A low power flexible halide perovskite-based threshold switching memristor as an artificial nociceptor

Abstract

The emergence of electronic devices with bionic capabilities has propelled advancements in artificial intelligence, enabling artificial nociceptors to mimic sophisticated human-like self-protective mechanisms, thereby enhancing the performance of humanoid robots in complex tasks. In this study, we introduce a low-power and flexible threshold switching (TS) memristor, employing a halide perovskite (OIHP), as an artificial nociceptor or a pain sensor. The device, characterized by its Au/Ag/poly(methyl methacrylate) (PMMA)/Ag doped halide perovskite (OIHP:Ag)/ITO/polyethylene naphthalate (PEN) structure, demonstrates outstanding ultra-low standby power consumption (∼200 fW) and ultra-low operating power consumption (∼25 nW), attributed to its exceptionally low leakage current (∼50 pA) and low threshold voltage (<0.3 V). Remarkably, it replicates advanced nociceptor behaviours encompassing threshold detection, relaxation, absence of adaptation, allodynia, and hyperalgesia. Moreover, the TS behavior of the device demonstrates outstanding environmental, thermal, and mechanical stability. Comprehensive material analysis techniques such as SEM-EDS, XRD, XPS, and UV-vis spectroscopy confirmed the successful incorporation of Ag⁺ ions into OIHP films, forming the basis for establishing a threshold switching mechanism model. This study is poised to advance the application of flexible artificial nociceptors based on halide perovskite TS memristors and enhance energy efficiency in humanoid robot systems.