Master project - Novel devices for physical neural networks

Conventional silicon-based computer systems have several limitations in machine learning computations using neural networks. Resistive switches (memristors) are new types of devices that emulate the behavior of synapses in the brain, and can significantly improve the performance and energy efficiency of such computations

Recently, it was demonstrated for the first time that incorporation of protons into a material can be utilized to modulate its electrical conductivity, with great potential for memristor applications (Figure 1).1 These types of devices may also pave the way towards human-like artificial general intelligence!

Figure 1: Schematic (a) and SEM image (b) of memristor device with WO3 memristor material.1

Activities and methods

Projects can involve combinations of the following activities and methods. Get in touch for an informal chat about specific projects that suits your interests!

  1. Synthesis and thin film deposition (PLD) of novel memristor materials
  2. Electrical characterization of materials upon proton incorporation
  3. Fabrication of devices using photolithography techniques (in collaboration with MinaLab)
  4. Characterization of materials and devices by XRD, SEM, XPS, etc.
  5. Atomistic simulations using density functional theory (DFT)

Collaboration

Co-supervisors from other groups will be included depending on the main methods.

References

  1.              Yao, X. et al. Protonic solid-state electrochemical synapse for physical neural networks. Nature Communications 11, 3134-3134 (2020). https://doi.org/10.1038/s41467-020-16866-6
Publisert 19. feb. 2025 13:42 - Sist endret 27. mars 2025 16:25
Del på e-post