Fig. 1: Thin films of Cr2O3 with a strain gradient.
From: Flexomagnetism and vertically graded Néel temperature of antiferromagnetic Cr2O3 thin films
a Schematic of a Cr2O3 thin film prepared on a Al2O3 substrate. The film experiences a strain gradient along its thickness that leads to a gradient of the Néel temperature. The Néel temperature at the top surface is probed via magnetotransport relying on the transversal spin Hall magnetoresistance in a Pt electrode adjacent to the Cr2O3 film. The Néel temperature in the interior of the film is probed via NV magnetometry. b Cross-sectional bright-field TEM image showing the Al2O3/Cr2O3/Pt stack for a 30-nm-thick sample. The sample does not reveal the presence of grain boundaries (see also Supplementary Note 1). c High-resolution TEM of the Al2O3/Cr2O3 interface. Two misfit dislocations are indicated in white. The dislocations are located about 5 nm away from the Al2O3/Cr2O3 interface. d Positron lifetime component τ2 and the corresponding calculated vacancy cluster size (assuming spherical geometry) in the top and bottom regions of the Cr2O3 films of different thickness t. The vacancy cluster size increases for thinner samples, indicating larger strain. Error bars correspond to the standard deviations of the fit. e The gradient of the vacancy cluster size along the film thickness (nVac/t) for the samples of different thickness. This parameter is a qualitative measure of the strain gradient across the sample thickness. The error bars correspond to the standard deviations of the fit. Average values of the in-plane (a; (f)) and out-of-plane (c; (g)) lattice parameters and the respective c/a ratio (h) estimated based on the XRD, Raman spectroscopy and TEM measurements of Cr2O3 samples of different thickness. Error bars of the lattice parameters obtained from XRD and Raman measurements are estimated from instrumental precision of the diffractometer and from the uncertainty of the Raman spectra peak fitting accordingly.(I, j Change of the c/a ratio along the film thickness estimated for the 30-nm-thick (i) and 50-nm-thick (j) samples. The c/a ratio for the pseudomorpic layer is highlighted with the gray background. Error bars of the lattice parameters are estimated from instrumental precision of the diffractometer and from the uncertainty of the peak fitting.
