The fact that Dr. Harold Winters passed away last year – at the age of 83 – has led to initiatives in tribute of Harold and his profound contributions to plasma science. One of these initiatives is a Special Issue Tribute to Harold Winters of the Journal of Vacuum Science and Technology. We have also contributed to this special issue with an article entitled “Revisiting the growth mechanism of atomic layer deposition of Al2O3: A vibrational sum-frequency generation study” [1] written by my former PhD student (now postdoc) Vincent Vandalon and myself. Another initiative is a special session which was held recently during the 64th AVS International Symposium and Exhibition (Tampa FL, Oct. 29 – Nov. 3, 2017). This session, which was an all-invited session, was entitled “The Science of Plasmas and Surfaces: Commemorating the Career of Harold Winters”. I had the honor to speak during this special session and here I will outline the story behind my presentation. But before doing so, I will first briefly describe what the profound contributions of Harold Winters to plasma science have been.
Harold Winters was a well-known research scientist who worked at the IBM San Jose Research Laboratory for 30 years (from 1963 to 1993). He became most famous with his work on plasma etching, also called reactive ion etching. Plasma etching emerged as a new dry etching technique in the late 1970s and it revolutionized pattern transfer in integrated circuit manufacturing. He and his colleague Dr. John Coburn studied the underlying mechanisms of this method which were not well understood at that time. To understand the complex processes originating from the interaction of electrons, ionic and reactive neutral species generated in the molecular plasmas used for etching, he and John took an experimental approach in which they exposed surfaces to collisionless directed beams of the various species present in the plasmas. To do so, they constructed specifically designed vacuum systems and carried out well defined experiments with in situ and ex situ analysis techniques.
Harold Winters at his retirement from IBM in 1993.
The most famous figure from Winters and Coburn, which is known to everybody in the field of dry etching and plasma processing, is shown below. This figure published in 1979 in their key article entitled Ion- and electron-assisted gas-surface chemistry—An important effect in plasma etching [2] shows the etch rate of poly-Si when exposed to beams of reactive species. In the first 200 s, the surface is exposed to the very reactive XeF2 gas, which can be considered as a simulant of reactive plasma neutrals (“radicals”) as it basically delivers F to the surface. At 200 s, a beam of Ar+ ions with an energy of 450 eV is turned on while leaving the XeF2 beam on. This XeF2 beam is switched off at 660 s while keeping the surface being exposed to the Ar+ ions. The etch rate with only the XeF2 beam or Ar+ beam on is not negligible but it is much smaller than for the case that the surface is exposed to both beams simultaneously. This clearly demonstrates the synergy between the Ar+ and the F, and more generally, between ions and reactive neutrals. This synergy has been demonstrated also for other etch chemistries and lies at the basis of anisotropic etching by plasmas. In narrow features being etched (e.g., trenches), the horizontal bottom of the feature is exposed to both radical species and energetic ions. The ions gain their energy when accelerated in the plasma sheath which gives them a high velocity (energy) in the direction perpendicular to the surface. The vertical sidewalls of the feature, however, are only exposed to radicals such that the vertical etch rate within the feature is much higher than the horizontal etch rate. This leads to anisotropic etching. The ability to etch anisotropically is a key merit of etching with plasmas. This important merit, explained by figure below by Winters and Coburn, makes the figure below truly “iconic”. I always make sure that the students who take my class on Plasma Processing Science & Technology know it by heart by the time they have finished the class. By the way, inspired by the work of Winters and Coburnn, such figures have also been generated by several other researchers to demonstrate synergistic effects in plasma processing. At the TU/e there is also a rich history in the field of fundamental studies on plasma etching (initialized by Prof. Herman Beijerinck) and in an article published in 2008 we demonstrated a similar synergistic etching effect for HfO2 [3].
Iconic figure published by Coburn and Winters demonstrating the synergism in etching between reactive neutral species (“radicals”) and energetic ions [2]. This figure explains the fact that plasmas can be used for fast anisotropic etching as indicated in the cartoon. The radicals have an isotropic distribution whereas the ions are directional due to their acceleration over the plasma sheath.
My research has largely been influenced by the research approach taken by Winters and Coburn. I have outlined this in the aforementioned presentation that I gave at the special tribute session. This presentation was entitled Surface science aspects of (plasma) ALD reactions, in analogy with the review paper published by Winters and Coburn in 1992: Surface science aspects of etching reactions in Surface Science Reports [4]. Another (indirect) way in which Coburn and Winters had a large impact on my academic career, is by the fact that I received the so-called Coburn and Winters Student Award of the Plasma Science & Technology Division in 1999. At that stage I was a 4th year PhD student at the Eindhoven University of Technology and I presented my work at the 46th AVS International Symposium which was held in Seattle WA. The award was given on the basis of my research to that date and the presentation itself. As can be seen from the presentation slides shown in the figure below, my PhD research dealt mainly with plasma deposition (plasma-enhanced chemical vapor deposition, PECVD) of thin films of hydrogenated amorphous silicon (a-Si:H). I was mainly interested in understanding the plasma chemistry in the remote SiH4 plasma used for high-rate deposition of a-Si:H as well as in revealing the underlying surface reactions mechanisms. For this, I employed an extensive set of plasma and surface analytical techniques: (threshold ionization) mass spectrometry, cavity ringdown spectroscopy, optical emission spectroscopy, sticking probability measurements, ellipsometry and attenuated total reflection infrared spectroscopy. The latter method I learnt during my 3 month stay within the group of Prof. Eray Aydil at the University of California Santa Barbara.
Two slides of my presentation given at the 46th AVS International Symposium in Seattle WA in 1999. On the basis of this presentation and my research to that date, I received the Coburn and Winters Student Award from the Plasma Science and Technology Division of the AVS. Prof. Richard van de Sanden and Prof. Daniel Schram were my thesis advisors.
After receiving my PhD degree and taking up some postdoc positions, I remained interested in the surface reactions during thin film growth by PECVD. When applying for a fellowship position at the Netherlands Royal Academy of Sciences in the Netherlands (in those days one important way to get a tenure track position at a Dutch university), I chose this as research focus. In 2001, I submitted a proposal with the title: The Surface Science of Plasma Deposition. In the figure below you see the first page of the supplemental section of the proposal in which I presented the proposed research in detail. When one reads it, it is clear that the research approach is very much influenced by the research approach taken by Winters and Coburn and many of their “followers”. A main difference was however that the proposed research focused on deposition instead of etching. Apart from the general topic (and title) itself, their influence can be derived from the fact that the first reference is to the work of Coburn and Winters. Moreover the application of advanced surface diagnostics is described as well as the construction of a specifically designed high vacuum system with beam-like ion and radical sources with which the deposition process can be mimicked. The model system of hydrogenated amorphous silicon was chosen (the proposal mentions also hydrogenated amorphous carbon although in the end no research on this material was carried out in the project) although the research was aimed at gaining fundamental insight in general mechanisms during PECVD of thin films. This is all quite analogous to the approach taken by Coburn and Winters. An obvious novel aspect of the proposed research was however the plan to study the film growth also by nonlinear optical spectroscopy. The proposal, which was accepted for funding by the jury members at the end of 2001, describes the plan to study the surface reactions during film growth by the spectroscopic method of broadband sum frequency generation. As I will describe below, this was only realized about 15 years later.
First page of the supplemental section of the proposal “The Surface Science of Plasma Deposition” that I submitted to the Netherlands Royal Academy of Sciences in 2001. This proposal was aimed at obtaining a fellowship and hence a tenure track position at the TU/e. The proposal was successful and mid 2002 I started as an assistant professor. The comments highlight some parts of the proposed research which is clearly influenced by the work of Coburn and Winters (and many of their other followers).
After starting as an assistant professor at the Eindhoven University of Technology, many aspects mentioned in the proposal were realized. Obviously this was only possible with the help of several PhD students and postdocs. The design and construction of a high vacuum system to carry out well-defined growth studies was a very important part of the work. A schematic and photograph of this system are shown below. The key idea behind the design was to put the substrate centrally in the system such that the substrate could be accessed optically both from its front and back side. The system was split up in two vacuum chambers which were both pumped to avoid deposition at the back side of the sample but in such a way that no tight vacuum seal was necessary for the substrate. In this way, all kind of substrates could be employed, including attenuated total reflection crystals. The substrate could be heated radiatively from the back while various plasma, radical, ion and gas sources could be used to prepare or treat films on the front side of the surface. Some more details can be found in a paper by my first PhD student Johan Hoefnagels [5]. Many studies on a-Si:H film growth were carried out with the system. Experiments in which the optical methods of spectroscopic ellipsometry, attenuated total reflection infrared spectroscopy, and optical second harmonic generation were all applied at the same time (in situ and real time!) are a real highlight in this respect [6,7]. Another highlight is the study of a-Si:H film growth and the interaction of a-Si:H films with atomic hydrogen as studied with the method of evanescent-wave cavity ringdown spectroscopy [8,9]. This complicated experiment involved a very special optical element called a “folded cavity” and making it work was really a major achievement. Note that the experiments with respect to a-Si:H growth were published over the course of many years and the latest publication about the evanescent-wave cavity ringdown spectroscopy experiments [9] even only appeared recently in the Special Issue Tribute to Harold Winters of the Journal of Vacuum Science and Technology.
The high vacuum system for fundamentals studies on film growth. It consists of two vacuum chambers separated by a flange on which the substrate is mounted. The vacuum seal between the chambers does not have to be very good which relaxes the requirements for the mounting of the substrate. It does however prevent film deposition on the back side of the substrate. Film growth and/or treatment can be achieved by various plasma, radical, ion and gas sources mounted at the front side of the system. The substrate can be heated radiatively and it can be probed by several optical methods at the same time (both from the front and the back) to carry out advanced studies of the surface reactions.
During the academic year 2004-2005, I did a sabbatical in the group of Prof. David Graves at the University of California Berkeley. This is where I got to know Harold Winters and John Coburn personally. At that time, they were already retired for quite a while from IBM but they were still interested in being involved in research with PhD students. So that’s why they came in into David’s lab every Tuesday (at least during the time I was there) to join the group meetings and to discuss with the PhD students about the experiments going on. For me it was extremely nice and also very inspiring to interact with Harold and John. They had plenty of time and we had many chats about science and also about many other topics. Harold was still working on a manuscript about silicon etching during that period. This manuscript dealt with the penetration of F into Si and appeared in the Journal of Vacuum Science and Technology A in 2007 [10]. It was the last paper he published. I found it amazing to see him still working so fanatically on the topic. He also presented the work during the 27th International Conference on Phenomena in Ionized Gases (XXVII ICPIG 2005) which was held in Eindhoven in the summer of 2005. Harold was invited to this conference to speak during the special session organized on the occasion of the retirement of Daan Schram (one of my thesis advisors). Harold and Daan had been friends for many years. During the conference there was also an excursion and conference dinner. Harold did not want to join the excursion (to the theme park De Efteling) but he wanted to join the dinner. Therefore Harold and I used the opportunity to do a visit to our labs at the university and afterwards we stopped by at my house. After some coffee, I had to do some other business and he was ok to wait for me until I came home again. When I did, he had fallen asleep in the armchair.
Harold Winters giving a presentation during the 27th International Conference on Phenomena in Ionized Gases (XXVII ICPIG, Eindhoven, 18-22 July 2005). He presented during a special session organized on the occasion of the retirement of Prof. Daan Schram. Note that he was still using an overhead projector at that time.
When I did my sabbatical in Berkeley, I had already starting working on atomic layer deposition (ALD). How I got involved in ALD, is a story in itself but it had to do with two main aspects. First of all, I figured that with my experience on processing plasmas and plasma-enhanced CVD, I could really contribute to the field of ALD. People were starting to employ plasmas for plasma-enhanced ALD, but in many cases their background in processing plasmas wasn’t that extensive. Secondly, I figured that it would be easier to understand the surface reactions during film growth for the case of ALD because the surface chemistry is not as complex as during plasma-enhanced CVD (although it can still be quite complex for sure!). Even for the case of plasma-enhanced ALD, this is the case because the plasmas used in ALD are much less complex as those used for plasma-enhanced CVD. When I arrived in Berkeley, the group of David Graves had just published a paper related to plasma-enhanced ALD about TiN to which John Coburn had a large contribution [11]. Berkeley (in particular Caffe Strada at Bancroft Way) is also the place where I finished our first ALD paper ever. This paper was dealing with plasma-enhanced ALD of TiN as well [12].
After the start-up phase in which it was important to demonstrate a couple of plasma-enhanced ALD processes in our lab, we also started to investigate the surface chemistry during ALD growth. Over the years we implemented several methods to obtain insight in the surface reactions. These included in situ spectroscopic ellipsometry, quartz crystal microbalance, mass spectrometry, optical emission spectroscopy, and in situ infrared spectroscopy (see e.g., some of our studies on ALD of Al2O3 [13,14,15]). We employed these methods for both thermal and plasma-enhanced ALD and the studies were carried out in our regular ALD systems, especially in our home-built tools as they allow for some more flexibility in terms of implementing a combination of methods at the same time. Eventually we also took the step towards very well-defined ALD studies on our high vacuum system discussed before. We started to study the thermal ALD process of Al2O3 (we didn’t want to start with the experimentally more difficult case of plasma-enhanced ALD) with the nonlinear optical technique of broadband sum-frequency generation. This was one of the methods mentioned in my research proposal submitted to the Netherlands Royal Academy of Sciences in 2001. So after 15 years or so I realized my dream of studying thin film growth by this very advanced method. This experimentally challenging work was carried out by Vincent Vandalon as part of his PhD thesis. The laser system and high vacuum system are shown below. The first manuscript about the work appeared in Applied Physics Letters in 2016: What is limiting low-temperature atomic layer deposition of Al2O3? A vibrational sumfrequency generation study [16]. A more extensive follow-up paper appeared in the Special Issue Tribute to Harold Winters of the Journal of Vacuum Science and Technology as already mentioned in the beginning of this blog post [1]. With our work, we have shown that several open questions, in the field of ALD can be solved by the unique features of the broadband sum-frequency generation method. This holds even for the well-studied and relatively well-understood process of ALD of Al2O3. A strong point of the method is that it can be used to obtain cross-sections for the reactions of precursor molecules with the surface. These can be translated into reaction or sticking probabilities. Extracting quantitative information about surface reactions is a main goal in surface science and Harold Winters has always been promoting this in his research on etching reactions. By studying the surface science aspects of (plasma) ALD reactions, or of reactions during atomic-scale processing in general, we can extend the legacy of Harold Winters.
A view in the lab: on the left side the optical table with the femtosecond laser systems to generate 800 nm and tunable infrared photons and on the right side the high vacuum system in which films can be grown by ALD. With this equipment broadband sum-frequency generation studies were carried out during Al2O3 film growth.
=====================================================================================
The AVS newsletter also published testimonials from Past AVS Plasma Science Technology Division Harold Winter Award Recipients. Below is my testimonial:
Prof. Erwin Kessels at the Eindhoven University of Technology, the 1999 recipient, added “it goes almost without saying that winning the award was a critical step in my career. It definitely was the trigger for my interest in staying in academia and it provided the necessary confidence that I could contribute to cutting edge research in the field of plasma processing and thin films. Yet the award started to mean even much more to me when I got to know John Coburn and Harold Winters in person. In 2004/2005 I spent a sabbatical in the lab of David Graves at U.C. Berkeley and both John and Harold came in every Tuesday to help the students and postdocs with their research. During this period I had a lot of interaction with Harold and I experienced his passion about research and his dedication to try to understand things to the root of the matter. I will always remember the many in-depth discussions that we had. This certainly holds for the discussions about the next step in understanding of etching of silicon by fluorine atoms, as even after several decades of research, there was still much more to understand! What struck me most however was his wonderful personality. When he visited me in the Netherlands in 2005 it felt that we had become friends. Harold and his work will continue being a source of inspiration to me.”
=====================================================================================
Special thanks go out to Prof. Richard van de Sanden, my thesis advisor and mentor, who played a very important role during my career. With his enthusiasm he got me interested in the fundamental aspects of thin film growth and the importance of his contributions to all written above cannot be overstated. He also contributed to the genesis of this blog post.
References
[1] Revisiting the growth mechanism of atomic layer deposition of Al2O3: A vibrational sum-frequency generation study, V. Vandalon and W.M.M. Kessels, J. Vac. Sci. Technol. A 35, 05C313 (2017).
[2] Ion- and electron-assisted gas-surface chemistry—An important effect in plasma etching, J.W. Coburn and H.F. Winters, J. Appl. Phys. 50, 3189 (1979).
[3] Ion-radical synergy in HfO2 etching studied with a XeF2/Ar+ beam setup, P.M. Gevers, H.C.W. Beijerinck, M.C.M. van de Sanden, and W.M.M. Kessels, J. Appl. Phys. 103, 083304 (2008).
[4] Surface science aspects of etching reactions, H.F. Winters and J.W. Coburn, Surface Science Reports 14, 161 (1992).
[5] Plasma-surface interaction and surface diffusion during silicon-based thin film growth, J.P.M. Hoefnagels, E. Langereis, W.M.M. Kessels, and M.C.M. van de Sanden, IEEE Transactions on Plasma Science 33, 234 (2005).
[6] a-Si:H/c-Si heterointerface formation and epitaxial growth studied by real time optical probes, J.J.H. Gielis, P.J. van den Oever, M.C.M. van de Sanden, and W.M.M. Kessels, Appl. Phys. Lett. 90, 202108 (2007).
[7] Real-time study of a-Si:H/c-Si heterointerface formation and epitaxial Si growth by spectroscopic ellipsometry, infrared spectroscopy, and second-harmonic generation, J. J. H. Gielis, P. J. van den Oever, B. Hoex, M.C.M. van de Sanden, and W.M.M. Kessels, Phys. Rev. B 77, 205329 (2008).
[8] Absolute in-situ measurement of surface dangling bonds during a-Si:H growth, I.M.P. Aarts, A.C.R. Pipino, M.C.M. van de Sanden, and W.M.M. Kessels, Appl. Phys. Lett. 90, 161918 (2007).
[9] Atomic hydrogen induced defect kinetics in amorphous silicon, F.J.J. Peeters, J. Zheng, I.M.P. Aarts, A.C.R. Pipino, W.M.M. Kessels, M.C.M. van de Sanden, J. Vac. Sci. Technol. A 35, 05C307 (2017).
[10] Penetration of fluorine into the silicon lattice during exposure to F atoms, F2, and XeF2: Implications for spontaneous etching reactions, H.F. Winters, D.B. Graves, D. Humbird, S. Tougaard, J. Vac. Sci. Technol. A 25, 96 (2007).
[11] Fundamental beam studies of radical enhanced atomic layer deposition of TiN, F. Greer, D. Fraser, J. W. Coburn, and D.B. Graves, J. Vac. Sci. Technol. A 21, 96 (2003).
[12] Plasma-assisted atomic layer deposition of TiN monitored by in situ spectroscopic ellipsometry, S.B.S. Heil, E. Langereis, A. Kemmeren, F. Roozeboom, M.C.M. van de Sanden, and W.M.M. Kessels, J. Vac. Sci. Technol. A 23, L5 (2005).
[13] In situ reaction mechanism studies of plasma-assisted atomic layer deposition of Al2O3, S.B.S. Heil, P. Kudlacek, E. Langereis, R. Engeln, M.C.M. van de Sanden, and W.M.M. Kessels, Appl. Phys. Lett. 89, 131505 (2006).
[14] Reaction mechanisms during plasma-assisted atomic layer deposition of metal oxides: A case study for Al2O3, S.B.S. Heil, J.L. van Hemmen, M.C.M. van de Sanden, and W.M.M. Kessels, J. Appl. Phys. 103, 103302 (2008).
[15] Surface chemistry of plasma-assisted atomic layer deposition of Al2O3 studied by infrared spectroscopy, E. Langereis, J. Keijmel, M.C.M. van de Sanden, and W.M.M. Kessels, Appl. Phys. Lett. 92, 231904 (2008).
[16] What is limiting low-temperature atomic layer deposition of Al2O3? A vibrational sum-frequency generation study, V. Vandalon, and W.M.M. Kessels, Appl. Phys. Lett. 108, 011607 (2016).
Leave a Reply