Non-volatile computing-in-memory macros that are based on two-dimensional arrays of memristors are of use in the development of artificial intelligence edge devices. Scaling such systems to three-dimensional arrays could provide higher parallelism, capacity and density for the necessary vector–matrix multiplication operations. However, scaling to three dimensions is challenging due to manufacturing and device variability issues. Here we report a two-kilobit non-volatile computing-in-memory macro that is based on a three-dimensional vertical resistive random-access memory fabricated using a 55 nm complementary metal–oxide–semiconductor process. Our macro can perform 3D vector–matrix multiplication operations with an energy efficiency of 8.32 tera-operations per second per watt when the input, weight and output data are 8, 9 and 22 bits, respectively, and the bit density is 58.2 bit µm^–2. We show that the macro offers more accurate brain MRI edge detection and improved inference accuracy on the CIFAR-10 dataset than conventional methods.

The degradation behavior and its mechanisms of E-mode GaN high electron mobility transistors (HEMTs) with p-GaN gate under electrostatic discharge (ESD) stress were investigated. Reverse short-pulse stress was generated by a transmission line pulse (TLP) tester in order to simulate the static electricity. The experiment results show that the reverse short-pulse stress leads to the characteristic degradation of the E-mode GaN HEMTs with p-GaN gate. The values of the threshold voltage and on-resistance increase, and the gate capacitance curve shifts positively. The low-frequency noises (LFNs) were obtained for the E-mode GaN HEMTs with p-GaN gate before and after the reverse short-pulse stress. The concentration of traps was extracted, and it has doubled after 700 cycles. The degradation mechanism could be attributed to the generation of traps at p-GaN/AlGaN heterointerface, AlGaN barrier, and GaN/AlGaN interface. Such an investigation can be a significant reference in the design and application of E-mode GaN power devices.

Aluminum electrolytic capacitors (AEC) are widely used in switching mode power supply (SMPS) as smoothing capacitor, but they are the weakest among power components in DC-DC converter. It is very important to monitor its electrical parameters, mainly the equivalent series resistance (ESR). A monitoring method of capacitor's ESR for buck converters is proposed in this paper. Based on the calculation model of ESR, the target parameter is monitored by sampling output ripple voltage. The proposed method needs no current sensor. Whether the circuit is in continuous conduction mode (CCM) or discontinuous conduction mode (DCM), it is convenient to monitor the capacitor. Implementation of the monitoring system is presented. The experimental results validate the effectiveness of the method.

3D Convolutional Neural Networks (CNNs) has been widely used for medical image analysis such as magnetic resonance imaging (MRI) and video recognition due to their intrinsic 3D characteristics. This letter presents 3D convolution operations realized by an 8-layers 3D vertical resistive random access memory (VRRAM) under a field-programmable gate array (FPGA)-controlled relay-matrix based test platform. As an implementation, 3D Prewitt operators are used for edge surface detection of 3D version MNIST handwritten digits with 16 × 16 × 16 pixels. The experimental results show that 3D convolution kernels can be correctly implemented on our in-house 3D VRRAM with higher parallelism than the conventional architecture. Besides, the proposed 3D VRRAM meets the demands of low power and high capacity for 3D CNN accelerator, paving the way of 3D VRRAM-based processing-in-memory (PIM) architecture for 3D CNNs.

The continued advancement of organic electronic technology will establish organic electrochemical transistors as pivotal instruments in the field of biological detection. Here, we present a comprehensive review of the state-of-the-art technology and advancements in the use of organic electrochemical transistors as biosensors. This review provides an in-depth analysis of the diverse modification materials, methods, and mechanisms utilized in organic electrochemical transistor-structured biosensors (OETBs) for the selective detection of a wide range of target analyte encompassing electroactive species, electro-inactive species, and cancer cells. Recent advances in OETBs for use in sensing systems and wearable and implantable applications are also briefly introduced. Finally, challenges and opportunities in the field are discussed.

Non-filament 3D vertical random access memory (VRRAM) is a promising technology for emerging high-density memory applications, which can significantly improve system performance by simplifying memory hierarchy. Reliability issue is one of the most challenging concerns precluding the commercial application of RRAM. The bulk of RRAM reliability papers focus on the electric stress, while the effect of moisture stress from ambient environment has seldomly been studied. In this brief we examine the aging kinetics of moisture stress through statistical measurements on state-of-the-art 8-layer 3D VRRAMs. The results reveal that RHRS and on/off window declines significantly while RLRS remains unaltered in terms of the average value of multiple devices. Meanwhile, reset current Ireset increase obviously. Moreover, different layers exhibit the same aging kinetics. We speculate the observed degradation is due to the moisture stress introduces defects at HfO2/TaOx interface and brings down the barrier.