Bi-layer structure has been widely adopted to improve the reliability of the conductive bridge random access memory (CBRAM). In this work, we proposed a convenient and economical solution to achieve a Ta2O5/TaOx bi-layer structure by using a low-temperature annealing process. The addition of a TaOx layer acted as an external resistance suppressing the overflow current during set programming, thus achieving the self-compliance switching. As a result, the distributions of high-resistance states and low-resistance states are improved due to the suppression of the overset phenomenon. In addition, the LRS retention of the CBRAM is obviously enhanced due to the recovery of defects in the switching film. This work provides a simple and economical method to improve the reliability of CBRAM.

In embedded neuromorphic Internet of Things (IoT) systems, it is critical to improve the efficiency of neural network (NN) edge devices in inferring a pretrained NN. Meanwhile, in the paradigm of edge computing, device integration, data retention characteristics and power consumption are particularly important. In this paper, the self-selected device (SSD), which is the base cell for building the densest three-dimensional (3D) architecture, is used to store non-volatile weights in binary neural networks (BNN) for embedded NN applications. Considering that the prevailing issues in written data retention on the device can affect the energy efficiency of the system’s operation, the data loss mechanism of the self-selected cell is elucidated. On this basis, we introduce an optimized method to retain oxygen ions and prevent their diffusion toward the switching layer by introducing a titanium interfacial layer. By using this optimization, the recombination probability of Vo and oxygen ions is reduced, effectively improving the retention characteristics of the device. The optimization effect is verified using a simulation after mapping the BNN weights to the 3D VRRAM array constructed by the SSD before and after optimization. The simulation results showed that the long-term recognition accuracy (greater than 10⁵ s) of the pre-trained BNN was improved by 24% and that the energy consumption of the system during training can be reduced 25,000-fold while ensuring the same accuracy. This work provides high storage density and a non-volatile solution to meet the low power consumption and miniaturization requirements of embedded neuromorphic applications.