I. Stopping Cross Section Additivity for 0-2 MeV ⁴He Ions in Solids. II. Magnetite Thin Films: Fabrication and Electrical Properties

Citation

Feng, Joseph Shao-Ying (1975) I. Stopping Cross Section Additivity for 0-2 MeV ⁴He Ions in Solids. II. Magnetite Thin Films: Fabrication and Electrical Properties. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/1C3P-AH34. https://resolver.caltech.edu/CaltechTHESIS:04062012-150944579

Abstract

Part I:

Rutherford backscattering can be used to determine the depth distribution of the constituent elements in the first micron of a sample. The interpretation of the spectra requires an accurate knowledge of the rate at which the probe ions lose their kinetic energy. The microscopic rate of energy loss, the stopping cross section, has been fairly intensively investigated, both experimentally and theoretically, in elemental targets.

In 1905, Bragg and Kleeman proposed that the rate of energy loss in a compound is a linear superposition of the atomic stopping powers. Because of the experimental difficulties, the uncertainties in the tests of this assumption have been 5-10%. Within the sensitivities of these experiments, Bragg's rule has usually been reported to be valid at high ion velocities (E≳I MeV/amu). We describe two new methods of testing Bragg's rule in which the apparent sensitivity limit is below 1%.

The first test requires that Bragg's rule be extended in the obvious way to include at toys and non-stoichiometric compounds. This experiment requires a multi-layered sample in which the components of these layers can somehow be redistributed. If there is no chemical interaction with the ambient, including the substrate, the total energy loss in this multi-layered structure should be independent of the distribution of the constituent elements. This test was applied to two-layered structures of Au-Ag, Au-Cu, Au-Al, and their alloys or compounds. The total energy loss before and after the two layers were mixed was reproducible to within 1%, as predicted by Bragg's rule.

The second test is particularly useful in those targets in which one of the component elements(or chemical radicals) is not readily available as a separate layer. Some examples that were included in this experiment are the oxides, SiO_2 and Al_2O_3. The analytical procedure required that three assumptions in addition to Bragg's rule be invoked. When this procedure was applied to MgO, SiO_2, Al_2O_3, Fe_2O_3 and Fe_3O_4, it was possible to demonstrate that there is a unique contribution by oxygen to the molecular stopping cross sections of these compounds. However, this value is apparently 6-22% lower than the value expected from the measured stopping cress section of molecular O_2 in the gas phase.

Part II:

A low-temperature process for manufacturing magnetite (Fe_3O_4) thin films by converting hematite(α-Fe_2O_3) thin films is described. The films produced are unambiguously identified as magnetite.

Resistivity, dc Hall effect and transverse magnetoreslstance measurements were performed on these films from 104°K to room temperature. The Verwey transition is observed at 123°K, about 4°K higher than reported for stoichiometric bulk magnetite. The ordinary and extraordinary Hall coefficients are both negative over the entire temperature range, consistent with negatively charged carriers. The extraordinary Hall coefficient exhibits a ρ^(1/3) dependence on the resistivity above T_V and a ρ^(2/3) dependence below T_V. The magnetoresistance is negative at all temperatures and field strengths and its magnitude increases monotonically with the magnetic field at all temperatures. The planar Hall effect signal was below the sensitivity of the present experiment.

One particular anomalous result observed in these measurements is the elevated Verwey transition temperature. To account for this unusual behavior, the Verwey transition in magnetite thin films was investigated by measuring the temperature dependence of the sheet resistivity. It was demonstrated that substrate-induced stresses are responsible for the elevated Verwey transition temperature. The ratio of the resistances in the two states, as evaluated at the transition temperature, is apparently proportional to the thickness of the film and independent of the substrate. The combination of these two results suggests that there is a 600-1200 Å layer that remains in the high conductivity state at all temperatures and that it is probably at the free surface of the magnetite film.

Thesis Files