hero among the African leaders, and by '68-'69 had written some very good, very articulate books on how Africa should develop. So that was why Tanzania appealed to me, one of the reasons. I sent them a letter with my CV and some references. Low and behold I got a job offer, somewhat of a surprise, since I really thought I wouldn't get anything. So I lined up a job at the University of Hawaii, signed off the papers and made sure I had something to come back to, and then went to Tanzania for a year. It was a marvelous experience. I taught physics.
PH - Undergraduate or …..
CF - Undergraduate; they had a masters program but it was pretty small. So my first course was quantum mechanics for physics majors at the University of Dar es Salaam. It was really wonderful, I really enjoyed it. It was a good time in general in Africa. There were at that point in time a lot of countries in Africa that were about 10 years into independence and aid was flowing in from all over because the cold war was still cold…or do I mean hot. Everybody wanted to keep Africa from falling the other way. So the US was building a road, the Chinese were building a railway, the Swedes were building a chicken farm, East Germans and West Germans were building two different projects here and there. The Russians and Chinese were all around. Nyerere was a smart guy, he used everybody, and he didn't make anybody really mad. So it was a pretty encouraging outlook I think for Africa, especially that country. A lot of new development projects, little factories building things they use to have to import, processing things they use to have send to out as raw materials. So I really, really enjoyed it.
PH - Did you have any research program active or was it straight teaching.
CF - No, no strictly teaching and well gee, I had a great time just doing the undergraduate labs. I think I built a couple of them, one on ESR--electron spin resonance. I still have all the lab instructions; I rewrote the instructions while I was there. It was strictly a teaching experience but it was a good one.
PH - Then you came back to Hawaii?
CF - Then I came back to the University of Hawaii Chemistry Department. The University also, was in a good situation in the late 60's; it had grown a lot and had a pretty dynamic president. The Chemistry Department was just moving into a new building, and as part of the equipping of it, was able to negotiate enough money to buy one of the Hewlett-Packard 5950 spectrometers, at a 1972 price of $178k. That was a lot of money at the time. (Ed. note: This would inflate to $600-700k in 2005 dollars, still not a bad startup package.) In fact I think one of the permanent curses of surface and interface science is always being expensive.
PH - It has always required high dollar value equipment.
CF - So I thought "geez, I could live in paradise, I could have the first, it really was the first Cadillac or Mercedes, whatever you want to call it, of electron spectrometers then. So that is how I started my group, with that machine. That was where Ron Baird was the first Ph.D. to come out of the group and he did his work on angle resolved XPS for surface analysis, thinking about roughness effects and plasmons and all that.
PH - I think that is when you and I first meet. I think it was a Physical Electronic Conference. I presented a paper on surface roughness and you asked about it, but I was dealing with Auger electron spectroscopy rather then XPS. That was a while back.
CF - Yes, I remember that. Anyway, that got that aspect of our work started. Even before I got to Hawaii, we had noticed in Sweden that these angular dependent measurement, when you got the x-ray incidence angle very low, there was a significant increase in the intensity. There was a little peak that would come up just before the "sun sets" on the x-ray, just before you turn off the intensity all together. I didn't really know at that point what it was. However Burt Henke, at about the same time in '72 wrote a beautiful paper in Physical Review that explained it all and had all the x-ray optical theory in it. He was just next door in the physics building in Hawaii. So we became close collaborators and some of our first papers also then on angle-resolved stuff were trying to quantitate how total reflection could turn on surface sensitivity or enhance surface sensitively. It sounds kind of satisfying, I am sure you have seen it with your own work too, that 20 years later or something, I don't know what it was, Terence Jach at NIST and Jun Kawai in Japan both started to publish papers on total reflection as a technique for reducing the inelastic background in XPS spectra, which is a nice thing. And JEOL even now has a spectrometer that is specifically designed to operate in that mode.
PH - It is always interesting when some phenomenon that you understood and talked about early on really becomes established.
CF - Then came photoelectron diffraction in a more quantitative way. We started our by just measuring patterns above a single crystal, and interpreting them in terms of very simple Kikuchi theory. Ron Baird was involved in that too. But then the first real structural determination we ever did with x-ray photoelectron diffraction was oxygen on copper, c(2x2) oxygen on copper. We modeled these sorts of intensity patterns using a single scattering cluster program written initially by Shozo Kono, then a postdoc in the group. We actually got a decent hunk of a the final answer. Keith Mitchell, who is at Vancouver, much later on and with LEED and STM finally sorted out what that structure really is. There is a missing row structure in it that we couldn't assess with photoelectron diffraction. We had the right position of the oxygen: we knew that oxygen had to be down in a crack, coplanar or sort of collinear with these remaining things that were there. It was kind of nice and he appreciated that we had obtained a piece of the answer in our first XPD study too.
PH - A lot of time to answer those questions you have to use a lot of different analytical techniques to come at it a lot of different ways. So the scanning probe microscopy just verified a lot of that early work.
CF - Yes that is right. This revolution started in the 80's is still going on.
PH - You had done angular resolved measurements. You had done diffraction effects. What else did6 you study?
CF - In Hawaii, another major thing that I did, but haven't done since I left there, was angle-resolved valence-band spectroscopy, because very early on we noticed that the density of states things kind of fit. But when synchrotron radiation came along, it wasn't clear exactly what model was going to be appropriate to describe valence spectra. So we wrote some programs to simulate spectra in terms of direct transitions, and we began in '74 to do measurements at Stanford on the synchrotron. Some of the first ones were of the type in which we would just sit with a fixed emission direction from a copper crystal and vary the photon energy so that what you are really doing is moving the excitation point through the Brillouin zone. We didn't know that for sure at the time, though. In fact there were other ideas that the variations in spectra were density-of-states in origin or matrix elements or matrix-element-weighted densities of states. Dave Shirley's and my groups were sort of on opposite sides of some of this for a short time as we went through all this, although congenially. Then we wrote a computer program, which was probably the first of its kind, although there must be many around now, just to figure out where in the reduced Brillouin zone an electron has originated if you measure it in your spectrometer with a certain high wave vector k. Just a simple projection back into the reduced Brillouin Zone. Larry Wagner, a post doc from Bill Spicer's group, was the guy who put that program together for the first time. We showed for some synchrotron radiation data that we could explain how the peaks went up and down for copper as photon energy changed. Copper, as you may know, is sort of the gold standard of angle-resolved photoemission. If it doesn't work for Cu, it won't work for anything.
Then there was a question of what happens in XPS at much higher energies. We for a while thought that you could see direct-transition effects, for gold in particular, and published one paper saying that. It seemed like you could see that you were actually exploring different parts of the Brillouin zone. But around the same time, Nigel Shevchik from Stonybrook, a very smart guy that we unfortunately lost rather early to a youthful death, pointed out that at these high energies you really have to worry about phonon effects. That is, when the electron leaves, the lattice in a sense recoils, you create phonons, and you smear out what you think is a definite thing and turn it into a fuzzy thing in the Brillouin zone. This smearing is directly related to the Debye-Waller factor which is familiar from x-ray diffraction theory, and it turned out to be very important at XPS energies around a kilovolt or so. Then with Zahid Hussain for his graduate work, we thought 'lets look around for a material that is really rigid', where the phonons are pretty quiet. Tungsten was the case we chose. We were able to show that you could raise the temperature of tungsten and turn on the density of states if you went up in temperature, or turn it off and turn on direct transitions if you go down in temperature. That was really very nice.
PH - You were using the synchrotron at Stanford and Bill Spicer was also using that a lot, with a lot people working with him.
CF - Dave was also using it, too. Dave Shirley must have started in '72 or '73.
PH - How much overlap did you have between these other groups?
CF - We shared in what they were doing, Dave especially. His stuff was more related to ours. Bill was working on semiconductors a lot, looking at surface adsorption and reaction systems. Dave's and my group were the closest, so that was the closest interaction. He was doing photoelectron diffraction also but in another mode. He always did it by scanning energy; we always did it by scanning angles, so we sort of kept from competing too directly, although we did have some differences of opinion on theoretical interpretation early on.
PH - Maintained congeniality. How much work did you do on your ESCAs in Hawaii versus on the synchrotron in Stanford?
CF - Well, in Hawaii, 80% of the experimental work I guess in 20 years would have been done in Hawaii. Later on I got a VG ESCA lab. I still have both of those machines and they both still run, actually. So there were two spectrometers there for angle resolved stuff. A little further along in Hawaii, there was this herculean experiment, one of those for which, when you look back you think 'geez, were we crazy enough to do that?' But Boris Sinkovic and Brent Hermsmeier, two really superb students, were given the task of using an yttrium M-zeta X-ray source to get the photon energy down to about 190 eV, so that spin-dependent scattering effects associated with multiplets get strong enough that you can seen them . If you go to 1 keV, the spin-dependent scattering effects really fall off. So whatever is going on is just masked. They put together a suitable yttrium electron target and had to use these very thin windows on the x-ray tube, because 190 eV x-rays don't travel through much. These are soft x-rays, with which I now work with as a fulltime business. There was also a special little liquid helium transfer line because we needed to cool below the Neel temperature of the antiferromagnet. The helium had to be ordered from the mainland. By the time you got it, I think you had lost 1/3 or ½ of the tank because it came out on a boat.
PH - You had to really want to do this experiment.
CF - Yes, you had to want to do this experiment. But that was really a nice experience because it showed something was going on way above the Neel temperature in two antiferromagnets, KMnF3 and MnO. Because we would see the spin up/spin down peak ratio just make a rather sharp change in it. The only thing it could be is spin-dependent things because the two peaks are really close together in kinetic energy, so there is no other funny business that could have happened and there are no structural transformations happening in these temperature regimes. So really how I got into magnetism, per se, was through that. Then, lets see, later on, or maybe simultaneously, we had done things like trying to see whether the multiplet splitting in a ferromagnet or the valence-band densities of states changes as you go above the Curie temperature. At least with the resolutions of that day, we didn't see anything, so I don't know if we ever even published it because nothing happened. Which was in some sense proof positive that the local moment in iron doesn't really change much when you go through the Curie temperature. They are there but they are sort of flopping around, they are randomized. I am sure that by now, people must have looked just at the temperature dependence of the valence band with higher resolution and probably see something, I don't know, but at the time we just did not see anything. So I got into magnetism sort of early on. In fact, one of my most cherished papers was with Peter Wohlfarth, a very nice man who was a professor at Imperial College when I was a postdoc in Sweden in '70-'71. I don't know how we got in contact. But he wrote me and said 'why don't we write a paper together'. So we wrote one of these Comments on Solid State Physics. It was a special journal, publishing not necessarily new results, but just comments with hopefully some wisdom in them. We wrote on what changes, and does not change, in the ferromagnets as you go through the Curie temperature.
PH - So you were really into magnetism early on. An enduring love with your work I see.
CF - Then the more normal type of photoelectron diffraction, for a while, went in different directions, as for example, adsorbate structures. Both adsorbate structures and the valence band studies went on through my first sabbatical in '78-79 in Paris and then another sabbatical in '86-'87 in Paris with both being at LURE, the synchrotron source there. Yves Petroff and Jean Lecante were my hosts there and really great guys that I worked with. From there we were doing a mixture of these two things: valence band studies and looking for spin-dependent magnetic things that would change at high temperature. I think by then we were also thinking about holography, but it did not come along in a real way until the early 90's. We were trying to look for other short-range order changes in this multiplet splitting as a monitor of short-range magnetic order. For one reason or another we never got much data that was convincing. It was a tough experiment. We wanted to look at iron. Iron as you heat it bubbles out this nasty stuff, with sulfur and other things coming to the surface. Also if you go a little too high above the Curie temperature it changes phase: bcc to fcc. It is a classic monster. Sometimes you destroy the crystal if you go just a little too high. So that is kind of the Hawaii experience, two sabbaticals in Paris and actually in '80-'81 I went to the University of Utah for a year. I thought of moving there. I don't think I mentioned that in the bio did I?
PH - To the department of physics?
CF - A joint appointment with physics and chemistry. They made a very nice offer and it is a good university and I thought very seriously about it but decided to go back to Hawaii.
PH - Did they have a spectrometer there?
CF - They were buying one. They would have bought me one and set up a new lab.
PH - But you couldn't turn down Hawaii?
CF - It was hard; it was hard. I am not a skier. If I had been a fanatic skier, it would have been paradise. But I don't like the snow much.
PH - So how did the transition between Hawaii to Berkeley and the Lawrence Berkeley National Laboratory (LBNL) and the Advanced Light Source (ALS) happen?
CF - The ALS was being put together and I had actually been involved on a couple of evaluation committees for different projects and what should be built when and how it should be built. That sort of got my feet more and more into the synchrotron radiation game. Dave Shirley, when he put this project together, said 'why don't we promote some joint appointment between different campuses of the UC system and the Lab., and call them ALS Professorships, to promote interactions between the ALS and the campuses. I am pretty sure he said campuses, plural, because it could have been just Berkeley, but I don't think he did. I know he didn't, because Davis certainly has two now, Steve Cramer and I. UCSF has one, Carolyn Larabell. Finally there are three, but early on I think Berkeley was thinking of creating another. Oh yeah, and Riverside had one person who for a while who was an ALS Professor. I am biassed, of course, but I think it was a great idea. Because half of my salary is paid by LBNL, and therefore I only teach a halftime load. I love teaching but a halftime load is really just about right in terms of keeping up with research and everything else.
PH - So what is that, one course per semester?
CF - One course per year if we were on semesters, but one and a half courses per year since we're on quarters. That immediately looked attractive to me. I am not sure I need to give you any of these intimate details of teaching requirements or anything. I got to California and I was beginning to realize it was getting a bit tiresome hauling a manipulator from Hawaii to California and being a long distance user in the early days. I think it is easier now because the facilities have more support personnel and they realize that they have to host people and the experiments have to get done right. But in the early days, it was Piero Pianetta, a poor graduate student of Ingolf Lindau's at Stanford, who helped us with our first experiments, getting up in the middle of the night and all that. By '90, undoubtedly it was better at Stanford, but still it was a pain and I always enjoyed living in Berkeley or the Bay Area. So I thought 'hey let's try this' and it worked out. So here I am.
PH - So do you have a spectrometer at Davis or do you have all of your equipment at LBL?
CF - It is a mixture. I have two HP ESCAs at Davis. One was donated to UC Davis by IBM, through a former student, Brent Hermsmeier. It was essentially a shining virgin instrument that IBM hadn't used much. We use that for service things for the campus. The other one I brought from Hawaii, the original instrument that they bought for me, and I think it has had one PhD thesis done on it in California, but we don't use it much anymore. They are both still functional. Then there is another system I brought from Hawaii, a VG ESCA Lab that has a monochromater and a bunch of other stuff on it, that is kind of a stand alone instrument. That is in Berkeley. The main thing that we use, the main focus of my attention, is the machine that I flashed up today in my talk. That was the biggie that I went there to build, once again in collaboration with Zahid Hussain, every year we add some other little thing on it. A lot of the time all the stuff works at the same time. That is always an issue when you add one tool, the probability of success is so many percent, and you multiply them.
PH - So did holography become important after you came back?
CF - No, it was a little before that, in '86 or '87 I got this paper by Abraham Szoke at LLNL, and thought that is an interesting idea. So before I left Hawaii actually, we had done one experiment on a semiconductor, germanium, did the transforms and we could see atoms in the lattice and knew that it was interesting. We had begun to write, in fact, the programs we still use. They were written by Theva Thevuthasan, who is now at the Pacific Northwest National Lab. So he was really the genesis of all that effort to see what we could do with holography. Then in California we did early on this little thing on c(2x2) sulfur on nickel that I showed this morning. Gee, that looked pretty interesting.
Roughly simultaneously, a student named Pat Len at Davis began working with Gerd Materlik, an x-ray physics professor from Hamburg, who had an idea for turning the holography experiment backwards with x-rays. At that time, maybe because we were a little better able to model it theoretically, he got in touch with us. So they took the data and we analyzed it and there is this nice paper, which is really the first demonstration you could do this "inverse" x-ray holography. Then along the way we theoretically explored a few ways to make it better or fancier, including this spin polarized variant that I mentioned at the end of my talk. (I should mention here that Michel Van Hove became a key collaborator on theoretical modeling once I moved to California.) You measure a spin up hologram and a spin down hologram and take the difference or so you can somehow differentiate the spins in the scattering. That is still an experiment that I want to do or someone should do. I still would like to do that one, but it is just tricky enough and you need such high quality data that it is sort of waiting for a better spectrometer plus better, faster detectors. We may be able to do it with this new detector that we are just putting in our machine. Ultimately it could be a pretty nice tool to look at short-range magnetic order. For example, high Tc materials are thought to have antiferromagnetic things in them and that antiferromagnetism is very important. It will be tricky to do it on copper with only one unpaired spin. It is maybe not a next generation experiment but the one after that. But for other systems with higher moments, a lot of magnetic or real magnetic materials or strong antiferromagnets, it should be interesting.
PH - When did you get interested and start using the standing wave techniques?
CF - That one I actually had in a proposal to ONR in '91, right after I got to Berkeley, for using either total reflection or standing waves in this mode and I did have a lot more money at the time to try things, for a while. But we never got to it. It was somewhere there on the list. It wasn't really until about 6 years ago, when the combination of postdoc See-Hun Yang and grad student Simon Mun, plus the guys in the Berkeley Center for X-ray Optics permitted it to really got serious. Jeff Kortright at LBNL also helped, he is an x-ray optics expert and had done some standing wave stuff of a different kind before, with absorption measurements. We got together and finally it congealed and finally we saw these first sort of jiggly oscillations and it has just been getting better ever since. It may seem at first sight like a kind of complex or convoluted way to make your samples but we do a lot of complicated things in science. If this particular trick will give you enough extra information on the depth and is worth the effort, then people will develop it. So I look forward to working with other groups. I hope to explore the depth profiling stuff and other types of systems, not just magnetic things.
PH - It certainly looked like really excellent data. The signal-to-noise that you had on the data was quite good.
CF - Yeah, that was the best data set we have gotten, but you know how it is, I am sure. If you get a good data set once you can somehow get it again. If you never get a good data set, then you have to ask, maybe this technique is too hard or something.
PH - That brings us pretty much up to date. What do you see for the future? Where do you think the field is going to move to now?
CF - Well, it depends on how you define the field. I certainly know where my own directions will go. I think that I will try exploring more ways to use the standing wave effect and other things. Even solution chemistry at surfaces or whatever, is something I am trying to promote. I think this faster detector or other faster detectors that come along will permit us to do more measurements of full angular distributions of things and pursue a lot of directions in holography that we haven't before, including spin dependent things. Using the multiple spectroscopies, including photoemission, x-ray emission, and x-ray absorption, on more strongly correlated materials like the manganites or other things, even high Tc perhaps, would be interesting. Although there are a lot of people in high Tc, and I don't want to be just one more guy. If we have a special trick, and I think we may have a few, I may look at those. X-ray holography, too, is still sitting there somewhere as something I would like to pursue. We have a collaboration with a group in Hungary that still exists. Again that is an experiment where the detectors are still just a little bit slow to really do what you want, but I think that is a nice future direction. Another very interesting new direction for photoemission is doing it with hard x-rays, in the 5 to 15 keV range, and we are going to do some experiments of this type in June in Grenoble. You are never going to get the extremely high resolutions in the meV range that you get with 20-100eV excitation, but the groups already doing this in Europe and Japan can get down to 70 milivolts resolution at 6 kV, so it is nothing to sneeze at. I am sure that, as a more bulk sensitive material probe, it has lots to say for it. In fact dichroism will still be there as well, something I haven't even mentioned until now in this interview. This is a kind of special little thing, that is probably in Auger as well, and it can provide information on magnetic systems. There is also a famous thing called the Sudden Approximation, that we have always used to interpret photoemission, but whose validity one can question at lower energies. Well you are sure as hell at the sudden limit when you get to 10 kilovolts, so some of the theory will get a little cleaner, I think, as well, making interpreting the data easier.
PH - So there are still lots of opportunities.
CF - If you just look at only photoemission and synchrotron radiation, there are other things like focusing the spot down to 10 nanometer size, doing what they call nanophotoemission or nano angle-resolved photoemission (NanoARPES). I am not involved in that at all but there is some really beautiful work now from Eli Rotenberg et al. at Berkeley, where they scan the spot around on a surface then they can do band mapping within various nanocrystalline regions. They see these nice quasi-parabolic bands, or whatever distorted things there are in the actual band structure. Time- resolved things represent another big area that I have not touched yet but should be most interesting, where the x-ray pulses get down into the femtosecond regime so you can pump and look, pump and look. Or even at much longer times scales watching chemical reactions at surfaces that are much slower processes. If you have microseconds you might be able to watch some sorts of surface processes. That kind of thing we may do with this high-speed detector that I mentioned in my talk today. High-pressure photoemission--there are a couple of projects, 2 or 3 in the world I guess. One in Berkeley, where I was a little involved at the beginning of it but Miquel Salmeron is the main guy. They have a thin window between the finely focused x-ray beam and the sample and small orifice. The sample is in a chamber with a small orifice for the electrons to pass into a lens. The lens focuses the electrons back on another small hole and you pump like crazy in the intermediate region, and then you have another focusing lens and pump like crazy in two places. So you can end up getting like 10 Torr around the sample and still do photoemission. They have already done some really interesting work with that, where they watch reactions on surfaces at high-pressure, trying to bridge the pressure gap.
PH - Right, that has always been a problem.
CF - So like I said at the beginning of my talk today, I thank Einstein for giving us the photoelectric effect. It has provided us and will provide us with a lot of wonderful information.
PH - That is great. Anything else you would like to share with us today?
CF - Oh, I don't know, I think overall I would say that I just feel very fortunate. You like to think that you could tell any young person, which I would tell them, decide on your goal and just work hard. If it is something you want to do you will get there. At the same time I had a wonderfully supportive set of parents and a family that supports me very well and I didn't choose any one of those, that was just coincidence The fact that Stig Hagstrom showed up a Berkeley right at the time when I was about to start on a Ph.D., is purely a coincidence. I had actually started another Ph.D. project on nuclear something, but I wasn't very far into so it was easy to switch, and Dave Shirley was adventurous enough to jump on it. That was luck.
PH - There is certainly is some element of luck in a career.
CF - Both of us, I think, were also lucky in the sense that surface science was just a baby and taking off at that time.
PH - Exciting times.
CF - It was exciting then, it is exciting now. As an addition I have always enjoyed this particular organization. AVS, beyond just dealing with excellent and exciting things that span basic science and technology, is an especially congenial organization
PH - Well, Chuck thanks very much for spending some time with me today.
CF - Thank you for taking your time. It's been a real pleasure.