A self-rectifying TaOy/nanoporous TaOx memristor synaptic array for learning and energy-efficient neuromorphic systems

The human brain intrinsically operates with a large number of synapses, more than 1015. Therefore, one of the most critical requirements for constructing artificial neural networks (ANNs) is to achieve extremely dense synaptic array devices, for which the crossbar architecture containing an artificial synaptic node at each cross is indispensable. However, crossbar arrays suffer from the undesired leakage of signals through neighboring cells, which is a major challenge for implementing ANNs. In this work, we show that this challenge can be overcome by using Pt/TaOy/nanoporous (NP) TaOx/Ta memristor synapses because of their self-rectifying behavior, which is capable of suppressing unwanted leakage pathways. Moreover, our synaptic device exhibits high non-linearity (up to 104), low synapse coupling (S.C, up to 4.00 × 10−5), acceptable endurance (5000 cycles at 85 °C), sweeping (1000 sweeps), retention stability and acceptable cell uniformity. We also demonstrated essential synaptic functions, such as long-term potentiation (LTP), long-term depression (LTD), and spiking-timing-dependent plasticity (STDP), and simulated the recognition accuracy depending on the S.C for MNIST handwritten digit images. Based on the average S.C (1.60 × 10−4) in the fabricated crossbar array, we confirmed that our memristive synapse was able to achieve an 89.08% recognition accuracy after only 15 training epochs. A high-density array of memristors that can behave in a similar way to the human brain has been developed by scientists in South Korea. The brain processes information very differently and much more efficiently than a computer, and researchers are keen to engineer a system that can emulate a brain network. One possible building block for these artificial brains is the memristor-based synapse. The challenge is to create dense networks of memristors that can work a neural network capable of suppressing an undesired neural signal. Gunuk Wang from Korea University in Seoul and co-workers have shown that using memristor based on nanoporous tantalum oxide bilayer can suppress effectively the undesired neural signals between the artificial synapses, thus enabling the team to create the artificial neural network with high-accuracy and energy-efficient learning capability. A two-terminal self-rectfying TaOy/Nanoporous TaOx memristor synapse was fabricated based on anodization process. The device exhibits high non-linearity, low synapse-coupling (S.C), acceptable endurance, sweeping and retention stability, as well as essential synaptic functions such as long-term plasticity and spiking-timing-dependent-plasticity. Furthermore, crossbar array consisting of the only designed device without any selector shows relatively well-defined switching parameters with acceptable cell uniformity and capability of suppressing undesired pathways. The effect of S.C on recognition accuracy of MNIST patterns was also simulated for the first time. Based on experimental average S.C value, the device exhibited the high accuracy comparable to S.C = 0