Answers are grounded in this paper and cite their sources.
Neuromorphic computing requires highly reliable and low power consumption electronic synapses. Complementary-metal-oxide-semiconductor (CMOS) compatible HfO 2 based memristors are a strong candidate despite of challenges like non-optimized material engineering and device structures. We report here CMOS integrated 1-transistor-1-resistor (1T1R) electronic synapses with ultrathin HfO 2 /Al 2 O 3 bilayer stacks (<5.5 nm) with high-performances. The layer thicknesses were optimized using statistically extensive electrical studies and the optimized HfO 2 (3 nm)/ Al 2 O 3 (1.5 nm) sample shows the high reliability of 600 DC cycles, the low Set voltage of ∼0.15 V and the low operation current of ∼6 µ A. Electron transport mechanisms under cycling operation of single-layer HfO 2 and bilayer HfO 2 /Al 2 O 3 samples were compared, and it turned out that the inserted thin Al 2 O 3 layer results in stable ionic conduction. Compared to the single layer HfO 2 stack with almost the same thickness, the superiorities of HfO 2 /Al 2 O 3 1T1R resistive random access memory (RRAM) devices in electronic synapse were thoroughly clarified, such as better DC analog switching and continuous conductance distribution in a larger regulated range (0–700 µ S). Using the proposed bilayer HfO 2 /Al 2 O 3 devices, a recognition accuracy of 95.6% of MNIST dataset was achieved. These results highlight the promising role of the ultrathin HfO 2 /Al 2 O 3 bilayer RRAM devices in the application of high-performance neuromorphic computing.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: