ICMCTF2010 Session FP: Symposium F Poster Session

Uniform tungsten oxide (W₁₈O₄₉) nanowires with diameters of 15–20 nm and lengths of several micrometers were synthesized by thermal chemical vapor deposition (CVD) without using any catalyst via a two-step heating process in a tube furnace. Typical temperatures of this two-step process are 8 00 and 1000°C, respectively. The first-step heating was found crucial for the growth of small and uniform nanowires because it made the crystal cluster uniformly distributed on the p-type Si(100) wafer surface. Morphology, composition, and crystal structure were characterized by scanning electron microscopy (SEM); x-ray diffraction (XRD); transmission electron microscopy (TEM); and energy-dispersive X-ray spectroscopy (EDS), Raman, UV-visible, and cathodoluminescence (CL) spectroscopy. SEM images of high-density W₁₈O₄₉ nanowires clearly demonstrate that the nanowires have a uniform one-dimensional morphology high aspect ratio. The result of XRD, TEM, and EDX confirmed the formation of W₁₈O₄₉ nanowires ( lattice constants, a = 1.832 nm; b = 0.3784 nm; c = 1.403 nm) containing W and O atoms, with [010] as the major growth direction. The vapor-solid (VS) mechanism is responsible for the growth of W₁₈O₄₉ nanowires in this experiment since no catalyst were used. Blue emission was observed in both the UV-visible and CL spectra, indicating that the W₁₈O₄₉ nanowires exhibited a red-shift at an optical absorption wavelength due to oxygen deficiencies. The crystallinity and size distribution of the nanowires influenced the band gap. In the CL spectra, the blue emission was at shorter wavelengths than reported previously, which is attributed to the nanoscale size effect.

Optically-active Au:TiO₂ thin films were prepared and characterized with the aim to establish a correlation between the Surface Plasmon Resonance Effect (SPR) and the films’ effective thermal transport properties. Variations of the optical activity and of the SPR effect of the films were obtained by dispersing Au nanoclusters in an amorphous TiO₂ dielectric matrix, by varying the films’ Au content and by annealing the as-deposited films, thus tailoring the size, shape, and number of Au nanoclusters [1].

Different series of films containing amounts of gold of 10 - 30 at.% were deposited by dc reactive magnetron sputtering. Thermal annealing was carried out in vacuum atmosphere, using temperatures of 300 - 800ºC. A heating rate of 5 ºC/min., one hour at the desired temperature, and free cooling was applied in all annealing processes.

The depth profiles of the effective thermal transport properties of the films, measured by frequency-dependent Modulated IR Radiometry [2], exhibit a structure characterized by two characteristic lengths: At higher modulation frequencies, corresponding to smaller structures, the effects related to the Au nanoclusters are sensed, whereas at lower modulation frequencies, the thermal properties of the films as a whole are measured. Preliminary results show that the significant variations of the crystalline phase and of the clusters’ shape and size are correlated with the changes observed in the evolution of the thermal depth profiles, as a function of the increased annealing temperature.

[1] M. Torrell, P. Machado, L. Cunha, N.M. Figueiredo, J.C. Oliveira, C. Louro, F. Vaz, Development of new decorative coatings based on gold nanoparticles dispersed in an amorphous TiO₂ dielectric matrix, Surf. Coat. Technol. (2009), submitted.

[2] F. Macedo, A. Gören, A.C. Fernandes, F. Vaz, J. Gibkes, K.H. Junge, J.L.N. Fotsing, B.K. Bein, Potential of Modulated IR Radiometry for the On-Line Control of Coatings, Plasma Proc. Polym. 4, (2007) S857-864.

Zinc oxide (ZnO) is a semiconductor material with great potential in catalysis, photonics, and chemical sensing applications. Here we report the structural properties of ZnO thin films prepared by spin-coating a sol-gel precursor solution containing zinc acetate dihydrate and monoethanolamine in isopropanol solvent onto glass subtrates. By varying precursor concentration and number of spin-coats, film thicknesses between ~20 and 250 nm were obtained. Annealing at temperatures up to 500^oC was correlated to the structural, optical and morphological properties of the films.

The surface topography of the calcined films (characterized by high-resolution FIB-SEM), the optical absorption (measured by FTIR), and the extent of their c-axis orientation (characterized by XRD) varied with annealing and precursor solution concentration. At annealing temperatures of 100^oC and below, films exhibited the structure of a lamellar ZnO precursor, Layered Basic Zinc Acetate (LBZA). At higher temperatures, LBZA released loosely bound water and acetate groups, and Zn-OH groups condensed to form a film consisting of stacked ZnO nanograins (~20 nm diameter) having wurtzite structure. Micron-scale wrinkle patterns, in the form of skeletal branches, were observed in many of the films annealed at 500^oC. A minimum of surface wrinkles and a maximum in c-axis orientation were observed at intermediate precursor concentrations.

A simple model is presented to correlate wrinkle generation with the release of mechanical stress, which results from the removal of the solvent during the annealing process.

Residual stress gradients in shot peened TiN and TiCN coatings deposited by chemical vapor deposition are characterized using in-situ X-ray diffraction in a temperature range of 25-1100°C. The stress gradients are determined using the sin²ψ method and evaluated by applying inverse Laplace transformation. The experimental results show a tensile stress of about 0.6 GPa across the whole thickness of 9 μm in the as-deposited coatings. By shot peening high compressive stress can be produced up to a depth of about 2 μm. In the rest of the coating however the original tensile stress remains. The magnitude of the compressive stress strongly depends on the pressure used during the shot peening process. In-situ experiments indicate, however, that the temperature at which the relaxation of the peening-induced compressive stresses starts is influenced by the stress magnitude itself: in coatings with higher compressive stress a lower temperature is required to initiate relaxation processes.

In this study, we reported that the high-density ZnS nanowires were synthesized on a Si substrate using thermal evaporation at a synthesizing temperature of 1000 °C for 1 hr. A field emission scanning electron microscope (FESEM), a MAC glancing incident X-ray spectrometer with an incident angle of 0.5°, an x-ray photoelectron spectroscopy (XPS), a high-resolution transmission electron microscope (HRTEM) with an energy dispersive x-ray spectrometer (EDS) and a photoluminescence (PL) spectrometer were used to characterize the as-prepared ZnS nanowires, which shows that the ZnS nanowires are single crystalline hexagonal wurtzite structure. The photo-sensing measurement shows that the ZnS nanowire-based photodetectors have both high sensitivity and low response time, which demonstrates that ZnS nanowires have potential application in future UV photodetectors.

Oblique angle deposition (OAD) is widely used for fabricating thin-films with nanoengineered microstructures. The films deposited in this technique show the optical anisotropy originating from the microstructures and the porosity of the films increase due to shadow effects. These controls can be utilized to engineer thin films for specific applications such as three-dimensional photonic crystals, gradient index optical filters, broadband antireflection coatings, and linear polarizer, etc.

In this study, we investigate the optical and structural properties of linear polarization selector and polarization-discriminatory inverters. They are realized as a combination of quarter-wave plate, Bragg reflector and opposite quarter-wave plate. The two quarter-wave plates of opposite zigzag structure and the Bragg reflector of helical structure are fabricated by electron beam evaporation using OAD technique. The linear polarization selector transmits the normal incidence p-polarized light with wavelength lying in the Bragg regime, and reflects the s-polarized light as the p-polarized lights. The linear polarization-discriminatory inverters show that the incident linear polarized light becomes opposite linear polarized lights with Bragg effect at output.

Thin film silicon materials exhibit different structures, from amorphous to crystalline, which opens a wide range of applications, e.g. solar cells or MEMS devices. Comparatively to polycrystalline metallic thin film growth, stress evolutions in amorphous materials has been much less studied. However, a more simple behaviour may be expected due to the lack of elastic anisotropy and epitaxial strain. The purpose of this work is to study the stress development during magnetron sputtering growth of a-Si thin films. By varying the kinetic and energetic conditions, we attempt to clarify the origin of the compressive steady state stress observed in these amorphous layers.

In-situ stress evolution during deposition of amorphous Si was studied. The stress measurements were made in real-time during growth using a technique for substrate curvature measurement known as the multiple-beam optical stress sensor. The a-Si thin films were grown at room temperature on single crystal Si(001) substrates by either dc or rf magnetron sputtering in an ultra high vacuum system (~2. 10^-6 Pa) using Ar as working gas. Different target powers and Ar pressures were used to vary the deposition rate in the range 0.01 to 0.15 nm/s.

In a large range of pressure (< 0.44 Pa), a steady-state compressive incremental stress of ~-1.2 GPa is observed during growth of the a-Si layer and hardly independent on the deposition rate and substrate bias voltage. This suggests that the atomic peening mechanism does not play a significant contribution to the stress generation. Rather, the stress build-up is ascribed to surface stress effect, especially in the absence of epitaxial inheritance, similarly to what has been proposed by Floro et al [1] for a-Ge grown by thermal evaporation. At higher pressures, a stress relaxation is observed at larger film thickness, which can be related to a change in surface morphology and roughness, as revealed from ex-situ atomic force microscopy observations and X-ray reflectivity.

[1] J.A. Floro, P.G. Kotula, S.C. Seel and D.J. Srolovitz, Phys. Rev. Lett. 91, 096101 (2003).

In-situ impedance spectroscopy is a new method to acquire fundamental characteristics of plasma electrolytic oxidation (PEO) processes. In this technique, a variable frequency current waveform is used as a perturbation signal to obtain impedance characteristics of the electrolyser during the PEO processing. Previous studies have demonstrated feasibilities of this technique when operating in both the large and the small signal PEO modes [1;2]. It was however noticed that strong voltage electric fields have significant effects on characteristics of impedance spectra. In this research, to comprehend the effects of electric field distribution in the electrolyser, PEO processes were studied in equidistant and non-equidistant cell geometries.

During experiments, the voltage and current waveforms were acquired and processed using specially developed software to derive characteristics of impedance, admittance, and phase angle. To assist data analysis, the characteristics were refined to eliminate process-inherent wild points and noise, as well as to smooth and convolute data. The COMSOL software package was used to model and represent the distribution of electric field in the cells; these distributions were then correlated with immittance in the PEO electrolyser obtained by the impedance spectroscopy.

The modeling shows to be effective for estimation of the effects associated with electric field distribution in different geometry cells, thus promoting a better understanding of the PEO technique.

[1] E.V. Parfenov, A.L. Yerokhin, A. Matthews, Surf. Coat. Technol. 201 (2007) 8661.

[2] E.V. Parfenov, A.L. Yerokhin, A. Matthews, Surf. Coat. Technol. 203 (2009) 2896.

Composition analysis by electron probe microanalysis was performed to quantify the titanium and metalloid concentrations in the films. Similarly the composition was related to the deposition rate of the Ti-C(N,O) films. X-ray diffraction experiments revealed a face centered cubic structure for all the films but with a significant reduction of the lattice parameter from a = 4.336 Å to 4.292 Å as well as the grain size dropped from 40 to 20 nm, as the (O2 + N2) flow rate changed from 0 to 4 sccm. The structural analysis showed that the titanium carbide structure produced without O2 + N2 injection led to a progressive substitution of carbon by oxygen and/or nitrogen atoms in the lattice. Moreover, the (111) preferential orientation vanished when the O2+N2 gas mixture was supplied into the sputtering process. DC electrical conductivity of the films was gradually modified with the increase of the O2 + N2 gas mixture flow rate. Measurements versus temperature showed that a metallic to semiconducting evolution was observed, which was discussed taking into account the variations in chemical composition and the correspondent structural modifications in the films.

Coplanar XRD grazing incidence measurements (2θ scans) in parallel beam setup were evaluated by total pattern fitting by using of a new own software Mstruct. This is an extension of flexible modular system of Crystal Objects and Fox program [1] for studies of real structure of thin films. It includes the effects and corrections necessary for thin film diffraction geometries. Diffraction peak positions are determined by variable lattice parameters and zero shift error, and by possible residual stresses introduced in terms of weighted Reuss-Voigt model. Correction of peak positions for refraction is necessary for very low angles of incidence. Peak intensities calculated by the fObjCryst library from a known crystal structure model. They can be modified by absorption and the texture correction obtained from the known model of ODF after numerical integration over all crystallites with diffracting {hkl} planes perpendicular to direction of the measured diffraction vector, or they can be varied independently, if the texture is unknown. Peak profiles are given by numerical convolution of the known instrumental function described by the pseudo-Voigt function and physical profiles including several free parameters (models). The size broadening is described by the model of log-normal distribution of spherical crystallites or distribution given by a histogram, the strain broadening either by phenomenological microstrain or dislocation model. Stacking faults and twins can be included too. Individual layers in the system are characterized by their thickness and linear absorption coefficient.

It was shown that crystallization of amorphous films occurs above 220°C depending on their thickness and in all cases relatively large crystallites (> 100 nm) grow quickly from the very beginning of crystallization. By contrast as-deposited nanocrystalline films (crystallite size ~ 5 nm) or nanocrystalline powders showed the fine microstructure stable to quite high temperatures (above 500°C) before the strain relaxation and crystallite growth was observed.

During the crystallization of amorphous films tensile residual stresses were generated. They were of simple uniaxial character and strongly increase with decreasing film thickness. Their values obtained by the total pattern fitting and independently by the direct XRD stress measurement agreed well.

[1] V. Favre-Nicolin, R. Cerny, J. Appl. Cryst. (2002) 35, 734-743.

In thin hard coatings, thickness continuity is a very important factor for quality. The aim of this work was to compare the capabilities, advantages and disadvantages for measuring PVD coating thickness of three available analytical methods: Glow Discharge Optical Emission Spectroscopy (GDOES), Scanning Electron Microscopy (SEM) and the Kalotester Method. For this purpose, three mono-layer [CrN, TiN, TiAlN] and two multi-layer [Cr/CrN)x8, (CrN/TiN)x3] coating structures were deposited on tool steel by the arc evaporation method and their thickness was measured using the three methods. The complete method procedure for each analysis is described and discussed in this paper. Finally, advantages and disadvantages in the use of these three methods are concluded from this comparative study. The highly precise measurements obtained by SEM analyses is the main advantage of this method, however sample preparation is time consuming. GDOES method provides quick results but not highly accurate, especially when measuring multi-layered structures. SEM and GDOES analyses are usually expensive, being this one main disadvantage of these methods. The Kalotester method is the fastest and is inexpensive but not highly accurate. It was concluded from this work that no single method is better than the others. It will have to be a decision of the user which method to choose bearing in mind that the analysis will inevitably involve a compromise between time, cost and accuracy.