Artificial visual perception neural system using a solution-processable MoS2-based in-memory light sensor

Optoelectronic devices are advantageous in in-memory light sensing for visual information processing, recognition, and storage in an energy-efficient manner. Recently, in-memory light sensors have been proposed to improve the energy, area, and time efficiencies of neuromorphic computing systems. This study is primarily focused on the development of a single sensing-storage-processing node based on a two-terminal solution-processable MoS₂ metal-oxide-semiconductor (MOS) charge-trapping memory structure-the basic structure for charge-coupled devices (CCD)-and showing its suitability for in-memory light sensing and artificial visual perception. The memory window of the device increased from 2.8 V to more than 6 V when the device was irradiated with optical lights of different wavelengths during the program operation. Furthermore, the charge retention capability of the device at a high temperature (100 °C) was enhanced from 36 to 64% when exposed to a light wavelength of 400 nm. The larger shift in the threshold voltage with an increasing operating voltage confirmed that more charges were trapped at the Al₂O₃/MoS₂ interface and in the MoS₂ layer. A small convolutional neural network was proposed to measure the optical sensing and electrical programming abilities of the device. The array simulation received optical images transmitted using a blue light wavelength and performed inference computation to process and recognize the images with 91% accuracy. This study is a significant step toward the development of optoelectronic MOS memory devices for neuromorphic visual perception, adaptive parallel processing networks for in-memory light sensing, and smart CCD cameras with artificial visual perception capabilities.

Nanographene-Based Heterojunctions for High-Performance Organic Phototransistor Memory Devices

Organic phototransistors can enable many important applications such as nonvolatile memory, artificial synapses, and photodetectors in next-generation optical communication and wearable electronics. However, it is still a challenge to achieve a big memory window (threshold voltage response ∆V_th ) for phototransistors. Here, a nanographene-based heterojunction phototransistor memory with large ∆V_th responses is reported. Exposure to low intensity light (25.7 µW cm^-2 ) for 1 s yields a memory window of 35 V, and the threshold voltage shift is found to be larger than 140 V under continuous light illumination. The device exhibits both good photosensitivity (3.6 × 10⁵ ) and memory properties including long retention time (>1.5 × 10⁵  s), large hysteresis (45.35 V), and high endurance for voltage-erasing and light-programming. These findings demonstrate the high application potential of nanographenes in the field of optoelectronics. In addition, the working principle of these hybrid nanographene-organic structured heterojunction phototransistor memory devices is described which provides new insight into the design of high-performance organic phototransistor devices.

Fully Photoswitchable Phototransistor Memory Comprising Perovskite Quantum Dot-Based Hybrid Nanocomposites as a Photoresponsive Floating Gate

Tremendous research efforts have been dedicated into the field of photoresponsive nonvolatile memory devices owing to their advantages of fast transmitting speed, low latency, and power-saving property that are suitable for replacing current electrical-driven electronics. However, the reported memory devices still rely on the assistance of gate bias to program them, and a real fully photoswitchable transistor memory is still rare. Herein, we report a phototransistor memory device comprising polymer/perovskite quantum dot (QD) hybrid nanocomposites as a photoresponsive floating gate. The perovskite QDs offer an effective discreteness with an excellent photoresponse that are suitable for photogate application. In addition, a series of ultraviolet (UV)-sensitive insulating polymer hosts were designed to investigate the effect of UV light on the memory behavior. We found that a fully photoswitchable memory device was fulfilled by using the independent and sequential photoexcitation between a UV-sensitive polymer host and a visible light-sensitive QD photogates, which produced decent photoresponse, memory switchability, and highly stable memory retention with a memory ratio of 10⁴ over 10⁴ s. This study not only unraveled the mystery in the fully photoswitchable functionality of nonvolatile memory but also enlightened their potential in the next-generation electronics for light-fidelity application.

Fast Photoresponsive Phototransistor Memory Using Star-Shaped Conjugated Rod-Coil Molecules as a Floating Gate

With the explosive growth in data generation, photomemory capable of multibit data storage is highly desired to enhance the capacity of storage media. To improve the performance of phototransistor memory, an organic-molecule-based electret with an elaborate nanostructure is of great importance because it can enable multibit data storage in a memory device with high stability. In this study, a series of star-shaped rod-coil molecules consisting of perylenediimide (PDI) and biobased solanesol were synthesized in two-armed (PDI-Sol2), four-armed (PDI-Sol4), and six-armed (PDI-Sol6) architectures. Their molecular architecture-morphology relationships were investigated, and phototransistor memory was fabricated and characterized to evaluate the structure-performance relationship of these rod-coil molecules. Accordingly, the memory devices were enabled by photowriting with panchromatic light (405-650 nm) and electrical erasing using a gate bias. The PDI-Sol4-based memory device showed high memory ratios of 10 000 over 10 000 s and a rapid multilevel photoresponse of 50 ms. This achievement is related to the favorable energy-level alignment, isolated nanostructure, and face-on orientation of PDI-Sol4, which eliminated the charge tunneling barrier. The results of this study provide a new strategy for tailoring nanostructures in organic-molecule-based electrets by using a star-shaped rod-coil architecture for high-performance phototransistor memory.

Localized Surface Plasmon Resonance-Mediated Charge Trapping/Detrapping for Core-Shell Nanorod-Based Optical Memory Cells

For following the trend of miniaturization as per Moore's law, increasing efforts have been made to develop single devices with versatile functionalities for Internet of Things (IoT). In this work, organic optical memory devices with excellent dual optoelectronic functionality including light sensing and data storage have been proposed. The Au@Ag core-shell nanorods (NRs)-based memory device exhibits large memory window up to 19.7 V due to the well-controlled morphology of Au@Ag NRs with optimum size and concentration. Furthermore, since the extinction intensity of Au@Ag NRs gradually enhance with the increase in Ag shell thickness, the phototunable behaviors of memory device were systematically studied by varying the thickness of Ag shell. Multilevel data storage can be achieved with the light assistant. Finally, the simulation results demonstrate that the phototunable memory property is originated from the multimode localized surface plasmon resonance (LSPR) of Au@Ag NRs, which is in consistent with the experimental results. The Au@Ag core-shell NRs-based memories may open up a new strategy toward developing high-performance optoelectronic devices.

Synergistic Effects of Self-Doped Nanostructures as Charge Trapping Elements in Organic Field Effect Transistor Memory

Despite remarkable advances in the development of organic field-effect transistor (OFET) memories over recent years, the charge trapping elements remain confined to the critical electrets of polymers, nanoparticles, or ferroelectrics. Nevertheless, rare reports are available on the complementary advantages of different types of trapping elements integrated in one single OFET memory. To address this issue, we fabricated two kinds of pentacene-based OFET memories with solution-processed amorphous and β-phase poly(9,9-dioctylfluorene) (PFO) films as charge trapping layers, respectively. Compared to the amorphous film, the β-PFO film has self-doped nanostructures (20-120 nm) and could act as natural charge trapping elements, demonstrating the synergistic effects of combining both merits of polymer and nanoparticles into one electret. Consequently, the OFET memory with β-PFO showed nearly 26% increment in the storage capacity and a pronounced memory window of ∼45 V in 20 ms programming time. Besides, the retention time of β-PFO device extended 2 times to maintain an ON/OFF current ratio of 10(3), indicating high bias-stress reliability. Furthermore, the β-PFO device demonstrated good photosensitivity in the 430-700 nm range, which was attributed to the additive effect of smaller bandgap and self-doped nanostructures of β-phase. In this regard, the tuning of molecular conformation and aggregation in a polymer electret is an effective strategy to obtain a high performance OFET memory.