Laboratory based X-ray photoemission core-level spectromicroscopy of resistive oxide memories.

HfO2-based resistive oxide memories are studied by core-level spectromicroscopy using a laboratory-based X-ray photoelectron emission microscope (XPEEM). After forming, the top electrode is thinned to about 1 nm for the XPEEM analysis, making the buried electrode/HfO2 interface accessible whilst preserving it from contamination. The results are obtained in the true photoemission channel mode from individual memory cells (5 × 5 µm) excited by low-flux laboratory X-rays, in contrast to most studies employing the X-ray absorption channel using potentially harmful bright synchrotron X-rays. Analysis of the local Hf 4f, O 1s and Ti 2p core level spectra yields valuable information on the chemistry of the forming process in a single device, and in particular the central role of oxygen vacancies thanks to the spectromicroscopic approach.

Understanding the STM images of epitaxial graphene on a reconstructed 6H-SiC(0001) surface: the role of tip-induced mechanical distortion of graphene.

Epitaxial graphene (EG) grown on an annealed 6H-SiC(0001) surface has been studied under ultra-high vacuum (UHV) conditions by using a combined dynamic-scanning tunneling microscope/frequency modulation-atomic force microscope (dynamic-STM/FM-AFM) platform based on a qPlus probe. STM and AFM images independently recorded present the same hexagonal lattice of bumps with a 1.9 nm lattice period, which agrees with density functional theory (DFT) calculations and experimental results previously reported, attributed to the (6 × 6) quasi-cell associated with the 6H-SiC(0001) reconstruction. However, topographic bumps in AFM images and maxima in the simultaneously recorded mean-tunneling-current map do not overlap but appear to be spaced typically by about 1 nm along the [11] direction of the (6 × 6) quasi-cell. A similar shift is observed between the position of maxima in dynamic-STM images and those in the simultaneously recorded frequency shift map. The origin of these shifts is discussed in terms of electronic coupling variations between the local density of states (LDOS) of EG and the LDOS of the buffer layer amplified by mechanical distortions of EG induced by the STM or AFM tip. Therefore, a constant current STM image of EG on a reconstructed 6H-SiC(0001) surface does not reproduce its real topography but corresponds to the measured LDOS modulations, which depend on the variable tip-induced graphene distortion within the (6 × 6) quasi-cell.

Tip induced mechanical deformation of epitaxial graphene grown on reconstructed 6H-SiC(0001) surface during scanning tunneling and atomic force microscopy studies.

The structural and mechanical properties of an epitaxial graphene (EG) monolayer thermally grown on top of a 6H-SiC(0001) surface were studied by combined dynamic scanning tunneling microscopy (STM) and frequency modulation atomic force microscopy (FM-AFM). Experimental STM, dynamic STM and AFM images of EG on 6H-SiC(0001) show a lattice with a 1.9 nm period corresponding to the (6 × 6) quasi-cell of the SiC surface. The corrugation amplitude of this (6 × 6) quasi-cell, measured from AFM topographies, increases with the setpoint value of the frequency shift Δf (15-20 Hz, repulsive interaction). Excitation variations map obtained simultaneously with the AFM topography shows that larger dissipation values are measured in between the topographical bumps of the (6 × 6) quasi-cell. These results demonstrate that the AFM tip deforms the graphene monolayer. During recording in dynamic STM mode, a frequency shift (Δf) map is obtained in which Δf values range from 41 to 47 Hz (repulsive interaction). As a result, we deduced that the STM tip, also, provokes local mechanical distortions of the graphene monolayer. The origin of these tip-induced distortions is discussed in terms of electronic and mechanical properties of EG on 6H-SiC(0001).