Characterization of Magnetic-Field Dependent Electron Transport of Fe3Si Nanodots by Using a Magnetic AFM Probe

Abstract

The application of group-IV based ferromagnetic-nanodots (NDs) such as Fe3-silicide NDs to a floating gate in MOS memories has been attracting much attention from the view point of new functionality to control charging and discharging properties in NDs floating gate by the tunnel magnetoresistance effect between ferromagnetic NDs and gate electrodes [1-2]. In our previous work, we demonstrated the formation of high density Fe3Si-NDs on ultrathin SiO2 by exposing ultrathin Fe films deposited on Si-NDs, which were self-assembled on OH-terminated SiO2 layers by controlling the early stages of low-pressure chemical vapor deposition (LPCVD) using SiH4, to remote H2 plasma (H2-RP) [3]. In this work, we extend our research to the characterization of their local electron transport properties measured with a magnetic probe under external magnetic field.An atomic force microscope (AFM) image taken after the SiH4-LPCVD, and subsequent ultrathin Fe-film deposition and H2-RP exposure confirmed the formation of NDs with an areal density as high as ~2.7×1011 cm-2 and with an average height of ~5.9 nm which were evaluated from a size distribution of NDs obtained from the AFM image. In addition, from in-plane and perpendicular magnetization curves (M–H curves) measurements, a coercivities of 180 and 220 Oe were obtained for the in-plane and perpendicular directions, respectively. To evaluate local electron transport properties, a CoPtCr-coated AFM probe was scanned on the sample surface under application of an external magnetic field to take topographic and current images simultaneously where a DC bias of -2.0 V was applied to the AFM probe with respected to an Al bottom-electrode of the sample. The magnetization of the tip was initially set along the upward direction, and upward (positive) or downward (negative) external magnetic fields were applied to the NDs by placing the NdFeB magnet under the sample. When a positive field of 1.5 kOe was applied to the sample in the same direction as the initial tip magnetization, a clear correlation between the topographic images and current images was observed, namely the clear current image with a high contrast among dots and SiO2 (Fig. 1 (a) and (b)). However, a distinct decrease in the current level was detected when a negative field of 0.35 kOe was applied. Notice that, when a positive field of 1.5 kOe was applied to the sample in the same direction as the first magnetization, an almost complete recovery of the resistive state was observed. We also confirmed a reversal of the tip magnetization at a magnetic field of 1.5 kOe from the observation of the reversed contrast in a magnetic force microscope image of a floppy disk taken with the tip exposed to a field of 1.5 kOe. Based on these results, the increase in the resistance by the application of the magnetic field of 0.35 kOe in the opposite direction to the tip magnetization direction can be interpreted in terms of the reversal of the magnetization direction of the dots because the perpendicular coercivity of the dots is 220 Oe. After the re-magnetization at 1.5 kOe, the magnetization direction of the dots and tip can be aligned in parallel. These results indicate that with the application of a magnetic field of 0.35 kOe, the resistive state can be changed by controlling the magnetization directions of the dots. In conclusion, we evaluated the local electron transport properties of Fe3Si-NDs with the areal density as high as ~1011 cm-2 on an ultrathin SiO2/n-Si(100) by using the magnetic cantilever under the external magnetic field. We confirmed at room temperatures that there is a clear change in the contrast of the current images in which current through the NDs depends on the relative directions of magnetization between the NDs and the magnetic cantilever.Reference[1] Y. Nakamura, et al., Jpn. J. Appl. Phys. 50, 015501 (2011).[2] S. H. Sun, et al., Science 287, 1989 (2000).[3] J. Wu et al., 8th ISCSI, WP2-18 (Sendai, 2019). Figure 1