Now You See It Now You Don’t: A Soluble Memristor

This new memristor mimics the activity of neural networks, but can be thrown into water and dissolved if data is at risk of falling into the wrong hands

2 min read
Now You See It Now You Don’t: A Soluble Memristor
Images show the time sequence of a synaptic device dissolving in deionized water at room temperature.
Photo: School of Advanced Materials and Nanotechnology

Neurons are remarkable. Simply by firing in certain patterns and at particular frequencies in concert with one another, they endow us with something incredibly valuable: memory. It’s no wonder that scientists have long sought to mimic neural networks, which are both compact and efficient, with electronic devices. One way is with the development of memristors, devices that retain a “memory” of the direction and amount of electrical charge that has flowed through them. Over the past decade, researchers have made some progress in developing memristors, but the latest advance comes with a twist: This new memristor can dissolve in water.

While such a design could result in more environmentally friendly electronic devices, it is perhaps most useful for those who want to stop valuable information from falling into the hands of an enemy.

“Combining transient materials that can physically disappear on demand within a memristor device can be an effective way to achieve secure storage applications,” explains Hong Wang, a researcher at the School of Advanced Materials and Nanotechnology, at Xidian University. “For example, it can be more convenient for us to throw a transient storage device in water when information security is under serious threat. This [holds] great value for military applications, especially.”

Wang’s team recently published a description of their design for a soluble memristor in IEEE Electron Device Letters [Read the paper on Xplore for free here]. Using a water-assisted transfer printing method, they were able to transfer layers of silver and magnesium oxide, in combination with tungsten-based electrodes, onto transient substrates. They arranged these components so that the end result is a memristor that mimics the signaling between neurons.

Just as a threshold amount of calcium ions dictates how many signaling molecules neurons are released to their neighbors, a threshold amount of voltage can be used to control the flow of silver ions through the memristor.

But with neurons, the frequency at which they fire is key to creating a strong neural network. Neurons that fire in rapid succession are more likely to form stronger connections with one another. This phenomenon is captured by the common saying, “Neurons that fire together, wire together.” This same effect can be seen in the silver–magnesium oxide memristor; applying voltage more frequently strengthens the conductance of the device.

This system closely mimics the short-term plasticity of our own neural system, Wang notes. “It proves that such a transient synaptic device can emulate short-term neural activity effectively, which paves the way to achieve security neuromorphic computing applications.”

This new design will need a bit more work before it’s commercialized, but it does demonstrate a desirable method for stopping information from falling into enemy hands. Simply throw it into some deionized water and the key components will dissolve within 30 minutes.

The Conversation (0)

3 Ways 3D Chip Tech Is Upending Computing

AMD, Graphcore, and Intel show why the industry’s leading edge is going vertical

8 min read
A stack of 3 images.  One of a chip, another is a group of chips and a single grey chip.
Intel; Graphcore; AMD

A crop of high-performance processors is showing that the new direction for continuing Moore’s Law is all about up. Each generation of processor needs to perform better than the last, and, at its most basic, that means integrating more logic onto the silicon. But there are two problems: One is that our ability to shrink transistors and the logic and memory blocks they make up is slowing down. The other is that chips have reached their size limits. Photolithography tools can pattern only an area of about 850 square millimeters, which is about the size of a top-of-the-line Nvidia GPU.

For a few years now, developers of systems-on-chips have begun to break up their ever-larger designs into smaller chiplets and link them together inside the same package to effectively increase the silicon area, among other advantages. In CPUs, these links have mostly been so-called 2.5D, where the chiplets are set beside each other and connected using short, dense interconnects. Momentum for this type of integration will likely only grow now that most of the major manufacturers have agreed on a 2.5D chiplet-to-chiplet communications standard.

Keep Reading ↓Show less