Neuromorphic Impedance Signatures and Oxygen-Related Interfacial Dynamics in Pt/YBCO Memristive Junctions
Abstract Memristive junctions based on high-temperature superconducting cuprates are promising building blocks for oxide-based neuromorphic electronics; however, the small-signal dynamics underlying their state-dependent behavior remain largely unexplored. Here, we use zero-bias impedance spectroscopy (IS) to probe memristive Pt/YBa2Cu3O7–δ (YBCO) junctions programmed into a series of nonvolatile resistive states along and beyond the resistive hysteresis switching loop (RHSL). Over the full temperature range (80–300 K) and for every state, the response is well captured by equivalent-circuit models built from two interfacial blocks, consistent with the two-interface picture of bipolar resistive switching. The low-resistance state requires an interfacial element with a negative-resistance branch, in which a negative differential resistance (NDR) is limited by a slow internal variable acting as a chemical inductance; intermediate states are described by two parallel-RC relaxations whose parameters evolve consistently with a progressive depletion of the interfacial oxygen content. Upon further programming toward high resistance, the most resistive states develop a qualitatively distinct signature, with the Bode phase dropping below –90° as the negative-resistance branch becomes dominant. We interpret this crossover as compatible with two regimes of the coupled ionic–electronic dynamics: a slow response set by the scarcity of oxygen vacancies and a faster one enabled by their abundance. Cross-sectional scanning transmission electron microscopy of an electrically overstressed junction reveals a crystalline-to-amorphous degradation pathway that bounds the structural changes compatible with the milder switching explored here. These results are consistent with field-driven oxygen redistribution at the Pt/YBCO interface acting as the control parameter of both the accessible resistance states and the emergence of impedance responses with a negative-conductance band, and they identify the small-signal spectroscopic fingerprint of a FitzHugh–Nagumo (FHN)-like circuit topology in a complex-oxide memristive junction.
Authors
- Myriam H. Aguirre (ORCID: https://orcid.org/0000-0002-1296-4793)
- C. Acha (ORCID: https://orcid.org/0000-0002-8650-4343)
- P. Paturi (ORCID: https://orcid.org/0000-0002-6240-2801)
- Esteban Acosta Rodríguez
- H. Huhtinen (ORCID: https://orcid.org/0000-0002-2166-5939)
Institutions
- University of Turku (FI)
- Universidad de Buenos Aires (AR)
- Universidad de Zaragoza (ES)
Publication Details
- Journal
- ACS Applied Electronic Materials
- Published
- 2026-09-25
- DOI
- https://doi.org/10.1021/acsaelm.6c01363
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00