PlumX Metrics
Embed PlumX Metrics

Ta/HfO-based Memristor and Crossbar Arrays for In-Memory Computing

Memristor Computing Systems, Page: 167-188
2022
  • 1
    Citations
  • 0
    Usage
  • 2
    Captures
  • 0
    Mentions
  • 0
    Social Media
Metric Options:   Counts1 Year3 Year

Metrics Details

Book Chapter Description

Computing hardware systems based on memristors can eliminate the data shuttling between the memory and processing units, offer massive parallelism, and hence promise substantial boost in computing throughput and energy efficiencies. However, such non-von Neumann computational approach imposes high requirements on specific characteristics of memristive devices. In this chapter, we first describe the electrical behavior of our newly developed Ta/HfO2 memristor, in particular, how it meets most requirements for in-memory computing in artificial neural networks. We then examine the underlying mechanism of this device by using electrical and physical characterizations, and attribute the resistance switching to the composition modulation of conduction channel(s) through motion of both cation and anions. Finally, we showcase large-scale one-transistor-one-resistance switch (1T1R) arrays fabricated by integrating the Ta/HfO2 memristors with foundry-made metal–oxide–semiconductor transistors, and their applications in artificial neural networks and hardware security.

Bibliographic Details

Provide Feedback

Have ideas for a new metric? Would you like to see something else here?Let us know