An increasing demand for nonvolatile memory has driven extensive research on resistive switching memory because it uses simple structures with high density, fast switching speed, and low power consumption. To improve the storage density, the application of multilevel cells is among the most promising solutions, including three-dimensional cross-point array architectures. Two-dimensional nanomaterials have several advantages as resistive switching media, including flexibility, low cost, and simple fabrication processes. However, few reports exist on multilevel nonvolatile memory and its switching mechanism. We herein present a multilevel resistive switching memory based on graphene oxide (GO) and MoS 2 fabricated by a simple spin-coating process. Metallic 1T-MoS 2 nanosheets, chemically exfoliated by Li intercalation, were successfully embedded between two GO layers as charge-trapping sites. The resulting stacks of GO/MoS 2 /GO exhibited excellent nonvolatile memory performance with at least four resistance states, >10 2 endurance cycles, and >10 4 s retention time. Furthermore, the charge transport mechanism was systematically investigated through the analysis of low-frequency 1/ f noise in various resistance states, which could be modulated by the input voltage bias in the negative differential resistance region. Accordingly, we propose a strategy to achieve multilevel nonvolatile memory in which the stacked layers of two-dimensional nanosheets are utilized as resistive and charge-storage materials.
Internet‐of‐things (IoT) edge devices with a memristive neuromorphic system can more effectively enhance daily lives. However, cyberattacks remain critical concerns for smart IoT edge devices that process a vast body of information via networks. Herein, a highly secure neuromorphic system is reported, which can be implemented using a physically unclonable function (PUF) that exploits the high entropy achieved via the stochastic switching of a poly(1,3,5‐trivinyl‐1,3,5‐trimethyl cyclotrisiloxane) (pV3D3)‐based memristor. The excellent insulating property of pV3D3 enhances the stochasticity of the tunneling distance for randomly ruptured Cu filaments. The pV3D3 memristor‐based PUF (pV3D3‐PUF) achieves near‐ideal 50% averages for uniformity and uniqueness, excellent reliability under conditions of mechanical stress and water immersion, and reconfigurability‐bolstering security without additional hardware. Using stochastic in‐memory computing, the pV3D3‐PUF shows resilience to machine learning attacks. Furthermore, a cryptography protocol is demonstrated, which enables artificial intelligence service implementation without security issues for PUF‐integrated pV3D3 memristor‐based neuromorphic systems.
Filament type resistive random access memory (RRAM) based on polymer thin films is a promising device for next generation, flexible nonvolatile memory. However, the resistive switching nonuniformity and the high power consumption found in the general filament type RRAM devices present critical issues for practical memory applications. Here, we introduce a novel approach not only to reduce the power consumption but also to improve the resistive switching uniformity in RRAM devices based on poly(1,3,5-trimethyl-3,4,5-trivinyl cyclotrisiloxane) by inserting multilayer graphene (MLG) at the electrode/polymer interface. The resistive switching uniformity was thereby significantly improved, and the power consumption was markedly reduced by 250 times. Furthermore, the inserted MLG film enabled a transition of the resistive switching operation from unipolar resistive switching to bipolar resistive switching and induced self-compliance behavior. The findings of this study can pave the way toward a new area of application for graphene in electronic devices.