Memristive Properties and Synaptic Plasticity of Iodobismuthates: Handwritten Character Recognition and Detection of Vocalizations of Trichosurus Vulpecula in a Physical Reservoir Computer

This work explores the structure-property-function relationship in low-dimensional halobismuthates incorporating various organic cations and halides (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{I}^{-}, \text{Br}^{-},\ \text{Cl}^{-}$</tex>), with a focus on their application in the physical reservoir computing. Structural variation driven by anionic composition was correlated with semiconducting behaviour through experimental and Density Functional Theory (DFT) analyses. Memristive devices based on bismuth iodide complexes exhibited analog plasticity and nonlinear dynamics essential for neuromorphic function. Integrated into multiterminal architectures, these materials performed benchmark tasks-including MNIST classification and raw signal recognition—without preprocessing or trainable layers. The results demonstrate the feasibility of sustainable hardware-level reservoir computing using bismuth-based memristors).