PhD student, Aleksandra Koroleva 1,2
1 Université Grenoble Alpes, CNRS, Grenoble INP, LMGP, 38000, Grenoble, France
2 Université Grenoble Alpes, CNRS, Grenoble INP, TIMA, 38000, Grenoble, France
Study and development of La2NiO4 memristive devices for bio-inspired computing
Abstract
Memristive devices based on functional oxide materials are promising candidates for next-generation memory and neuromorphic computing systems due to their analog switching capabilities, scalability, and energy efficiency. Particularly, the memristive behavior of the p-type semiconductor La2NiO4+δ was recently thoroughly investigated
This seminar will present recent progress in the development of the TiN/La2NiO4+δ/Pt memristive devices, focusing on the physical mechanisms underlying their resistive switching behavior. The La2NiO4+δ thin films are deposited by Pulsed Injection Metal-Organic Chemical Vapor Deposition (PI-MOCVD) on platinized silicon substrates. Post-deposition annealing under controlled atmospheres (Ar and O2) is employed to finely tune the oxygen content and investigate its effect on device performance. Furthermore, the effect of the reduction in deposition temperature is investigated to assess compatibility with CMOS-compatible process windows.
A combination of structural, spectroscopic, and microscopic techniques is employed to analyze the material properties. The memristive performance metrics, such as variability, endurance, retention, and analog capabilities, are evaluated through electrical characterization. The study also explores the switching kinetics and investigates how device geometry and stack design influence the forming voltage, energy consumption, and analog performance. Based on the systematic data, a unified mechanism involving both interfacial redox reactions and conductive filament formation is proposed. The comprehensive analysis of memristive properties combined with the insights into the physical origins of the observed phenomena contributes to the optimization of memristive devices for applications in neuromorphic computing.
[1] T.-K. Khuu, A. Koroleva, A. Degreze, E.-I. Vatajelu, G. Lefèvre, C. Jiménez, S. Blonkowski, E. Jalaguier, A. Bsiesy, M. Burriel, J Phys D Appl Phys 2024, 57, 10LT01.
[2] A. Koroleva, T. Khuu, C. Magén, H. Roussel, C. Jiménez, C. Ternon, E. Vatajelu, M. Burriel, Adv Electron Mater 2024, 10, 2400096.
[3] T. Khuu, A. Koroleva, C. Moncasi, A. Stangl, D. Cooper, G. Lefèvre, F. Wilhelm, A. Rogalev, M. Weber, C. Jiménez, M. Burriel, Adv Electron Mater 2024, 2400313.
Short Bio/CV
.Alexandra Koroleva is a Research scientist with strong experience in the field of functional oxide materials for emerging memory technologies. She completed the Master of Science degree at the Moscow Institute of Physics and Technology, specializing in interdisciplinary research at the interface of nanotechnology and neuroscience. She is currently in the final year of her PhD at Université Grenoble Alpes, leading a collaborative project between LMGP and TIMA laboratories focused on the development of La2NiO4+δ-based memristive artificial synapses for neuromorphic computing hardware. Her skillset includes thin-film deposition (MOCVD, ALD), cleanroom microfabrication, and a broad range of characterization techniques such as XPS, XRD, SEM, AFM, and I-V/T measurements. She is actively engaged in student mentoring, international collaborations, conference presentations, and peer-reviewed publications.