Conductance quantization in memristive devices with electrodeposited Prussian blue-based dielectrics

Le résumé fourni par la source

Identifying new, scalable materials for memristive devices is critical to advance next-generation memory and neuromorphic technologies. In this context, electrodeposited Prussian Blue (PB), a mixed-valence iron hexacyanoferrate compound, is emerging as a highly promising candidate due to its low-cost synthesis, CMOS compatibility, and rich redox chemistry. Here, we report both experimental evidence and theoretical modeling of conductance quantization in memristive devices employing PB as the active dielectric layer. PB thin films were synthesized via electrodeposition and integrated into a conventional metal–insulator–metal ( M I M ) structure (Ag/PB/Au), which exhibits robust and reproducible resistive switching behavior. Notably, we observe quantized conductance steps at integer and half-integer multiples of the quantum of conductance ( G 0 = 2 e 2 / h ), indicative of atomic-scale filament formation and ballistic electron transport. To interpret these findings, we use a quantum transport model based on the finite-bias Landauer formalism, incorporating a series resistance and a non-ideality parameter ( α ), which successfully reproduces the experimental I − V characteristics. An algorithm is also introduced to extract the model parameters directly from measured data. The emergence of quantized states is attributed to the properties of PB due to its open-framework structure, mixed F e 2 + / F e 3 + valence, and reversible ionic mobility, which allow the formation of atomic conduction channels. These results highlight the potential of PB-based memristors for multilevel memory storage and neuromorphic computing, while offering a scalable, CMOS-compatible, and sustainable materials platform.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.