ICMCTF2015 Session C4-1: Thin Films for Energy Related Applications

The efficient use of renewable energies represents a major economic and environmental issue. Among the potential solutions, the Dye Sensitive Solar Cells (DSSC) present many advantages. In order to improve the efficiency of DSSC, TiO2 nanoparticles, which are usually used as photo-anode, could be replaced by nanostructured TiO2 thin films. Indeed, a photo-anode of ordered porous nano-columnar TiO2 would provide large surface area for dye absorption, fast electron transfer path [1] and would lead to a better impregnation of the organic polymer used as electrolyte, leading to a better charge collection [2]. In this work, nanostructured TiO2 thin films are synthesized by reactive magnetron sputtering combined with a Glancing Angle Deposition (GLAD) setup in order to control both the phase constitution and the morhology of the films. The substrate temperature, the substrate bias voltage and the rotation speed were varied in order to determine the best experimental conditions leading to ultra-porous films with anatase TiO2 columns. The chemical composition, the crystalline structure and the morphology of the films were analyzed by XPS, XRD, SEM and AFM, respectively. Preliminary data reveal that the increase of the substrate temperature up to 450°C leads to anatase film with an increase of the roughness degree (from 30 to 35 nm). On the contrary, denser rutile films were obtained when the substrate bias voltage was increased. Starting at -100V, the energy provided by the ions becomes too large and promotes scale structure by merging columns while the roughness decreases sharply (from 30 to 10 nm). On the other hand, by rotating the substrate (from 0.06 to 3 rpm) during the deposition process, films with larger columns (from 20 to 100 nm) and higher degree of roughness (from 30 to 36 nm) were obtained due to an enhanced shadowing effect. A substrate heated at 450°C and biased at a low voltage equal to -50V seems to be an optimal condition to synthesize an anatase columnar and porous thin film.

[1] H-Y Yang and al, Thin Solid Films 518 (2009) 1590–1594.[2] L González-García

Nanostructured silicon has emerged as a leading candidate for improved lithium-ion battery electrode design. The combined highly accessible surface area and nanoscale spacing for volumetric lattice expansion of nanostructured thin films have shown improved cycle lifetime over bulk-like silicon films. Additionally, ultra-thin passivation layers have been reported to increase the longevity and stability of silicon thin film electrodes. Very little in-situ information has been reported on silicon films during the complicated lithiation process. Furthermore, what information that is available has been limited to the study of bulk-like thin films. The advantageous geometry of glancing angle deposited (GLAD) thin films allows for strain management of individual nanostructures in comparison to the bulk response. For this reason, alumina passivated GLAD silicon slanted columnar thin films (SCTFs) were grown for use as working electrodes in half cell electrochemical experiments.

The spatially coherent silicon SCTF electrodes have intrinsic biaxial optical properties. Therefore, generalized ellipsometry was employed to investigate the electrode’s physical response to lithium intercalation during an electrochemical cyclic voltammagram cycled against pure lithium metal in a conductive anhydrous electrolyte solution. In-situ ellipsometric monitoring of directional optical constant changes determined by the homogeneous biaxial layer approach are presented. Both the optical and electrochemical responses express reversible phase and morphological transitions. The ability to nondestructively monitor complex nanostructured thin films during lithium-ion processes provides new avenues for high storage battery electrode design.

Efficient control of p-type doping in InN remains one of the most significant issues towards material implementation in advanced electronic devices. We have demonstrated that doping with Mg with concentration from 1×10¹⁸ cm^-3 to 9.0×10¹⁹ cm^-3 results in p-type conductivity in InN [1-2]. Further increase of Mg concentration leads to switching the conductivity to n-type, but the Mg-induced donor defect has not been identified yet. Our results also showed that crystal quality deteriorates with increasing [Mg] [2]. In order to get further insight into the effect of Mg doping on microstructure we have performed comprehensive transmission electron microscopy (TEM) and high-resolution scanning TEM (HRSTEM) studies of InN:Mg with [Mg] = 2.9×10¹⁹ cm^-3 and InN with [Mg] = 1.8×10²⁰ cm^-3 exhibiting p- and n-type conductivity, respectively. The InN:Mg epitaxial films were grown on sapphire by MBE using an undoped InN buffer layer on a GaN template [1].

TEM reveals that threading dislocation (TD) densities in the InN films with p- and n-type conductivity, i.e. the samples on both sides of the p-type window, are very similar, ~7-8×10⁹ cm^-2. The majority of the TDs originate at the interface between the GaN template and the undoped InN buffer layer. The higher doped, n-type sample ehibits a high density of stacking faults (SFs) of type I₁ and I₂ in the Mg doped InN layer. We further observed a zig-zag contrast in the transition region between the undoped InN buffer and the InN : Mg layer, which may be associated with a change in polarity from In to N. In contrast, no SFs or zig-zag contrast were found in the lower doped, p-type sample. It is believed that Mg induced donor defects are responsible for the universally observed change in InN conductivity from p-type to n-type with increasing [Mg]. However exact identification of such defects has not yet been reported. Our results indicate that TDs are not likely to be the cause for n-type conductivity in InN:Mg since TD densities in both the p-type and n-type samples are very similar. On the other hand SFs in n-type InN, is the only structural difference observed. Mg atoms that have atomic radii much different from N or In, cause change in stacking and produce SFs once the Mg concentration reaches 2.9×10¹⁹ cm^-3. Negatively charged point defect in the vicinity of SF can also contribute to the observed n-type conductivity. Our results suggest that SFs play an important role for crystal quality deterioration and change in conductivity and polarity in InN doped with Mg.

[1] A. Yoshikawa, X. Wang, Y. Ishitani, and A. Uedono, Phys. Status Solidi A 207, 1011 (2010)

[2] M.-Y. Xie et al., J. Appl. Phys. 115, 163504, 2014.

Six flat sample plates with (7.8 cm x 7.8 cm x 0.1 cm) of ferritic stainless steel 430 were polished with emery papers and gamma alumina. The films of zinc sulphide were deposited on the sample plates using improved chemical bath deposition (CBD) method technique at 32^oC to 38^oc under intense solar radiation from 12.00am to 5.00pm respectively. The thermal emittance values of the polished and coated sample plates were determined before and after deposition of films respectively using thermocouple potentiometer. The average thermal emittance of polished samples plates was 0.180 ± 0.01. The thermal emittance values of deposited zinc sulphide (ZnS) thin film vary from 0.150 to 0.190 ± 0.01 and why the thickness of the deposited zinc sulphide (ZnS) thin films varies from 1.790 to 6.340 ± 0.01 µm. The thermal emmitance values compared well with those obtained for selective absorbers using other deposition techniques. The chemical bath deposition techniques could be developed for the deposition of the films at different temperatures with suitable deposition time to fabricate selective surfaces for solar energy applications.

Keynote: Thermal Emittance, Chemically Deposited Zinc Sulphide Thin Films, Surface Solar Energy Colectors, ferritic stainless steel plate.

Among the numerous laser applications, laser surface melting by using a pulsed-laser is an innovative technology in the field of surface treatments. This technique presents many advantages. It only modifies the surface properties by keeping the mechanical properties of the bulk. It requires neither addition of other compounds nor contact, so it is quite economical and it does not pollute the material. It allows the treatment of complex shapes into closed spaces with difficult access.

This treatment consists in focusing a nanopulsed laser beam on the surface of the material, leading to the rather immediate melting of the surface through a micron depth, immediately followed by an ultra-fast solidification occurring with cooling rate up to 10¹⁰ K/s.

By using different techniques of analysis, we showed that the combination of these processes leads to various modifications of surface properties.

Glow discharge optical emission spectrometry (GDOES) and transmission electronic microscopy (TEM) were used to establish the segregation of chemical elements and the growth of a new oxide layer with new properties.

X-Ray diffraction (XRD) with grazing incidence was employed to identify the change of crystallographic structure, scanning electron microscopy (SEM) was performed to promote the diminution of the surface defects and the roughness was also characterized.

Applications in the fields of nuclear, corrosion, and decontamination will be presented. Those results showed that the new surface properties strongly depend on the laser parameters.

To optimize the laser parameter for a deliberate application, it is necessary to model the physico-chemical mechanisms involved during the interaction laser-matter. But the modelling is made difficult by the speed of the phenomena to be observed. We set up a fine instrumentation that allows following those phenomena: the time resolved reflectometry and infrared pyrometry. These two techniques give such information as the duration of the melting bath, the kinetics of temperature and appearance of oxides. These data allow to validate and/or to fail the models.