Fifth international conference on hot-wire CVD (Cat-CVD) process

Thin Solid Films 517 (2009) 3413–3414 Contents lists available at ScienceDirect Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f Preface Fifth international conference on hot-wire CVD (Cat-CVD) process This is the fifth in the series of hot-wire chemical vapor deposition (HWCVD) (or catalytic CVD) conferences held for the purpose of exchanging information and highlighting research advances in this expanding technology. The first conference was held in Kanazawa, Japan in November 2000, and was organized and hosted by Prof. Hideki Matsumura. Succeeding conferences have since been held biannually in Denver, USA, hosted by Dr. Harv Mahan, in Utrecht, the Netherlands, hosted by Prof. Ruud Schropp, then in Takayama, Japan, hosted by Prof. Shuichi Nonomura, and the most recent gathering was in Cambridge, USA, hosted by Prof. Karen Gleason. The next conference is scheduled to be held in Palaiseau, France, in September 2010. While some attrition has occurred amongst the original conference participants, new researchers have continued to join the HWCVD field, enabling conference participation to remain at relatively constant levels. In the present conference, 80 papers were presented, with 110 researchers from 15 countries in attendance. Fifty-three papers, approved by the review process, are included in this volume of Thin Solid Films. It is important to remind readers of the most important advantages of the HWCVD (or Cat-CVD) process. First, since the technique is thermal and/or catalytic in nature, relying on a heated metal filament to decompose the gas species, it is a ‘gentle’ technique, and does not generate any energetic ion bombardment and thus minimal substrate damage occurs. This is important for depositing passivation or gas barrier films on organic devices. Second, due to the low pressures that can be used for gas phase decomposition, a high flux of atomic H can be achieved, which can be used for material etching or photo-resist removal. Third, since no plasma is needed, the substrate is decoupled from the deposition process, enabling substrates to be easily introduced and removed from the deposition chamber without disturbing the deposition. Moreover, step coverage is excellent and uniformity can easily be optimized. Fourth, it is an easily scalable method by expanding the spanned area of the catalyst, which is particularly important for industrial implementation. A general observation to be presented in this preface is that new and innovative uses for HWCVD (or Cat-CVD or initiated CVD), continue to be found. In the first two conferences, the great majority of the presentations were devoted to gas–phase and deposition chemistry of silicon related materials (hydrogenated amorphous silicon, microcrystalline silicon, polysilicon, epitaxial silicon, silicon alloys with carbon and germanium) and their applications (solar cells and thin film transistors). In the third and fourth conferences, an increasing variety of thin film materials have been obtained with this method, to name a few: silicon nitride, silicon dioxide, aluminum oxide, aluminum nitride, Si–O–C, Si–N–C alloys, diamond, carbon nanotubes, nanowalls or nanoparticles. Moreover, transition metal oxide nanoparticles have been synthesized using HWCVD for applications like gas sensors or electrochromic windows. A novel HWCVD variant called initiated CVD was discussed in the fourth conference and was demonstrated to be a convenient single step synthesis method to produce 0040-6090/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2009.01.016 high quality polymer thin films. Discussion of this technique has expanded rapidly in the present conference. A recurrent issue in HWCVD is the lifetime of the catalytic metallic filaments. A long life is necessary in order to prevent the need for frequent wire changes, so as to deposit a long series of thin films. In the case of the deposition of silicon layers using tungsten wire, there is rapid formation of tungsten silicide. This explains why tungsten has been progressively replaced by tantalum as a filament material. Researchers at Utrecht Univ. found that annealing the Ta filament at 2100–2200 °C in vacuum is a straightforward method for regenerating the Ta wire. The same group studied the reactions of a heated Ta filament with ammonia and silane — of great interest for the deposition of silicon nitride — as well as with ammonia, methane and hydrogen for the deposition of diamond and carbon nanotubes. Researchers at IPV Jülich found that rhenium shows the best performance in terms of long-term stability in a monomethylsilanehydrogen environment for the growth of microcrystalline silicon carbide. Concerning the deposition of metal oxides, iridium was used as a filament material by Shizuoka Univ. as it exhibits the best resistance against oxidation. Chromel was chosen by Kanagawa Inst. Tech. as the best catalyst for depositing alumina from a trimethylaluminum–oxygen environment. Fundamental research in the area of Si deposition was limited in the presentations. These focused on HWCVD hydrogenated microcrystalline silicon (µc–Si:H) deposition, for example on the structure of low substrate temperature µc–Si examined by high resolution TEM (Univ. Western Cape), the addition of HCl to suppress growth of the a–Si:H phase during µc–Si deposition (Seoul National Univ.), and the examination of µc–Si growth at chamber pressures similar to those used for PECVD µc–Si deposition (4–12 Torr) (Nagoya Univ.). An important work also examined µc–Si deposition on rough (Ag/ZnO) substrates, where it was shown by XTEM that if the substrate surface local opening angle is smaller than a certain value, micro-cracks form in the valleys below these surface structures, causing structural defects that deteriorate solar cell performance. As a result of this work, a criterion for the morphology of light-trapping substrates for photovoltaics can be set (Utrecht Univ.). On the other hand, an important new research area in Si deposition, led by researchers at NREL, has been the demonstration of high quality epitaxial Si growth by HWCVD. The motivation behind this research is that if epitaxial thickening of a high quality seed layer, deposited on an inexpensive substrate, can be performed, this may be a route to large-area c–Si thin film layers for photovoltaics. HWCVD has been demonstrated to be a uniquely viable deposition candidate in this regard, as it satisfies the requirements of high deposition rate, high quality, low substrate temperature during growth (compatible with borosilicate glass), and industrial scalability. In particular, using a substrate temperature of 650– 725 °C and a W filament operated at 2100 °C, researchers at NREL have achieved growth rates as high as 300 nm/min, have grown epitaxial layers as thick as 45 µm, and have measured dislocation densities as low as 3414 Preface 105 cm− 2. Further, reliable control over both n-and p-type dopant densities has been demonstrated. Using Re filaments and monomethylsilane, IPV Jülich explored the HWCVD deposition of (n-type) µc–SiC:H in an n–i–p solar cell configuration, with illumination through the n-layer. The i-and p-layers were deposited by (VHF) PECVD, with an Ag/ZnO back reflector. With a filament temperature between 1900–2000 °C, E04 window layer valuesN 3.0 eV were achieved, resulting in a very high device Jsc of 28.1 mA/cm2 and a solar cell efficiency of 9.2%, obtained using a µc–Si i– layer thickness of 2.0 µm. High efficiency solar cell results were also reported using a HWCVD a–SiGe:H i–layer, with this i–layer containing 42 at.% Ge, exhibiting a Tauc gap of ~1.42 eV and deposited at a growth rate of 4.1 Å/s (NREL). No bandgap grading was used for this i–layer deposition. The best achieved efficiency of 8.64% compares quite favorably with those for PECVD a–SiGe:H cells also grown at high growth rate (~ 5 Å/s), but which use a double (graded) a–SiGe:H i–layer. A direct correlation between solar cell performance and i-layer material structure was demonstrated. For the future, there is an increasing interest for fabricating solar cells on flexible substrates, i.e. at low temperature. Two approaches were examined. In the first, a PEN/ZnO:Al superstrate a–Si:H p–i–n cell was fabricated, with a wide bandgap buffer layer inserted at the p–i interface to limit recombination (Univ. Minho); this choice of (plastic) substrate limited the substrate temperature to a value of 150 °C. In the second, Corning 1737 glass substrates, again coated with Ag/ZnO back reflectors, were used (Univ. Barcelona) in a substrate approach. In this work, the electrical and optical properties of the individual a–Si:H based solar cell layers, all deposited at a substrate temperature of 150 °C by HWCVD, were investigated. In both cases, initial solar cell efficiencies as high as 4.6% were achieved. An important application of the HWCVD technique concerns the deposition of silicon nitride films. Japanese industry is particularly active in this area and SiNx films that can be deposited up to 7 nm/s have been shown to be highly transparent, dense and to exhibit good dielectric properties. They can therefore be used as passivation layers, as high permeation barrier films for organic devices, or as gate dielectrics in thin film transistors. Material Design Factory Co. has succeeded in the growth of organic-inorganic hybrid materials (total coating thickness of 300 nm) on PEN substrates leading to a water vapor transmission rate of 10− 3 g/m2.day. The quantity of reports on activities concerning the deposition of diamond, carbon nanotubes and carbon nanowalls has settled down. Researchers at Ajou Univ. showed that the growth of CNTs by HFCVD can be very rapid and non linear versus time. CNRS Strasbourg correlated the growth of CNTs with simulated data based on gas phase and surface chemistry. Following an opposite trend, the method of initiated chemical vapor deposition (iCVD) pioneered by the MIT group of Prof. Karen Gleason, has expanded rapidly as a unique and versatile form of HWCVD, as evidenced by a dozen oral presentations. These presentations covered a range of topics from basic materials research to commercialization efforts. The iCVD method allows the facile synthesis of organic polymeric thin films through free radical polymerization. In iCVD, both monomer species and an initiating species are introduced through the gas phase. The susceptibility of the initiating species to thermal activation enables the use of low filament temperatures (200 to 500 °C) and low substrate temperatures (as low as 25 °C). This allows functional organic groups in the monomer species to be retained and the polymeric films utilized for a variety of diverse applications, including sensing, biological compatibility, and antifouling surfaces. The Triton Systems Company presented its work on a novel detector for peroxide explosives based on iCVD, while MIT discussed a novel resistive switch which exploits the chemically specific interaction between a given iCVD polymer and a desired analyte. Drexel University and Institut Quimic de Sarra-Univ. Ramon Llull, described highly wettable and swellable hydrogel thin films grown by iCVD as platforms for cell growth. A group from the Hong Kong Inst. Sci. Tech. described a biologically interesting functional polymer that could be synthesized by iCVD but not by solution chemistry. These biomaterials may also provide effective coatings for implantable medical devices and complement extremely stable biocompatible dielectric coatings, as described by researchers from MIT. Low dielectric constant films, as desired for advanced microelectronics applications, in the form of “low-k” iCVD materials were described by the groups of Northeastern and MIT. The latter group showed that feature sizes of 25 nm could be achieved by combining iCVD with colloidal lithography. Novel patterning methods to achieve controllable multifunctional surfaces using two different iCVD materials were demonstrated by MIT: hydrophilic/hydrophobic checkerboards and orthogonal “clickable” surface for regionally segregated attachment of different proteins. The ability of iCVD to conformably coat features having a high aspect ratio and its application to the surface modification of advanced membranes was demonstrated by MIT. The new possibilities opened up by basic research and synthesis of iCVD polymers has demanded the rapid development and commercialization of iCVD for specific applications. The nature of these efforts, and the commercial iCVD systems developed to undertake their realization, were discussed by the GVD Corporation. In another application unique to HWCVD, highly crystalline metal oxide nanoparticles have been produced at high chamber pressures (N35 Torr) by purposely evaporating a W (Mo) filament in a partial oxygen atmosphere (NREL, Univ. Colorado). When thin films are formed from these nanoparticles using an electrophoresis technique, they exhibit an enhanced surface area, thus facilitating the absorption and extraction of Li ions for electrochromic and battery applications. Furthermore, because of the small nanoparticle size (10–20 nm) and resultant film porosity, volume expansion and contraction upon Li absorption and extraction is less severe than that exhibited for more compact films containing larger particles, enabling an enhanced cycling stability. This cycling stability, with minimum degradation in performance after the first cycle, was demonstrated for 100 cycles in WO3 nanoparticle films (as an active electrochromic layer), and for 150 cycles in MoO3 nanoparticle films (as an anode in a lithium ion battery). Finally, the editors wish to thank many people as well as organizations. Without their help and support the conference would not have been held. First, we thank Professor Karen Gleason for taking on the challenge of organizing and hosting this conference in the pleasant city of Cambridge, MA. Not only did we enjoy the conference and the interesting tour of GVD, in addition to the banquet and the entertaining Boston Duck Tour, but just after the conference ended some of the participants were able to enjoy the local Cambridge Festival. We are also grateful to the Local Arrangement Committee for making the conference run smoothly and to the members of the International Advisory Committee for their help with refereeing the manuscripts. We are deeply grateful to the governmental and industrial sponsors for their interest in this field of research and for their generous financial support, which made this conference possible. Please note that the lists of committee members and of corporate sponsors are shown on a separate page. For ease of reading, the 53 papers of these Proceedings have been classified into 7 sections and sub-sections. The Guest Editors Jean-Eric Bourée Laboratoire de Physique des Interfaces et des Couches Minces, Ecole Polytechnique, 91128 Palaiseau, France Corresponding author. E-mail address: jean-eric.bouree@polytechnique.edu. A. Harv Mahan National Renewable Energy Laboratory, 1617 Cole Blvd., Golden, CO 80401, USA E-mail address: harv_mahan@nrel.gov.