May 7, 2024
9:05am - 9:35am
CH01-virtual
Laura Bocher1,Ibrahim Koita1,Tizei Luiz H. G.1,Jean-Denis Blazit1,Xiaoyan Li1,Benoît Corraze2,Julien Tranchant2,Marcel Tencé1,Laurent Cario2,Etienne Janod2,Odile Stephan1
Université Paris-Saclay1,Université de Nantes2
Laura Bocher1,Ibrahim Koita1,Tizei Luiz H. G.1,Jean-Denis Blazit1,Xiaoyan Li1,Benoît Corraze2,Julien Tranchant2,Marcel Tencé1,Laurent Cario2,Etienne Janod2,Odile Stephan1
Université Paris-Saclay1,Université de Nantes2
Taming abrupt resistive transitions in functional oxides is a promising approach for developing advanced information processing and storage systems. V<sub>2</sub>O<sub>3</sub> is considered a prototypical system of metal-to-insulator transitions (MITs) where they can be activated under external stimuli such as temperature (T), pressure, or chemical doping<sup> [1]</sup> but hardly technologically feasible. Recent demonstrations from electric pulses yield MIT in (V<sub>1-x</sub>Cr<sub>x</sub>)<sub>2</sub>O<sub>3</sub> systems with real potential capabilities for non-volatile memories and neuromorphic applications<sup> [2]</sup>. Hence, understanding the V<sub>2</sub>O<sub>3</sub> electronic phase separation and its local mechanisms governing the insulator/metallic (I/M) domain dynamics across the IMTs remains of interest. However, all these electronic transitions rely on their relationship between structural and electronic degrees of freedom. For instance, when cooled below 160 K, V<sub>2</sub>O<sub>3</sub> presents a symmetry breaking, associated with a large volume change (+1.4%) and an MIT yielding a resistivity change of 7 orders of magnitude. This T-activated MIT has been extensively studied at the macroscopic scale <sup>[1]</sup> and remains still a perfect arena to probe <i>in situ</i> the V<sub>2</sub>O<sub>3 </sub>structural and electronic evolutions at a very local scale. Recently, the microscopic electronic coexistence of I/M domains has been mapped in (V<sub>1-x</sub>Cr<sub>x</sub>)<sub>2</sub>O<sub>3</sub> by <i>in situ </i>scanning photoemission spectroscopy <sup>[3]</sup>, PEEM <sup>[4]</sup> and nano-IR <sup>[5] </sup>with 25nm spatial resolution at best. In addition, combined micro-XRD and nano-IR experiments have highlighted competitive mechanisms between structural and electronic contributions during this MIT <sup>[6]</sup>. These latest investigations also demonstrate the cautions and possible experimental limitations when it comes to accurately mapping the dynamics of mechanisms within regions of interest of a few tens of nm by combining different instruments, hence the need to perform structural and electronic experiments within the same instrument.<br/><br/>Advanced monochromated electron spectromicroscopy emerged this last decade as real game-changers for nanomaterials characterization. Here we performed <i>in situ</i> monochromated STEM/EELS experiments on the NION CHROMATEM 200 MC with variable-T options under cryo-conditions thanks to a double-tilt HennyZ cryo-holder using MEMS to vary continuously the temperature conditions across the IMTs. For each probed temperature, we associated mapping of relevant spectroscopic electronic excitations (from IR to soft X-ray) with an ultra-high EELS resolution at the nm scale and the local structural features (symmetry and lattice parameters) determined by 4D STEM nano- and microdiffraction. During low-T thermal cycling through the resistive transition, EELS spectra acquired in the low-loss regime present a characteristic signature at 1.1eV only in the metallic phase. Upon cooling, the abrupt MIT was monitored while the coexistence of I/M nanodomains was evidenced upon heating over a few degrees yielding the propagation of the electronic I/M domain wall. 4DSTEM nanodiffraction experiments reveal the local distribution of monoclinic/rhombohedral phases coexisting in the insulating domains. The observation of the low-T insulating hexagonal phase suggests an analogous paramagnetic insulating (PI) phase of large volume, as confirmed by the jumps in lattice parameters observed at the transition. These PI-like phases emerge also at the I/M domain wall, as a precursor of the metallic phase.<br/><br/>[1] D. B. McWhan et al. <i>Phys. Rev. B</i> <b>2 </b>(1970); [2] E. Janod et <i>al. </i><i>Adv. Funct. Mater</i>. <b>25</b> (2015); [3] S. Lupi,<i>et al</i>., Nat. Commun.,<b>1</b> (2010); [4] A. Ronchi <i>et al.</i>. <i>Phys. Rev. B</i> <b>100</b> (2019); [5] McLeod <i>et al.</i> <i>Nature Physics</i> <b>13</b> (2017); [6] Kalcheim, Y. <i>et al. </i><i>Phys. Rev. Lett. </i><b>122, </b>(2019); [7] The authors acknowledge funding from the EDPIF, the National Agency for Research under the JCJC program IMPULSE and the program of future investment TEMPOS-CHROMATEM, and the European Union’s Horizon 2020 research and innovation program under grant agreement No 823717 (ESTEEM3).