An interface type multi-state resistive variable memory based on electrode stack and its preparation method

Abstract

The invention belongs to the field of multi-state storage, and discloses an electrode stack-based interface type multi-state resistive variable memory and a preparation method thereof. The multi-state resistive variable memory includes a substrate and a bottom deposited sequentially on the substrate from bottom to top electrode, oxygen ion storage layer, active electrode stack layer, and top electrode; the active electrode stack layer includes several electrode layers arranged in sequence from bottom to top, and several electrode layers are made of active metal materials with different activities, and the bottom electrode layer and The oxygen ion storage layer is in contact and the uppermost electrode layer is in contact with the top electrode. Under the action of an applied voltage, the oxygen ions in the oxygen ion storage layer can oxidize the active electrode stack layer step by step, change the oxidation thickness of the active electrode stack layer, and then change the interface barrier height of the active electrode stack layer, resulting in different conductivity. state, realizing a resistive variable memory with multiple resistance states.

Description

technical field [0001] The invention belongs to the field of multi-state storage, and relates to an interface type multi-state resistive memory based on an electrode stack and a preparation method thereof. Background technique [0002] At this stage, the field of artificial intelligence still uses the traditional von Neumann architecture and implements binary operations. The information transmission process between the memory and the processor still needs to be realized through the bus, which means that the data transmission on the bus will generate a large amount of power. consumption loss. In comparison, the power consumption of the human brain is much lower when processing the same amount of data, so people gradually realize that artificial intelligence based on the von Neumann architecture cannot achieve true brain-like intelligence, only by optimizing the hardware structure. Only brain-like computing can be realized. The human brain nervous system is composed of vario...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the disadvantages of poor electrical performance stability of the existing multi-state resistive memory due to the randomness of the formation and fusing of conductive filaments in the above-mentioned prior art, and to provide an interface-type multi-state multi-state based on electrode stacks. Resistive state memory and its preparation method

Images