Hot-Filament Metal Oxide Deposition ( HFMOD ) : A Novel Method for Depositing Thin Films of Metallic Oxides

O presente artigo descreve um novo metodo para a sintese de filmes finos de oxidos metalicos. Um filamento de metal, que pode ser aquecido por uma fonte de corrente alternada e instalado numa câmara de vacuo. Por meio de um fluxometro de massa, oxigenio pode ser admitido no interior da câmara. Da reacao entre o oxigenio e o metal do filamento aquecido, especies de oxidos volateis MexOy, onde Me e o metal, sao formadas, se condensando num substrato colocado proximo ao filamento, formando o filme. Foi observado que os filmes finos de WxOy e MoxOy podem ser depositados de forma satisfatoria por este novo metodo. Embora varias outras tecnicas de analise tenham sido usadas para caracterizar os oxidos, esta nota enfatiza os resultados obtidos por espectroscopia de fotoeletrons de raio-X (XPS).


Introduction
Metallic oxides are important materials from the standpoint of both fundamental and applied science.Particularly, tungsten and molybdenum oxide films have been the focus of extensive scientific investigations due to their prospective technological applications in (i) electrochromic devices, 1,2 (ii) gasochromic sensors, 3,4 and (iii) electrocatalysis. 5,6However, their most intensively investigated property so far is the electrochromism.WO 3 films are considered one of the most viable options in emerging electrochromic technology, 1 being applicable for regulating the throughput of radiant energy in smart windows and antidazzling mirrors.These oxides have also been used as "templates" in the synthesis of composites of the type transition metal oxide/conducting polymer, generating promising results in the area of cathodic materials in secondary lithium batteries. 7veral deposition techniques such as sputtering, [8][9][10] thermal evaporation, 3,11 plasma-enhanced chemical vapor deposition 12 and sol-gel 13,14 have been used to obtain tungsten and molybdenum oxide films.This report describes the development of a new deposition method, which we have called hot filament metal oxide deposition (HFMOD), using a metallic filament heated in a rarefied oxygen atmosphere.The film is formed on a substrate positioned near the filament and the deposition rate is controlled by the filament temperature and the oxygen pressure.Both the thermochemistry of the process and the kinetics of film formation are currently under investigation.It is clear, however, that the film is formed from volatile Me x O y precursors, where Me is the metal, generated on the heated tungsten surface from reactions between oxygen and tungsten.The investigations so far carried out in our laboratory show that the films can be deposited with a good stoichiometry control, with relatively high deposition rates and present good adhesion to both metallic and dielectric substrates.Tests on the electrochromical properties carried out on samples of WO 3 show that their optical efficiency is higher than those of WO 3 films obtained by the above-mentioned techniques.It is also important to remark that this technique differs from a deposition technique called hot filament chemical vapor deposition (HFCVD), which have been used to deposit siloxane 15 and diamond-like films, 16 because the filament used here is not just a "catalyst" used to activate chemical species; it is also a reactant in the reaction.

Experimental
Figure 1 depicts the experimental deposition setup that was designed in our laboratory for the deposition of transition metal oxides.The filament (F), which is made of the metal of the oxide to be deposited, is resistively heated by an ac current supply.Oxygen is admitted to the chamber via an electronic mass flowmeter.Pressure measurements are made using a capacitance manometer.During the depositions the chamber is continuously pumped by a Roots pump and the oxygen pressure is adjusted using the flowmeter.The chamber base pressure is about 2.0 x 10 -2 Pa.Substrate temperatures are measured using a chromel-alumel thermocouple and filament temperatures are determined with an optical pyrometer through a glass viewpoint in the wall of the deposition chamber.
In order to characterize the structure of the deposited films, several techniques were used such as Infrared Reflection-Absorption Spectroscopy (IRRAS), Rutherford Backscattering (RBS), X-ray Photoelectron Spectroscopy (XPS), UV-vis Absorption Spectroscopy, Single Wavelength Ellipsometry (SWE) and Raman Spectroscopy.The electrochromic properties of the films were also investigated by means of spectroelectrochemical measurements in the visible range.Manuscripts describing all these results are currently being prepared and will be submitted to scientific journals.
For the purposes of this report, only XPS results will be presented and discussed.For the XPS analyses, a hemispherical spectrometer using the unmonochromatized Kα X-ray line of aluminum was employed.To investigate the possible tungsten or molybdenum valence states, the 4f-doublet peak or the 3d-doublet peak, respectively, were fitted with Gaussian peaks corresponding to known bonding states of tungsten and oxygen.

Results and Discussion
The XPS data were gathered for a W x O y and a Mo x O y sample deposited with the conditions given in Table 1.
The tungsten 4f XPS spectrum (squares) for sample WO is shown in Figure 2. As can be seen from that figure, the 4f profile can be fit by two Gaussian peaks centered at 37.7 and 35.5 eV which are, respectively, the binding energy of electrons in the 4f 5/2 and 4f 7/2 levels of tungsten in the W +6 valence state, 17 indicating that the film is composed of stoichiometric WO 3 .
Figure 3 presents the XPS spectrum of the Mo3d core levels for sample MoO.As can be seen from that figure, the 3d profile can be fit by two pairs of Gaussian peaks.The most intense pair is centered at 235.8 and 232.7 eV which  are, respectively, the binding energy of electrons in the 3d 3/2 and 3d 5/2 levels of molybdenum in the Mo +6 valence state.The other pair is centered at 234.6 and 231.6 eV corresponding to the binding energy of electrons in the same 3d levels, but in MoO +5 valence state.For both valence states, these binding energies are in close agreement with literature values. 18,19Therefore, it was concluded that the Mo atoms were in mixed valence states, Mo +6 and Mo +5 , with a high predominance of the former over the latter.Thus the film was in the overall MoO x stoichiometry, with x smaller but close to 3. The observation that the film stoichiometry is close to that of MoO 3 , is consistent with the expectation that the film is formed from MoO 3 and MoO 2 species desorbed from the Mo filament and that the desorption rate of the former is greater than that of the latter.
The spectra of Figures 2 and 3 are representative of other spectra of samples prepared with oxygen flow rates in the interval between 6.0 and 21 sccm and the same filament temperatures.Thus the chemical composition of the films does not significantly change in this oxygen flow rate range.

Conclusions
By means of the unexpensive and simple deposition technique described in this note, it was possible to deposit W x O y and Mo x O y films with x close to 3. Several details about the role played by the deposition parameters on the overall structure and properties of the films are underway in our laboratory.

Table 1 .Figure 1 .
Figure 1.Schematic representation of the experimental arrangement inside the chamber.F-filament of 0.4 mm (tungsten) and 0.6 mm (molybdenum) diameter metal wire, shaped into a seven-turn coil of 8 mm diameter and 35 mm length; S -substrate; SH -copper substrate holder of 50 X 30 x 7 mm 3 ; TC -thermocouple; filamentto-substrate separation: 30 mm.

Figure 2 .
Figure 2. XPS spectrum of the W4f core levels for sample WO (Filament temperature of 1320 °C and oxygen pressure of 0.8 Pa).The thicker line is the fitting of the experimental data (squares) by the Gaussian peaks (thin lines).

Figure 3 .
Figure 3. XPS spectrum of the Mo3d core levels for sample MoO (Filament temperature of 1550 °C and oxygen pressure of 1.7 Pa).The thicker line is the fitting of the experimental data (squares) by the Gaussian peaks (thin lines).