AVS2014 Session TF+PS-ThM: Advanced CVD and Chemical Vapor Infiltration Methods

Since the discovery of the carbon nanotube (CNT) 20 years ago, extensive effort has been made to utilize their exceptional intrinsic properties toward industrial applications. However, availability has significantly thwarted these endeavors. In one section of my presentation, I will describe our efforts toward the economical mass-production of single-walled carbon nanotubes (SWCNT) based on the water-assisted chemical vapor deposition technique, from which highly efficient synthesis of vertically aligned SWCNTs grow from substrates (SWCNT forests). These SWCNT forests form through the self-assembly of individual SWCNTs when grown in sufficient density and have been shown to be useful templates for various applications from energy device electrodes to MEMS materials due both the continuous nature and high porosity. Further, I will describe the forest, and present a few examples of how we have infiltrated these material to create a new material with enhanced properties.

The world-wide shortage of the ³He isotope has led to a need for novel designs of neutron detectors. A detector based on the isotope ¹⁰B, in the form of thin films, has been suggested by the European Spallation Source (ESS). The detector design uses ¹⁰B₄C films, ≥1 µm, deposited on both sides of neutron transparent substrates such as Al blades.[1] The melting point of Al (660 °C) sets a strict upper temperature limit for CVD of the ¹⁰B₄C films. Also, metallic Al will be badly affected by corrosive by-products, like HCl. This means that traditional B₄C CVD routes based on BCl₃ and CH₄ cannot be used. An alternative CVD route is to use oragnoboranes, i.e. molecules with direct B-C bonds, as such molecules are very reactive and do not produce corrosive by-products.

We have demonstrated the synthesis of thin, X-ray amorphous, boron-carbon films at low temperature (400-600 °C), by thermally activated CVD using triethylboron, B(C₂H₅)₃, (TEB) as single precursor on both single crystalline Si (100) and Al substrates.[2] Films with B/C-ratio of 4.6 with density 2.42 g/cm³ (bulk B₄C density 2.52 g/cm³) and 3.6 with density 2.14 g/cm³ were deposited at 600 °C in hydrogen and argon ambient respectively, the impurity levels in the films was about 4 at.% of H at 600 °C. Further studies of TEB as precursor at higher temperatures (700-1200 ^oC) on Si substrates show that films with a B/C ratio of 4.5 and 3 were obtained from films deposited at 700 ^oC in hydrogen and argon ambient respectively with < 0.24 at.% of H. A threshold temperature of 1000 °C for the deposition is identified above which the B content decreases dramatically. Based on our results, a chemical mechanism for boron-carbon films from TEB, where the TEB molecule is decomposed to BH₃ and hydrocarbons, is suggested.

Plasma Enhanced CVD using trimethylboron, B(CH₃)₃, (TMB) is also explored to further lower the deposition temperature. Results from CVD and PECVD will be compared with state of the art PVD of ¹⁰B₄C.

[1] R. Hall-Wilton et al. IEEE NSS/MIC conference record, 4283 (2012)

[2] H. Pedersen et al. Chem. Vapor Deposition 18, 221-224 (2012)

In this paper, a new CVD method for ZnO film growth using the reaction between dimethylzinc (DMZn) and high-temperature H₂O produced by a catalytic reaction on Pt nanoparticles is presented [6]. H₂ and O₂ gases were admitted into a catalyst cell containing a Pt-dispersed ZrO₂ catalyst, whose temperature increased rapidly to over 1300 K due to the exothermic reaction of H₂ and O₂ on the catalyst. The resulting high-temperature H₂O molecules were ejected from a fine nozzle into the reaction zone and allowed to collide with DMZn ejected from another fine nozzle. ZnO epitaxial films were grown directly on a-plane sapphire substrates at substrate temperatures of 773-873 K with no buffer layer. Growth rates were 0.02-0.13 μm min^-1, and film thicknesses were 2-8 μm. X-ray diffraction patterns exhibited intense (0002) and (0004) peaks. The smallest FWHM value of the ω-rocking curve of ZnO(0002) was less than 0.1º (194 arcsec). The Hall mobility and residual carrier concentration of the epilayers were in the ranges 140-197 cm²V^-1s^-1 and 5.8×10¹⁶-6.0×10¹⁷ cm^-3 at 300K, respectively. This Hall mobility is very large compared with ZnO films grown directly on sapphire by other deposition methods. PL spectra at 10 K showed a strong emission peak at 3.360 eV, attributed to the neutral donor-bound exciton D^o_x. The FWHM was as low as 0.9 meV, which is smaller than that previously reported for ZnO obtained by MBE (5.5 meV) [4], and by PLD on a sapphire(0001) substrate (1.7 meV at 2K) [3]. .

[1] M. Sano et al., Jpn. J. Appl. Phys., 42 (2003) L1050. [2] H. Tampo et al., Appl. Phys. Lett., 84 (2004) 4412. [3] E. M. Kaidashev et al., Appl. Phys. Lett., 82 (2003) 3901. [4] A. Ohtomo et al., Semicond. Sci. Technol., 20 (2005) S1. [5] J. Dai et al., J. Cryst. Growth, 290 (2006) 426. [6] K. Yasui et al., MRS Symp. Proc., 1494 (2013) 127

The ability to fill a high aspect ratio feature with a thin film material enables the fabrication of many nanoscale devices. Examples include shallow trench isolation or inter-metal dielectric in microelectronics. One approach is to use chemical vapor deposition under conformal coating conditions. However, as film builds up on the sidewalls the width of the feature shrinks and the aspect ratio increases sharply. This often results in incomplete filling, leaving a narrow void or ‘seam’ of low-density material along the central axis. One solution is to taper the feature into a ‘V’ shape, such that uniform deposition causes the apex of the V to move smoothly upwards.

To achieve complete filling, the flux of deposition precursor down the axis of the feature must be sufficient to maintain a uniform growth rate. Precursor transport is typically modeled using the diffusion equation under quasi-static conditions. We show that for high aspect ratio features, the diffusion equation significantly under-estimates the flux of material that is deposited deep in the feature. This occurs because the diffusion formalism assumes a mean transport distance between collisions that is proportional to the feature size. However, in molecular flow some of the transport events occur at glancing angles to the feature sidewall and afford very long flight paths. These events move precursor species to the bottom of the feature, an effect that enhances filling.

We have developed a ballistic transport model based on computing the emission/capture probability between all points on the surface and coding the result as a matrix. Species transport from a starting distribution is found by matrix multiplication to afford the distribution of final positions. We first show how the results of this model compare with the diffusion formalism: the bulk of the transport is similar, but the ballistic model predicts a ‘tail’ of long-range events. We then simulate the filling of V-shaped features as a function of the apex angle and sticking coefficient. The result is a prediction of regimes that can afford complete filling.

Finally, we consider the effect of growth rate saturation under high precursor flux, an effect that is physically significant and vastly improves conformal growth. We derive from the ballistic model the total flux arriving at each position, and self-consistently calculate the effective sticking probability. We simulate the coating profiles on rectangular and V-shaped features and determine that rate-saturated growth conditions, in combination with long-range precursor transport, greatly expand the regime that affords complete filling.

To address these challenges Applied Materials has developed a new CVD technology we call FCVD™ to enable synthesis of high quality dielectric films including silicon oxides, silicon nitrides, silicon carbo-nitrides, silicon, low-k dielectrics, and carbon with a mechanism of film growth that promotes void-free fill irrespective of structure dimension and shape; this technology demonstrates capability to fill re-entrant structures with opening size <5nm and aspect ratio >20, flexibility to address multiple material systems and has been productized to address volume manufacturing requirements. In this paper we will demonstrate that we can achieve a void-free, profile-insensitive gap fill with multiple materials in a CVD reactor.

A recent short history of reactive evaporation by Mattox (1) described various methods for producing oxides, nitrides, carbides, and some compound materials using this special process. However, no mention was made producing hydrides using this method. A study was performed in the mid 1970’s at the General Electric Company (GE) Neutron Devices Department (GEND) in Largo, FL, by the author to study preparation of thin film hydrides using reactive evaporation and to determine their unique characteristics and properties.

Films were produced of scandium (Sc), yttrium (Y), titanium (Ti), zirconium (Zr), and the rare earth praseodymium (Pr), neodymium (Nd), gadolinium (Gd), dysprosium (Dy) and erbium( Er) hydrides by hot crucible filament evaporation in atmospheres of deuterium and tritium gas. All metal vacuum systems were used and dedicated for this special processing. Thin film test samples ~ 5,000Å thick were prepared on half-inch diameter molybdenum disk substrates for each occluder material.

Loading characteristics (i.e., gas –to-metal atomic ratios), oxidation characteristics, film structure, and stress properties were determined and showed near maximum gas-to-metal atomic ratios, variable oxidation properties, platelet type film structures and minimum film stress levels as determined by a double resonator technique. Also, stress aging characteristics were determined for some hydride films prepared in a radioactive tritium gas atmosphere.

The timeless data obtained showed gas-to-metal atomic ratios varied from 1.8 to 2.0, surface oxide levels varied from ~ 80Å to over 1,000Å, and initial normalized differential (tensile) stress levels were (1.0 to 4.0) x 10⁸ dyne/cm² for tritium loaded samples and (1.0 to 2.0) x 10⁹ dyne/cm² for deuterium loaded samples. Tritium loading, however, had the undesirable characteristic of having to dispose of the internal processing system fixtures, but the method generally produced desirable thin films.

† Formerly, Principal Member of the Technical Staff at Sandia National Laboratories,

Albuquerque, NM (Retired).

(1) Mattox, D.M., “History Corner- A Short History of Reactive Evaporation”, SVC Bulletin, p.50 –

51, Spring 2014.

A new technique called cathodic cage plasma deposition (CCPD) was used for growing TiN and TiO₂ films on silicon substrate. In this technique, the samples are positioned inside a cage having uniformly distributed round holes with fixed diameter, and onto an alumina insulator disk, so that the plasma acts on the cage and not on the sample surface, eliminating possible defects usually formed during the conventional plasma deposition. The CCPD technique produces films with high uniformity and permits a good control of roughness and crystallinity. The main advantages of this technique are the uniformity, tri-dimensionality, and high rate of deposition of the deposited films, as well as low cost of production.

TiN coatings increase the surface hardness and decrease the friction coefficient, thus enhancing the lifetime of components and tools employed in the metalworking industries. Thin films of TiO₂ have attracted considerable attention because of different applications on many fields due to their unique properties, such as chemical stability, no toxicity, low cost, high refraction index, high permittivity, wide valence band, etc. The TiO₂ main crystalline phases are: anatase, brookite, and rutile. The TiO₂ thin films can have a mixture of these phases showing hybrid properties. Many studies have focused on relationship of the different phases as dependent of the deposition method and parameters. Each one of these phases has its own characteristic properties, leading to different applications.

In this work, the influence of the parameters (temperature and gas atmosphere) in the characteristics of the deposited films was investigated. The TiN and TiO₂ thin films were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), and Raman spectroscopy in order to identify their crystalline phases and estimate their thicknesses. The combination of XRD and Raman spectroscopy results indicates that only a TiN crystalline phase was detected for the TiN films, and mainly the anastase phase was detected for the TiO₂ film. High crystallinity and uniformity of the films were observed by XRD, Raman, and SEM, confirming that this low cost technique is effective in producing high quality TiN and TiO₂ films.