Up until now, scientists have struggled to explain how memristors maintain nonvolatile memory, as models have failed to match experimental results. According to Jingxian Li, a U-M doctoral graduate in materials science and engineering and the study's first author, "While experiments have shown devices can retain information for over 10 years, the models used in the community show that information can only be retained for a few hours."
To address this, the team focused on resistive random access memory (RRAM), a promising alternative to conventional RAM used in computers. The specific type studied, filament-type valence change memory (VCM), sandwiches a tantalum oxide insulating layer between two platinum electrodes. When a voltage is applied, a conductive filament forms, representing a "1" in binary code. Applying a different voltage dissolves the filament, returning the device to a high resistance state or "0."
Initially, it was thought that information persisted because oxygen atoms were slow to diffuse back to the tantalum oxide layer. However, Yiyang Li, U-M assistant professor and senior author of the study, explained, "In these devices, oxygen ions prefer to be away from the filament and will never diffuse back, even after an indefinite period of time. This process is analogous to how a mixture of water and oil will not mix, no matter how much time we wait, because they have lower energy in a de-mixed state."
The team tested memory retention by raising the temperature of the devices, simulating the effects of long-term use. One hour at 250 C equated to roughly 100 years at 85 C, the average temperature for a computer chip. Through atomic force microscopy, they imaged the filaments, which were only five nanometers wide - just 20 atoms across - inside the one-micron wide RRAM devices. "We were surprised that we could find the filament in the device. It's like finding a needle in a haystack," said Li.
The study found that filament size affects memory retention. Smaller filaments, under 5 nanometers, dissolved over time, while larger ones became more stable. This variation could not be explained by diffusion alone, but rather by thermodynamic principles and phase separation.
By leveraging these findings, the team extended memory retention in radiation-hardened chips from one day to over 10 years. These chips are designed for space exploration, where reliable long-term memory is critical. Other potential applications include energy-efficient AI, in-memory computing, and electronic skin (e-skin) for prosthetics and robotics, allowing sensory feedback and tactile capabilities.
"We hope that our findings can inspire new ways to use phase separation to create information storage devices," Li concluded.
Research Report:Study: Thermodynamic origin of nonvolatility in resistive memory
Related Links
University of Michigan
Computer Chip Architecture, Technology and Manufacture
Nano Technology News From SpaceMart.com
| Subscribe Free To Our Daily Newsletters |
| Subscribe Free To Our Daily Newsletters |
