Publication date: 22nd July 2026
The rapid advance of artificial intelligence demands energy-efficient hardware beyond traditional von Neumann architectures, positioning memristors as promising candidates to emulate biological synapses. Metal halide perovskites (MHPs) are promising candidates for memristive applications owing to their mixed ionic-electronic conductivity, yet their inherent instability and tendency toward abrupt resistive switching remain critical barriers to neuromorphic implementations. Here, we present a nanocomposite strategy in which MAPbBr3 nanocrystals are synthesized in situ within a nickel acetate matrix through a one-step, annealing-free, and glovebox-free process, enabling ambient-stable memristive devices. By modulating the perovskite volume fraction, the ionic-electronic dynamics are tuned, inducing a transition from abrupt, non-volatile digital switching in polycrystalline MAPbBr3 to gradual, volatile analog switching in the nanocomposite. Impedance spectroscopy and photophysical analyses attribute this transition to a shift from ion migration-driven switching in the bulk material to interfacial charge trapping and detrapping in the nanocomposite. The resulting volatile memristors emulate key synaptic functionalities, including short- and long-term plasticity, paired-pulse facilitation, and spike-dependent weight modulation, with stable performance over 1000 cycles under ambient conditions. Exploited in a reservoir computing framework, the nanocomposite memristor achieves ~89% classification accuracy on noisy digit patterns with a linear readout, rising to 98% with a deeper network. This scalable, solution-processed approach offers a viable route to analog memristors for neuromorphic and edge computing.
P.F.B. acknowledges his Grant CIACIF/2022/183 funded by the Generalitat Valenciana and, as appropriate, by “ESF Investing in your future”. This work was supported by the European Research Council (ERC) under the European Union's Horizon Europe programme (Grant No. 101171478, project PhoenixPV). PPB thanks Generalitat Valenciana for the funding via Plan Gent-T (grant ESGENT 010/2024). Financial support by the Spanish Ministry of Science and Innovation (CEX2021-001230-S grant funded by MCIN/AEI/10.13039/501100011033) is gratefully acknowledged.
