Asymmetrical memristor

Patent metadata

Field	Value
Publication number	US-2016019453-A1
Application number	US-201514799488-A
Country	US
Kind code	A1
Filing date	Jul 14, 2015
Priority date	Jan 14, 2013
Publication date	Jan 21, 2016
Grant date	—

Abstract

Embodiments of the present invention provide a memristor having a first electrode, a second electrode and a memristive layer arranged between the first electrode and the second electrode. Thereby, the memristor is adapted to obtain an asymmetrical current density distribution in the memristive layer.

First claim

1 . An artificial tripartite synapse, comprising:
a memristor, comprising a first electrode, a second electrode and a memristive layer arranged between the first electrode and the second electrode, wherein the memristor is adapted to obtain an asymmetrical current density distribution in the memristive layer; and a field effect transistor; wherein the memristor is connected to a gate of the field effect transistor; wherein the memristive layer comprises an asymmetrical doping density distribution or an asymmetrical trap density distribution in order to obtain the asymmetrical current density distribution in the memristive layer; and wherein the current density varies along an equipotential line within the memristive layer or wherein the current density within the memristive layer is asymmetrical along a current path from the first electrode to the second electrode.

2 . The artificial tripartite synapse according to claim 1 , wherein a conductive cross-section of the memristive layer varies along a current path from the first electrode to the second electrode.

3 . The artificial tripartite synapse according to claim 1 , wherein the memristor comprises an asymmetry with respect to an unchangeable structural feature of the memristor, to obtain the asymmetrical current density distribution in the memristive layer.

4 . The artificial tripartite synapse according to claim 3 , wherein the unchangeable structural feature is a geometrical feature.

5 . The artificial tripartite synapse according to claim 3 , wherein the unchangeable structural feature is a material feature of the memristor which is unaffected by a state of the memristor.

6 . The artificial tripartite synapse according to claim 1 , wherein an area of the first electrode is at least by a factor of 1.5 greater than an area of the second electrode, in order to obtain the asymmetrical current density distribution in the memristive layer.

7 . The artificial tripartite synapse according to claim 1 , wherein an area of the memristive layer contacting the first electrode is at least by a factor of 1.5 greater than an area of the memristive layer contacting the second electrode, in order to obtain the asymmetrical current density distribution in the memristive layer.

8 . The artificial tripartite synapse according to claim 1 , wherein the memristive layer comprises TiO 2 .

9 . The artificial tripartite synapse according to claim 1 , wherein the first electrode and/or the second electrode comprises Ti.

10 . A neural network comprising an artificial tripartite synapse according to claim 1 .

11 . The neural network according to claim 10 , further comprising:
a first neuron; a second neuron; wherein the artificial tripartite synapse is connected in series between the first neuron and the second neuron.

Assignees

• Fraunhofer Ges Forschung
• Univ Ilmenau Tech

Inventors

• Klefenz Frank
• Husar Peter
• Williamson Adam
• Schumann Lars
• Hiller Lars
• Hoerselmann Ingo
• Schober Andreas

Classifications

• G06N3/065

  Analogue means · CPC title

• G06N3/09 Primary

  Supervised learning · CPC title

• G11C2213/52

  Structure characterized by the electrode material, shape, etc. · CPC title

• G06N3/063 Primary

  using electronic means · CPC title

• G11C13/0007

  comprising metal oxide memory material, e.g. perovskites · CPC title