![]() |
|
Investigation of Hematite Film-Substrate Interfaces Using Rutherford Backscattering and Channeling TechniquesS. Maheswaran,(a,b) R. J. Smith,(a,c) S. Thevuthasan, and S. A. Chambers Supported by UWS Nepean Staff Development Fund, NSF, EMSL Operations, EMSP and BES/CSD. The growth of iron oxide thin films with good crystalline quality is of increasing interest due to the applications of these films in heterogeneous catalysis, magnetic thin films, surface geochemistry, corrosion, and integrated microwave devices (Geus 1986; Kung 1989; Fujii et al. 1989). Several high-quality, well-oriented, single-crystal, iron oxide films with various stoichiometry have been synthesized using the molecular beam epitaxial (MBE) growth method, and the structural properties of these have been analyzed by various surface and bulk sensitive techniques (Kim et al. 1997; Gao et al 1997a, b). We have used Rutherford backscattering (RBS) and channeling techniques to investigate the nature of the interface between epitaxially grown a-Fe2O3(0001) thin films and an a -Al2O3(0001) substrate. The lattice parameters along the c-axis are 4.76 and 5.03 Å, for a-Al2O3 and a-Fe2O3 respectively, resulting in a lattice mismatch of 5.4%. The growth of this film on this substrate is expected to be complex due to the strain associated with this large lattice mismatch. In fact, it has been shown that the initial growth was a layer-by-layer growth up to the thickness of ~10 Å and the growth was then transformed into an island growth mode (Gao and Chambers 1997). Finally the islands coalesce to create a well-ordered, uniform thin film. In this extended abstract, we summarize our investigation of the characterization of the film-substrate interface of epitaxially grown (Kim et al. 1997) 7-Å-, 70-Å- and 700-Å-thick a-Fe2O3(0001) films on a -Al2O3(0001) substrates using RBS and channeling techniques. The samples were mounted on a molybdenum backing plate using Ta clips, and a conventional alumel-chromel thermocouple was attached to the backing plate close to the sample for temperature measurements. The samples were heated to 200°-250°C to desorb hydrocarbons from the surface. The backscattering spectrum was collected using a silicon surface barrier detector at a scattering angle of 150° . The primary energy of the ions was 1.04 MeV and the incident ion beam was directed along the normal to the sample surface. Theoretical simulations of surface peak area were carried out using the VEGAS program. This computer program uses Monte Carlo calculations and a Moliere-screened potential to simulate the ion scattering interactions and is described elsewhere (Yi et al. accepted). The root mean square vibrational amplitudes are calculated using the Debye temperature reported for Fe3O4 (Frenken et al. 1986) and the vibrations are assumed to be isotropic. Aligned and random RBS spectra using 1.04 MeV He+ beam for the 700-Å-thick film are shown in Figure 6.8. The minimum yield (cmin) for the film is determined to be 2.7 ± 0.3 %. The first three peaks (at the high energy side) visible in the aligned spectrum correspond to the front Fe surface peak, Fe atoms visible to the ion beam at the interface (back surface of the film), and Al atoms at the interface (substrate surface), respectively. Since the Fe and Al atoms are visible to the ion beam at the interface, there must be some disordering of the iron oxide film at the interface. Although mixing of the substrate and the film is possible at the interface, no evidence for mixing has been observed in the random spectrum within the experimental resolution and uncertainties. The experimental peak area of the front Fe surface peak is determined to be 2.3 atoms/row and this area is approximately the same as the area of the back Fe surface peak. The simulated surface peak area using VEGAS code was 2.7 atoms/row and there is a reasonable agreement between the experimental and theoretical surface peak areas. A single Fe-layer terminated bulk-like hematite structure with appropriate surface relaxations (Thevuthasan et al. 1999) was used in this simulation. Simulation of the back surface peak using disordered interface structures is currently in progress.
Figure 6.9 shows the aligned and random spectra for the 70-Å-thick film, and only one peak is visible in the aligned spectrum. Since the film thickness is much smaller than the 700-Å-film, a single peak is possible due to the overlap of two peaks from the Fe atoms present at the front and the back surfaces of the thin film. The experimental area of this peak is approximately 2.5 atoms/row. The simulated surface peak area using VEGAS code was 2.6 atoms/row, which is in good agreement with the experimental value.
The aligned and random spectra from the 7-Å-thick film are presented in Figure 6.10. Since the surface peak areas are almost the same for the aligned and random spectra, it appears that those on the surface do not shadow the Fe atoms in the deeper layers. Several iron oxide phases, including the non-common iron oxide phase reported for this film (Yi et al. submitted), are currently under investigation.
ReferencesFrenken, J. W., R. M. Tromp, and J. F. Van der Veen, Nucl. Instrum. Methods B 17, 334 (1986). Fujii, T., M. Takano, R. Katano, and Y. Bando, J. Appl. Phys. 66, 3168 (1989). Gao, Y., and S. A. Chambers, J. Cryst. Growth 174, 446 (1997). Gao, Y., Y. J. Kim, S. Thevuthasan, P. Lubitz, and S. A. Chambers, J. Appl. Phys. 81, 3253 (1997a). Gao, Y., Y. J. Kim, S. A. Chambers, and G. Bai, J. Vac. Sci. Technol. A 15, 332 (1997b). Geus, J. W., Appl. Catl. 25, 313 (1986). Kim, Y. J., Y. Gao, and S. A. Chambers, Surf. Sci. 371, 358 (1997). Kung, H. H., Transmission Metal Oxides: Surface Chemistry and Catalysis (Elsevier Science Ltd., 1989). Thevuthasan, S., Y. J. Kim, S. I. Yi, S. A. Chambers, J. Morais, R. Denecke, C. S. Fadley, P. Liu, T. Kendelewicz, and G. E. Brown Jr., Surf. Sci. 425, 276 (1999). Yi, S. I., Y. Liang, S. Thevuthasan, and S. A. Chambers, accepted to Surf. Sci. Yi, S. I., S. Thevuthasan, and S. A. Chambers, submitted to Surf. Sci. Lett.
William R. Wiley Environmental Molecular Sciences Laboratory Feedback: webmaster@emsl.pnl.gov Revised: June 12, 2001 Security & Privacy PNNL-13147 |