Awards > Awardee Interviews > Interview

Interview: Sven Tougaard

2012 Albert Nerken Award Recipient
Interviewed by Paul Holloway, October 30, 2012

HOLLOWAY:  Good afternoon. My name is Paul Holloway. I’m a member of the AVS History Committee. Today is Tuesday, the 30th of October 2012. We’re at the 59th International Symposium of the AVS in Tampa, Florida. I have the pleasure of interviewing Dr. Sven Tougaard from the University of Southern Denmark. He is the 2012 Albert Nerken Award winner. His citation reads, “For the development of advanced methods to characterize thin-film nanostructures by X-ray photoelectron spectroscopy.” So congratulations on the award, Sven. Well deserved.
TOUGAARD:  Thank you very much, Paul.
HOLLOWAY:  Could you begin by giving us your date and place of birth?
TOUGAARD:  I was born on December 17, 1950 in Aalborg in Denmark.
HOLLOWAY:  Well, you’re just a young fellow.  [Laughter]

TOUGAARD:  Okay. I’m not that young anymore!  [Laughing]
HOLLOWAY:  What about your educational background?
TOUGAARD:  Well, I did my studies at the Technical University in Copenhagen. I was studying electrical engineering. For my master’s, I then did solid-state physics, semiconductor physics. For the master thesis, I studied photo-sensitization of zinc oxide.
HOLLOWAY:  Is that right?
TOUGAARD:  So that was my first experience in surface physics. I did that in air and the idea is that you put a  dye on the zinc oxide. This would introduce interface states, which made it possible to excite the electrons with the visible light into the conduction band of the zinc oxide.
HOLLOWAY:  What year was this?
TOUGAARD:  That would be in 1974, 1975. January 1975 I finished my master’s.
HOLLOWAY:  What year did you finish your bachelor’s?
TOUGAARD:  Bachelor, we didn't really have the bachelor degree at that time.
HOLLOWAY:  Well, your first degree.
TOUGAARD:  Well yeah, we just went straight through.
HOLLOWAY:  You went straight through to the combined degree.

TOUGAARD:  Yeah. We had a five-year… That came in later with a bachelor’s.
HOLLOWAY:  The zinc oxide that you were working with, was it powder? What was its’  form?
TOUGAARD:  That was single crystal zinc oxide I was working with at the time.
HOLLOWAY:  How did they produce zinc oxide single crystal?
TOUGAARD:  Well, they had actually hired a person who was skilled in doing that at this institute where I did the thesis work. I don't know exactly how he did it, but I know this was the reason why they hired this guy.
HOLLOWAY:  Remarkable.
TOUGAARD:  I think it was difficult to produce at that time.
HOLLOWAY:  It’s easy to produce now, but reproducibly making the same material with the same qualities is still not so easy to do.  [Right.]  So you went on for your Ph.D. then?
TOUGAARD:  Yes. Then I got a grant to do my Ph.D. in Odense University, which is nearby Copenhagen. That was in ’75. It was a young university. It was just building up things. So the Physics Department was only a couple of years old, so there was no strong structure in the research environment yet. At the same time, this was shortly after the student revolution, which was pretty much complete in Denmark. So  the full professor and the graduate student had the same vote at the University at that time.
HOLLOWAY:  Is that right?  [Yes.]  That is quite a revolution.
TOUGAARD:  Quite, yes. So a very flat structure. At the same time as I mentioned, they wanted to build up the research structure. So I was pretty much on my own a lot of the time. I was supposed to start doing laser physics. My supervisor left for a sabbatical in Paris shortly after I came there. But I had this old interest in the zinc oxide and the  surface and interface states. So I started to study papers by  several groups  working on a  quantum mechanical description of the surface states and interface states at that time.
HOLLOWAY:  So would you care to mention some of the mentors and professors that you had at that time?
TOUGAARD:  Well, what happened was that I was pretty much on my own for this work  with the studies of the quantum mechanical stuff there. But at the same time, there was a group interested in building up surface physics at the Institute. That was Jens Onsgaard  who was the professor. I had this experience with the zinc oxide, so I sort of joined the group and we started to look at zinc oxide. This was powdered zinc oxide and we were looking at the absorption, and photo-stimulated desorption of oxygen and the influence on the conductivity.
HOLLOWAY:  This was all well after Siegbahn had received his Nobel Prize. Is that true?

TOUGAARD:  No. This was before then. But  we didn't have any XPS in Denmark at the time.
HOLLOWAY:  Were you aware of it but just didn't have the equipment?

TOUGAARD:  I was aware of it by reading papers. But this was like in ’76, around that time. So I was aware of it, but we were just building up a chamber with  a mass spectrometer and then shooting with photons  on the zinc oxide to look at the oxygen and CO2 response on the mass spectrometer and measuring the conductivity of the zinc oxide. It was basic stuff.
HOLLOWAY:  So you observed photo-induced conductivity.
TOUGAARD:  Exactly. Yes.
HOLLOWAY:  From the defect generation?
TOUGAARD:  Yes, yes, and the loss of oxygen. The band bending  mechanism at the surface because  the negatively charged oxygen on the surface would bend the bands of the semiconductor and change the surface conductivity.
HOLLOWAY:  So how did you go from there to…?
TOUGAARD:  Well, from there what happened was that there was a professor from the University of Houston, Alex Ignatiev . He was on sabbatical at Aarhus  University in Denmark at that time. I met him a couple of times at meetings, and he asked me if I was interested in doing a post-doc with him for a year. So I did that right after  the Ph.D.. I was doing LEED, low energy electron diffraction, and also some reflection electron energy loss spectroscopy with him.
HOLLOWAY:  Did you do any Auger with retarding fields?
TOUGAARD:  Oh yes, yes. Actually just before I left for this post-doc position, we bought a Varian system in Odense with a LEED and retarding field analyzer for Auger and energy loss spectroscopy. I was working with a similar system in Houston.
HOLLOWAY:  What year was this now?

TOUGAARD:  That would be in ’78, In ’79 I went to Houston.

HOLLOWAY:  How long did you stay there?
TOUGAARD:  I stayed there for one year initially, and then came back to Denmark. But I went back several times over the next few  years to Houston on shorter stays.
HOLLOWAY:  These were post-doc visits or pre-doc visits.
TOUGAARD:  These were post-doc visits. And  what happened was that when I came back from Houston to Denmark, I got a grant from the Danish government. This was a very nice grant. It was a three-year grant, so my salary would be secured for three years without any obligations. So I had complete freedom to do whatever I wanted to do.
HOLLOWAY:  Nice! Wow. Unheard of!
TOUGAARD:  Yes. Nowadays you don't get these things.
HOLLOWAY:  Nothing comes free nowadays.

TOUGAARD:  They really believed in the freedom of research at that time.
HOLLOWAY  Freedom of exploration.
TOUGAARD:  Exactly, yes. Yes. So I got interested in many different things at that time. Another thing that happened while I was in Houston was that a professor had been hired at  Odense University. This was Peter Sigmund who is doing  theories for ion-sputtering and ion mixing and these kinds of things. He changed the environment a lot at the University because he started inviting a lot of people. Many American professors came by on short term and long-term visits. So that was very nice for me with all the inspiration I could get all of a sudden from speaking to a lot of people coming in.
HOLLOWAY:  Your position was still one of a grad student. Is that correct?
TOUGAARD:  No. I had  this  senior post-doc position with the three-year grant that I mentioned.
HOLLOWAY:  Grant support.  [Yes.]  Did you have any teaching responsibility or it was all research responsibility?
TOUGAARD:  That was very little, only what I wanted to do. They could ask me to do some teaching , but this would be like teaching within surface physics like what I was doing research in.
HOLLOWAY:  Were there other people that you worked with besides Sigmund?
TOUGAARD:  Yes. There was in particular a long-time visitor from IBM, Harold Winters.
HOLLOWAY:  Oh, Harold Winters.
TOUGAARD:  You know him, yes.
HOLLOWAY:  I know him very well.
TOUGAARD:  Yeah, yeah. He is a nice guy.  [Yes.]  He was there in Odense right when I came back from Houston. I worked with him in the lab and had a lot of discussions with him. He is a very good solid physicist, a very thoughtful physicist.
HOLLOWAY:  Yeah. He and John Coburn made a good pair in interaction of energetic particles, atoms, and molecules with solids.  [Yes.]  So you did have some high-power people to pick their brains.
TOUGAARD:  Yes. Another person was Rufus Ritchie from Oak Ridge National Laboratory.
HOLLOWAY:  I’m not so familiar with him.
TOUGAARD:  He did some fundamental work on electron transport- made  some theories for that. That meant a lot to me, discussing with him on the  applications of the Landau formula and  dielectric response theory. He introduced me to using this for the electron transport in solids, which I used. I was very much getting involved in the development of XPS data analysis, trying to understand the energy loss processes there.
HOLLOWAY:  Now why did you go from defect generation in zinc oxide to electron transport during XPS analysis?
TOUGAARD:  Yeah, this happened like… One day I was giving a seminar at the Institute after I came back from Houston. I had been doing some XPS there, and I was explaining what you could do with XPS. We had these peaks and  it was very surface sensitive. Peter Sigmund asked me afterwards,  “Why do you look at the peaks?” He would think in his mind that the main information would be not in the peaks but in the background, in the intensity increase  you  get after each peak  Probably because he had a background in  the interaction of solids with ions,  and the standard method there for  characterization  is Rutherford r backscattering where you have  no peaks but only the background increase. So he pointed that out to me, and said, “There must be something in there which is coming from deeper down in the solid.” So after that, I started to look at this , and I came up with a small model for that. Peter Sigmund liked it very much.  But I never published this because Peter  suggested to me that we do something together in more detail. Then we looked at this and set up  models for elastic and inelastic scattering of electrons in solids which we published about a year later- in ´82.
HOLLOWAY:  So that remark by Peter Sigmund set you off in a new direction then.
TOUGAARD:  Yes. That remark was what inspired me to start looking at this. Yes, exactly.
HOLLOWAY:  And you’ve done an excellent job in bringing that to a quantitative state that can be used to interpret the distribution with depth of elements in solid. Is that accurate?
TOUGAARD:  Well, I don't know.  [Laughter]  To some extent.
HOLLOWAY:  And you call that what? Your name for your procedure.
TOUGAARD:  Well, I don't know. I call it peak shape analysis. Some people call it  Tougaard Method.
HOLLOWAY:  I call it the Tougaard Method.  [Laughter]  When I’m talking to my students and explaining to them how to improve their data, that’s what I call it. So tell us a little bit about that, what’s involved?
TOUGAARD:  In the analysis? The idea is that the shape of the spectrum changes a lot with the depth distribution of the atoms, because the deeper the photon excited electrons  come from, the more you have of inelastically scattered electrons. Then by analyzing the shape of the spectra, you can find out, so to speak, the history of the electron, what it has experienced after it was excited because the inelastically scattered electrons  will follow like a trail after it.
HOLLOWAY:  Now this is not angle resolved XPS, right?
TOUGAARD:  This is not angle resolved XPS. You just look in one direction and the electrons lose energy on the way out. The longer they have traveled, the larger fraction of the peak electrons have lost energy and they are then found at larger and larger energy loss behind the primary peak. So it’s just one angle.
HOLLOWAY:  So it’s remarkably stable then to try to do the analysis of those data in contrast to the angle resolve which we’ve tried to do. The noise gets to be a very significant factor in the angle resolved data resolution of depth.
TOUGAARD:  Yes, that is right. One problem with the angle resolved XPS as I see it is that if you don't have completely flat surfaces, you have some shadowing effects which are hard to correct for.
HOLLOWAY:  That’s absolutely true. So you have formed a little company to sell this software. Is that accurate?

TOUGAARD:  That is correct. I call it QUASES. The reason why I did that was that around 1992 , we had pretty much the final theory, which is pretty much the same as we’re using today; The framework had been developed. I had expected that a lot of people would start immediately to use this. There was a lot of interest for the technique, and I was invited to give talks in many places. But apparently there was a barrier to start using the technique because people  had to make computer programs   and to set up the whole thing. After I realized that, I thought it was a good idea to make a software package that would enable other people to do this kind of analysis.
HOLLOWAY:  People tend to want to take something that they can plug into their computer and send the data to it and have something fall out that they find useful. 
TOUGAARD:  Exactly.
HOLLOWAY:  In your bio for the award, you talk about electron energy loss spectroscopy, reflected electron energy loss spectroscopy, REELS. Is that part of QUASES?
TOUGAARD:  This is not part of QUASES. This is a different package of software. In 1987 I had developed, together with a student Ib Chorkendorff  - he is now professor at the Technical University in Copenhagen - an algorithm to determine the cross section for energy loss of low energy electrons from analysis of measured REELS spectra. This was an important step for the development of the procedures for XPS, because I needed the inelastic cross section for a quantitative understanding of the backgrounds in XPS. For the same reason I had also worked on theoretical dielectric response models to calculate the cross section and I thought that combining the theory and experiment would be a good way to test the validity of the formalism. Then what happened is that around 1990  I had a graduate student, Francisco Yubero . I still have a lot of contact with him. He is in Spain now in Seville, and  we developed  a detailed dielectric response model  for reflection electron energy loss spectroscopy.
HOLLOWAY:  So do you have a software package that’s available for REELS as well as for…?
TOUGAARD:  Yes, we also made a software package for REELS called QUEELS. The thing there is that  you can determine the dielectric function and the optical properties of thin films from analysis of a REELS spectrum with this software.
HOLLOWAY:  Tell me a little bit about what types of problems you can apply this software to solve.
TOUGAARD:  What we have applied it for or…?
HOLLOWAY:  For example, a uniform layer on a surface versus a buried layer on a narrow surface. Can you distinguish between those two different conditions?
TOUGAARD:  Oh, this is very easy to see from analysis of the XPS spectra whether the atoms are on the surface as a layer or a little bit underneath the surface. You can also easily determine whether a layer covers the surface completely or whether it forms islands on the surface.
HOLLOWAY:  Now if I have a material with a thin layer of A buried in B, can I quantitate how deep A is in that material?
TOUGAARD:  Yes, you can. You can fairly easily do that. The resolution you would get that with depends on lambda. I mean everything scales with the inelastic mean free path,  lambda.
HOLLOWAY:  So do you recommend then working in electron spectroscopy with your QUASES with higher kinetic energy electrons?
TOUGAARD:  Yes. This is one of the things we are just starting to look at. I mean this should open up the possibility that we could look at structures that are very deep, like 100 nanometers or maybe even more if we’re using  synchrotron radiation with photon  energies of maybe 10 keV or more where lambda could be a factor of 5-10 times larger than with conventional XPS. We’re starting to look at  gate stacks.
HOLLOWAY:  So how complex a sample can I do an analysis on? For example, can I take an island sample, bury islands of material and tell that these are buried islands versus a buried layer?
TOUGAARD:  Well, I haven't thought about this. Let me see, let me see.
HOLLOWAY:  Not many samples realistically have that condition.
TOUGAARD:  Yeah. I was wondering  if we have looked at that — We can clearly   tell the difference  if we have it on the surface. Can we see it if it is buried…?  It must be possible also to identify that, but the accuracy with which we can tell if we have a buried layer or buried islands would depend on the size-distribution of the buried islands and how deep they are buried.
HOLLOWAY:  Where do you see surface analysis going in the future? You know, we talk about Rutherford backscattering versus XPS versus Auger versus ISS versus SIMS versus it’s really not appropriate. And—I should put “and” between each of those techniques. But right now it seems like XPS is becoming more dominant and Auger is less dominant, for example. Do you agree?
TOUGAARD:  Right. Yeah, yeah, I think this is what we’re seeing. I guess one of the strengths of Auger still is that you can do very small spot size analysis, which is still not possible with XPS. But we see developments now that you can start looking at sub-micron structures with XPS. Maybe this will continue in the future. But it’s difficult to get the high photon flux in very small spots except at the synchrotrons. That’s the problem with the synchrotrons. You cannot have synchrotrons in the labs everywhere. So for industry, XPS  is a routine tool. This can be a problem.
HOLLOWAY:  Synchrotrons don't come to you.
TOUGAARD:  Right. You come.
HOLLOWAY:  You come to the synchrotrons, and whenever they decide it’s time to run, you decide, “Well, it’s time to run,” even if it’s 3:00 A.M. in the morning. 
TOUGAARD:  Exactly. Another trend I see in the use of XPS is that it is more and more being done by non-experts and robust automated data analysis is therefore getting very important. I realized this around 2002 and then started to develop new algorithms for that. The detailed XPS analysis with QUASES is straight forward but some skilled operator interaction is needed. So I thought there would be a need for very simple algorithms that would of course be less accurate than the original algorithms but that would be very robust and therefore good for automated data analysis. I think these simpler algorithms could become very useful in the future. Especially in industry where speed and robustness of data analysis is often more important than getting a little better accuracy. We have also used these algorithms for 3D-XPS imaging where automated analysis is necessary because of the huge data sets you get there.
HOLLOWAY:  Tell me about your experience with the other analytical techniques besides XPS and Auger. Have you done much with SIMS?
TOUGAARD:  I have not done much with SIMS. I did some with ISS, and that was because I got involved in some studies of radiation  enhanced diffusion during one of my visits to the University of Houston. Then after that, in 1983 when my three-year position expired, I got a German Alexander von Humboldt grant and went to the University of Osnabrück in Germany   to do ion-surface-scattering together with Werner Heiland. He was one of the experts in the field at the time. I started to do that because I wanted to have very good surface sensitivity- to see only the surface layer. With XPS we would see a few layers. But I quickly gave up because it was not quantitative. It was impossible to do quantitative analysis with the ISS.
HOLLOWAY:  Now let me take us off in another slightly different direction. You can do dielectric constant determinations using electron spectroscopy.  [Yes.]  Can you apply that to thin oxides as in gate oxides and learn something useful about them?
TOUGAARD:  Oh yes, definitely. We have done recently together with Samsung Laboratories in Korea and Chungbuk University there some analysis of gate oxides and determined the dielectric functions for these from analysis of REELS for a range of materials there of the high-k gate  dielectrics.
HOLLOWAY:  Who did you work with at Chungbuk?
TOUGAARD:  This was…what was his name?

TOUGAARD:  Yes, Professor Kang.
HOLLOWAY:  What about other applications beyond gate oxides? What do you see the future of surface analysis being for high technology areas? Is it going to continue to grow, shrink, or remain constant?
TOUGAARD:  Well, I think it’s going to grow in strength. I mean it’s like we’re going to the nano business where the interesting part of the sample is only a few nanometers deep so you have—I mean what we see with XPS, the information depth we have  is the complete sample  in new nanostructure applications in industry.
HOLLOWAY:  You just mentioned that you had collaborations with Samsung in Korea.  [Yes.]  Did that come from students that came to Odense and Technical University?
TOUGAARD:  No, no. This came from meetings where I met  professor Kang from Chungbuk University both in Japan and at the AVS meetings. He and his students share some interest in this, and we started up a collaboration this way.
HOLLOWAY:  What about your collaborators in Japan? You have some there?
TOUGAARD:  Yes, I have some there. Dr. Tanuma and his group  at NRIM in particular. We’ve done several papers together also on the dielectric response theory. One thing that we’re applying this REELS analysis for is to determine the inelastic mean free path. That’s what he’s doing a lot together with Cedric Powell in the Tanuma, Powell and Penn dielectric model. So it’s difficult sometimes for some materials to find values for the dielectric function in certain energy ranges,  You may have it at low energies where you can use optical methods and very high energies where people have done maybe electron transmission experiments. But there is a range of the  medium energy loss where data is often  missing. So there this comes in very nicely and can fill in the gaps so they can get more accurate inelastic mean free paths.
HOLLOWAY:  Now all that goes towards quantitating the data from Auger or XPS. How accurate can we quantitate those data today? Is it a factor of 2 error or 50% error or 10% error or 1% error?
TOUGAARD:  I think it depends a lot on what you’re doing. If you just take the peak intensities or peak areas and use the ratios of these, then you’re probably easily off by a factor of 2. But if you do it more carefully and also analyze the details of the inelastic background  , probably down to better than 10%.
HOLLOWAY:  That’s an accuracy versus a systematic repeated value.
TOUGAARD:  Yes, yes. This will be an absolute accuracy. Yes.
HOLLOWAY:  Rather than precision.
TOUGAARD:  Right. 5-10%, I would say.
HOLLOWAY:  That’s remarkably good. I think that if you take corrected electron microprobe data that’s about as good as they can claim, too. Let me turn the conversation to Cedric Powell and ASTM E42. Have you worked with that group very much?
TOUGAARD:  Yes. I’ve interacted a lot over the years with Cedric Powell and the people involved in ASTM E42 . This interaction has also meant a lot to me as an inspiration to continue my work. From the early years when I started to attend the AVS meetings—that would be in maybe 1982 or 1983 the first time—I have continuously reported my data there. I’ve probably attended a little more than half of the AVS meetings. I always got a lot of interest and good response from people there. This has meant a lot to me to keep going in this direction seeing that people from industry and doing applied XPS could see a use of this.
HOLLOWAY:  Do any of the manufacturers include your software as standard part of their software package?
TOUGAARD: Well there are actually two parts of the algorithms. There is a simple algorithm which has become known as the “Tougaard background” and this is included in most software packages for XPS analysis. This algorithm is used just to get rid of the background. The other more involved algorithms are used to determine the nano-structure from analysis of the details of the background.  I don't think these algorithms are in the manufacturers’ software. I think Physical Electronics  has an earlier version of the software, some of the first algorithms I made. I think they have implemented that. Otherwise I think the manufacturers  recommend people to buy it as an add-on.
HOLLOWAY:  That’s probably better for you.  [Laughter]
TOUGAARD:   [Laughs]  Could be.
HOLLOWAY:  Did you attend the quantitative surface analysis session they had here on Saturday, I guess, right?
TOUGAARD:  Not this one on Saturday. I’ve attended several of the previous ones.
HOLLOWAY:  Right. They tend to be quite good. I’ve attended a few of those as well, and they’re very interesting and very stimulating.
TOUGAARD:  They certainly are. There’s a lot of time set off for discussions after the talks. Very important.
HOLLOWAY  I remember in particular one that was held prior to an AVS meeting in Seattle. But I must confess the thing that I remember was looking at the Northern Lights.  [Laughter]  Rather than thinking about quantitative surface analysis.
TOUGAARD:  That was the one we had out in the woods, was it?

HOLLOWAY:  Out in the woods, yeah.
TOUGAARD:  Yeah, I remember that. I was there also, yes.
HOLLOWAY:  Well, you see the Northern Lights up where you’re at much more than we see them here.  [Yes.]  But Gainesville is about 100 miles north from here, and I do remember one night in the middle of the wintertime that we were lying out in the yard by my house watching the Northern Lights in Gainesville, Florida.
TOUGAARD:  Oh really?  [Yes.]  That’s quite far south.
HOLLOWAY:  It was remarkably low in the hemisphere.  [Yes.]  So anyhow, we digress. That completes the subjects that I wanted to talk to you about. Do you have anything that you would like to added?
TOUGAARD:  No. I think this was pretty much the story.
HOLLOWAY:  Good. Well, congratulations again on the award. Very well deserved, and something that you should be proud of because it’s a nice award.
TOUGAARD:  Thank you very much, Paul.
HOLLOWAY:  Thank you, Sven.