The magnetic and magnetoresistance properties of ultrathin magnetite films grown on MgO substrate

Abstract

We report on the properties of ultrathin ͑2, 4, 6, and 8 nm͒ epitaxial films of magnetite, Fe 3 O 4 , grown on MgO ͑100͒. Atomic force microscopy image and V-I curves suggest that the films at this thickness are still continuous. The resistivity versus temperature results imply that the conductivity mechanism in all these films is similar. The resistivity of 4 nm thick film is much greater than that of 6 and 8 nm films. The films show ferrimagnetic instead of reported superparamagnetic behavior. The dead layer formed by Mg diffusion between MgO substrate and magnetite films and also the dead layer on the top uncapped film could be the possible reasons for the anomalous resistivity and magnetic properties of the ultrathin films. The effect of the "dead layer" in the thinner film is relatively greater than the one in the thicker film and should lead to a lower magnetoresistance.

Magnetic Properties of Ultrathin Magnetite Films Grown by Molecular Beam Epitaxy

“…We find that all the films down to 2 nm thickness exhibit a ferromagnetic behavior. Observation of a substantial (Giant) increase (95%) in magnetization for a 5-nm Fe O films is in stark contrast to the earlier notion of a reduction in magnetization or a superparamagnetic behavior [5], [9], [11]. Lower surface roughness of the substrates is a key to the nonobservation of superparamagnetism as opposed to previous reports.…”

“…The positive intercept on the y axis (1483 nm.emu/cm ) is the magnetic contribution to the film arising from thickness dependent contributions. Our data suggests that the earlier proposition of a dead layer model for the Fe O /MgO interface was an oversimplification [11]. The inverse thickness dependence of effective magnetization (M) given by the model (1) to separate out distinct bulk and interface contributions is questionable, in particular for the case of Fe O films on MgO (100) [11].…”

“…Our data suggests that the earlier proposition of a dead layer model for the Fe O /MgO interface was an oversimplification [11]. The inverse thickness dependence of effective magnetization (M) given by the model (1) to separate out distinct bulk and interface contributions is questionable, in particular for the case of Fe O films on MgO (100) [11]. Apart from the interfaceand surface-related contribution for ultrathin films, an additional thickness dependent magnetization contribution could result from a totally different origin, i.e., from the thickness dependence of antiphase domain size.…”

“…As the film thickness increases, the magnetization decreases. This is a curious observation in a sense that much of the earlier published work related to ultra thin magnetite films suggested either the formation of a magnetically dead layer or the presence of superparamagnetism [9], [11]. In these earlier investigations, reduction in the magnetization was ascribed to the frustrated exchange at the APBs.…”

“…A recent report by Zhou et al [11] Digital Object Identifier 10.1109/TMAG. 2008.2003173 Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org.…”