Nonvolatile ternary memory devices were fabricated using the composite of polystyrene (PS) and graphene oxide(GO) as active layers, which have an reliable intermediate state. The current-voltage (I-V) curves of the indium tin oxide (ITO)/PS+GO/Al device under the external applied voltages exhibited current tri-stability with three conductivity states, which clearly revealed ternary memory performance. Under the stimulus of the external voltage, a stable intermediate conductivity state was observed. In the write-read-erase-read test, the ITO/PS+GO/Al device exhibited rewritable, nonvolatile, ternary memory properties. The resistance as functions of the time indicated that three conductivity states held for 2 × 10⁵ s, suggesting that the good stability of the ITO/PS+GO/Al devices. HRTEM and XPS observation indicated that the Al top electrode reacted with oxygen within in GO.

Nonvolatile memory devices based on active layers of poly(vinyl alcohol) (PVA) + graphene oxide (GO) hybrid composites have been fabricated. The performance of the ITO/PVA + GO/Al device was compared with that of the ITO/PVA/Al device. The ITO/PVA + GO/Al device showed excellent performance compared to the ITO/PVA/Al device (an ON/OFF resistance ratio of 1.2 × 10² at 1 V, V_SET ∼ −1.45 V and V_RESET ∼ 3.6 V), with a higher ON/OFF resistance ratio of 3 × 10⁴ at 1 V and lower operating voltages of V_SET ∼ −0.75 V and V_RESET ∼ 3.0 V. Furthermore, endurance performance and write-read-erase-reread (WRER) cycle tests manifest that the presence of GO in ITO/PVA + GO/Al devices makes them have better stability and repeatability. The results show that the performance of hybrid devices can be effectively enhanced by the introduction of GO into the PVA matrix.

The electrical conductance switching behavior and nonvolatile memory effects in poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) PEDOT:PSS and single-wall carbon nanotubes (SWCNTs) composite thin films have been investigated. The effect of doping level on electrical conductance switching behavior of an indium tin oxide/PEDOT:PSS + SWCNTs/aluminum (ITO/PEDOT:PSS + SWCNTs/Al) sandwich structure has been reported. The ITO/(PEDOT:PSS + SWCNTs)/Al devices show current bistability. Moreover, the ON/OFF state current ratio is in an extent of 10²–10⁴. The OFF state current increased with the decreasing SWCNTs content in the composite film, and the ON/OFF state current ratio of the device increases with the increase in SWCNTs content of the composite film. Thus, by varying the doping level of SWCNTs in PEDOT:PSS composite thin films, the electrical conductance switching behavior of ITO/PEDOT:PSS + SWCNTs/Al can be modulated in a controlled manner.

This paper reports the first successful fabrication of an ITO/silkworm hemolymph/Al bio-memristor using silkworm hemolymph as the active layer. Experiments demonstrate that the silkworm hemolymph bio-memristor is a nonvolatile rewritable bipolar memory device with a current switching ratio exceeding 10³. The state of the bio-memristor can be retained for more than 10⁴ seconds and remains stable for at least 500 cycles. Tests of 1/f noise have shown that the resistance switching characteristics of the silkworm hemolymph bio-memristor are related to the formation and breaking of conductive filaments, which result from the migration of oxygen ions and the oxidation and reduction of metal cations in the silkworm hemolymph film. The naturally non-toxic silkworm hemolymph offers advantages for human health, environmental protection, and biocompatibility. The proposed nonvolatile rewritable bio-memristor based on silkworm hemolymph possesses great application potential.

A kind of devices Pt/Ag/ZnO:Li/Pt/Ti with high resistive switching behaviors were prepared on a SiO₂/Si substrate by using magnetron sputtering method and mask technology, composed of a bottom electrode (BE) of Pt/Ti, a resistive switching layer of ZnO:Li thin film and a top electrode (TE) of Pt/Ag. To determine the crystal lattice structure and the Li-doped concentration in the resulted ZnO thin films, X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) tests were carried out. Resistive switching behaviors of the devices with different thicknesses of Li-doped ZnO thin films were studied at different set and reset voltages based on analog and digital resistive switching characteristics. At room temperature, the fabricated devices represent stable bipolar resistive switching behaviors with a low set voltage, a high switching current ratio and a long retention up to 10⁴ s. In addition, the device can sustain an excellent endurance more than 10³ cycles at an applied pulse voltage. The mechanism on how the thicknesses of the Li-doped ZnO thin films affect the resistive switching behaviors was investigated by installing conduction mechanism models. This study provides a new strategy for fabricating the resistive random access memory (ReRAM) device used in practice.

Nanocarbon materials have the advantages of biocompatibility, thermal stability and chemical stability and have shown excellent electrical properties in electronic devices. In this study, Al/MWCNT:GQD/ITO memristors with rewritable nonvolatile properties were prepared based on composites consisting of multiwalled carbon nanotubes (MWCNTs) and graphene quantum dots (GQDs). The switching current ratio of such a device can be tuned in two ways. Due to the ultraviolet light sensitivity of GQDs, when the dielectric material is illuminated by ultraviolet light, the charge capture ability of the GQDs decreases with an increasing duration of illumination, and the switching current ratio of the device also decreases with an increasing illumination duration (10³–10). By exploiting the charge capture characteristics of GQDs, the trap capture level can be increased by increasing the content of GQDs in the dielectric layer. The switching current ratio of the device increases with increasing GQD content (10–10³). The device can be programmed and erased more than 100 times; the programmable switching state can withstand 10⁵ read pulses, and the retention time is more than 10⁴ s. This memristor has a simple structure, low power consumption, and enormous application potential for data storage, artificial intelligence, image processing, artificial neural networks, and other applications.