
Everyday technology—smartphones, artificial intelligence data centers, wearable health trackers—depends on memory chips that consume enormous amounts of energy.
As our reliance on digital devices grows, so does the challenge of creating memory systems that are smaller, faster, and more efficient.
A new study from Auburn University physicists points toward a solution by exploring how atom-thin crystals can revolutionize memory design.
The research, published in ACS Applied Materials & Interfaces, focuses on memristors—ultra-thin devices that can “remember” past electrical signals.
Unlike traditional memory chips, memristors can switch between conducting electricity like a metal and blocking it like a semiconductor.
This makes them promising candidates for future electronics, where efficiency and speed are critical.
At the heart of the Auburn team’s work are transition metal dichalcogenides (TMDs), crystals that can be peeled down to just a few atoms in thickness.
What makes TMDs especially interesting is how they respond to electrodes—the metals attached to the device that guide electrical signals.
The researchers showed that by carefully selecting these electrodes, they could control how easily the material switches between states. This means engineers could design memory that is both more reliable and consumes less power.
“This is fundamental science with very practical implications,” said Dr. Marcelo Kuroda, associate professor of physics at Auburn and senior author of the study.
“By choosing the right electrode, we can make these devices switch more reliably and at lower power. That is exactly what we need for the next generation of electronics.”
The implications are far-reaching. Because memristors behave somewhat like biological neurons—strengthening or weakening their connections based on activity—they could serve as the building blocks for neuromorphic computing, hardware that mimics how the brain learns.
Such systems could run artificial intelligence tasks far more efficiently than today’s chips. Beyond AI, TMD-based memristors could also be integrated into flexible and wearable devices, enabling medical implants that run for years on a single battery or clothing embedded with adaptive sensors.
To better understand the physics behind this switching behavior, the Auburn team used advanced computer modeling to study how TMDs change at the atomic level.
They discovered that the interaction between electrodes and tiny defects in the crystal lattice, called vacancies, helps ease the transition between insulating and metallic phases. These findings matched experimental results, showing that the effect is both real and potentially useful in future devices.
The research provides a new blueprint for designing memory systems that are not only more powerful but also more sustainable.
“Instead of fighting against the imperfections of these materials, we are learning how to use them,” Kuroda said. “What once seemed like a flaw may actually be the key to building the next generation of technology.”