# Neuron-inspired flexible memristive device on silicon (100)

## TL;DR

This paper presents the development of flexible aluminium oxide memristive devices on silicon (100) using CMOS processes, aiming to emulate brain-like energy efficiency and facilitate large-scale neural network integration.

## Contribution

It introduces a novel fabrication method for flexible memristive devices on silicon (100) using CMOS-compatible processes for large-scale neural modeling.

## Key findings

- Successful fabrication of flexible aluminium oxide memristive devices.
- Potential for ultra large scale integration (ULSI) of brain-inspired memory devices.
- Advancement towards physical modeling of the human brain.

## Abstract

Comprehensive understanding of the world's most energy efficient powerful computer, the human brain, is an elusive scientific issue. Still, already gained knowledge indicates memristors can be used as a building block to model the brain. At the same time, brain cortex is folded allowing trillions of neurons to be integrated in a compact volume. Therefore, we report flexible aluminium oxide based memristive devices fabricated and then derived from widely used bulk mono-crystalline silicon (100). We use complementary metal oxide semiconductor based processes to layout the foundation for ultra large scale integration (ULSI) of such memory devices to advance the task of comprehending a physical model of human brain.

---
Source: https://tomesphere.com/paper/1706.05645