Utilizing polymer wrapped single-walled carbon nanotubes in neuromorphic and memory applications

Loo, Hilbert van (2018) Utilizing polymer wrapped single-walled carbon nanotubes in neuromorphic and memory applications. Master's Thesis / Essay, Applied Physics.

Preview	Text 180814_Thesis_Final (1).pdf Download (35MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (96kB)

Abstract

Artificial Neural Networks recently became of great interest due to the upcoming field of machine learning. Using dedicated hardware that mimics the brain will revolutionize machine learning algorithms. In Artificial Neural Networks, designing Artificial synapses that can maintain several conductance state in a non-volatile way is a major challenge. In this work we present a proof-of-concept of a three terminal artificial synapse that utilizes common current hysteresis in s-SWCNT transistors. The device shows a strong plasticity with a dynamic range spanning several orders of magnitude. The conductance of the transistor can be tuned by applying pulses to the gate terminal of the device. The modulation of the conductance is, to some extent, non-volatile implying that these devices could be used in future artificial neural network. Furthermore in this work we demonstrate that high-end binary memory elements can be designed when s-SWCNT transistors are gated by the ferroelectric polymer P(VDF-TrFE). These binary memory elements are made with solution processable materials, showing potential applications in flexible, low-cost and disposable electronics. The memory on/off ratio is around 10^4 with retention time larger than 10^4s and read/write endurance > 200.

Item Type:	Thesis (Master's Thesis / Essay)
Supervisor name:	Loi, M.A.
Degree programme:	Applied Physics
Thesis type:	Master's Thesis / Essay
Language:	English
Date Deposited:	08 Oct 2018
Last Modified:	10 Oct 2018 10:16
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/18689

Actions (login required)

View Item