Search

Electronic synaptic memristor systems have the potential to bring revolutionary change to traditional computer structures and to lay a solid foundation for the development of computer architectures simulating artificial brains. Among them, silver (Ag) or copper (Cu) filament-based memristor devices have increasingly attracted attention due to their excellent functional properties in plasticity and as memristors. However, the randomly dynamic process of nucleation during device fabrication results in nonuniform switching parameters. Here, we demonstrate the viability of a high-performance neuromorphic memristor device based on a carbon conductive filament mechanism, with the advantages of high switching stability and low power consumption. The memristor is also able to emulate faithfully different functions of artificial synapses, including paired-pulse facilitation (PPF) and spike-timing-dependent plasticity (STDP). According to detailed electron energy loss spectroscopy (EELS) and transmission electron microscopy (TEM) characterization, it is confirmed that carbon conductive filaments are formed in aluminum nitride (AlN) films comprising the middle layer of the memristor. First principles calculations provide insight into the energetics of defects involved in the diffusion of carbon atoms into the AlN film. This work probes the viability of a new physical conduction mechanism for use in neuromorphic memristor performance, with evidence of improved device performance.

The Co-Fe-B/MgO system with perpendicular magnetic anisotropy finds extensive application in modern magnetic memories. Critical issues for the practical usage of magnetic tunnel junctions (MTJs) include the limited thermal stability of bit storage, and low switching efficiency. This study elucidates the deterministic influence of Co-Fe-B composition and the MgO interface on an MTJ's fundamental magnetic properties, which affect both thermal stability and switching current. Furthermore, the authors observe an anomalous trend in saturation magnetization, which is attributed to a change in magnetic anisotropy.

Using density-functional theory calculations, we investigate the dominant defects formed by boron (B) and carbon (C) impurities in a CoFe/MgO/CoFe magnetic tunnel junction (MTJ) and their influence on conductivity and tunneling magnetoresistance (TMR). We find that, in the O-poor conditions relevant to experiment, B forms the substitutional defect ${\mathrm{B}}_{\mathrm{Co}}$ and C forms the interstitial site ${\mathrm{C}}_{\mathrm{i}}$ at the CoFe/MgO interface. The C-doped MTJ is predicted to have a significantly higher TMR than the B-doped MTJ. This is due to interface state densities associated with the majority spin ${\mathrm{\ensuremath{\Delta}}}_{1}$-symmetry bands being more heavily suppressed by the ${\mathrm{B}}_{\mathrm{Co}}$ defects than by the ${\mathrm{C}}_{\mathrm{i}}$ defects. Our results indicate that carbon can serve as a viable alternative to boron as a dopant for MTJ fabrication.

Memristors with nonvolatile memory characteristics have been expected to open a new era for neuromorphic computing and digital logic. However, existing memristor devices based on oxygen vacancy or metal-ion conductive filament mechanisms generally have large operating currents, which are difficult to meet low-power consumption requirements. Therefore, it is very necessary to develop new materials to realize memristor devices that are different from the mechanisms of oxygen vacancy or metal-ion conductive filaments to realize low-power operation. Herein, high-performance and low-power consumption memristors based on 2D WS2 with 2H phase are demonstrated, which show fast ON (OFF) switching times of 13 ns (14 ns), low program current of 1 µA in the ON state, and SET (RESET) energy reaching the level of femtojoules. Moreover, the memristor can mimic basic biological synaptic functions. Importantly, it is proposed that the generation of sulfur and tungsten vacancies and electron hopping between vacancies are dominantly responsible for the resistance switching performance. Density functional theory calculations show that the defect states formed by sulfur and tungsten vacancies are at deep levels, which prevent charge leakage and facilitate the realization of low-power consumption for neuromorphic computing application.