A new type of memory device could solve the problem of overheating and battery drain in electronics. This is reducing the components to an extreme degree and redesigning their structure. Here, researchers have found a way to reduce energy loss instead of increasing it.
The result is a tiny memory unit that gets better as it gets smaller—something that was once thought impossible. This could pave the way for ultra-efficient smartphones, wearables and AI systems.
If yours The smartphone always feels warm after streaming video or quickly switching between apps, you’re experiencing a fundamental limitation of modern electronics. Every calculation, every file stored, every message depends on microscopic electronic circuits that consume power and emit heat as a byproduct.
At the heart of this problem is computer memory. From phones and laptops to servers that power artificial intelligence, memory systems rely on the movement of electrical charge to represent binary information: the familiar 0s and 1s. The more energy required to switch between these states, the more energy is consumed – and the more heat is generated.
Now, researchers are revisiting a more than half-century-old idea to tackle this inefficiency from the ground up.
A concept ahead of its time?
In 1971, scientists proposed a new type of memory called ferroelectric tunnel junction (FTJ). Unlike conventional memory, which stores information through charge accumulation, FTJs rely on a property called ferroelectricity.
Ferroelectric materials possess an intrinsic electrical polarization that can be switched between two states. Crucially, this shift changes how easily electrons can pass through the material. By switching between these states, the device can encode binary information – without needing as much current.
In principle, this approach offers a major advantage: dramatically reduced energy consumption. Less current means less heat and greater efficiency.
But for decades, FTJs remained largely experimental. The main hurdle was scaling. As devices became smaller—an essential step for modern electronics—the materials used for FTJs began to lose their desirable properties. Performance declined and the concept struggled to compete with established technologies.
A discovery material
The turning point came in 2011, when researchers discovered this hafnium oxide– a material already widely used in semiconductor manufacturing – can exhibit ferroelectric behavior when engineered at extremely small scales.
This was unexpected. Hafnium oxide had long been valued for its insulating properties in transistors, but its ability to maintain electrical polarization in ultra-thin layers opened a new avenue for memory design.
Building on this knowledge, Professor Yutaka Majima and colleagues at the Tokyo Institute of Science have pushed the concept to its limits, developing a ferroelectric memory device. 25 nanometers in width– about one-third the thickness of a human hair.
Rethinking the problem at the nanoscale
Shrinking electronic components to the nanoscale presents a well-known challenge: leak. In such small dimensions, the boundaries between microscopic crystal domains become weak pointsallowing unwanted current to pass. This leakage has historically limited further miniaturization.
Instead of trying to eliminate these imperfections, Majima’s team took a counterintuitive approach. They made the device even smaller.
At this reduced scale, the influence of grain boundaries is reduced and the structure behaves more like a single crystal. To further enhance this effect, the team developed a new fabrication technique. By heating the electrodes during production, they caused them to form a smooth, semi-circular shape, improving the uniformity of the material.
The result is a structure with fewer defects and reduced leakage paths—a solution that turns a long-standing problem into a design advantage.
When smaller is better
The most amazing result is not just that the device works, but that it works better as it decreases.
This directly challenges one of the fundamental assumptions in electronics: that miniaturization inevitably leads to diminishing gains and increased volatility. Instead, researchers have demonstrated that, at least in this case, downsizing can increase efficiency and functionality.
Such behavior can be transformative. If the memory can be made smaller AND more energy efficient at the same time, it can help maintain the long-term trend of improved computing performance without unsustainable increases in power consumption.
Implications for everyday technology
The potential applications are vast. Devices such as smartphones, smartwatches and wearable sensors can operate with significantly lower power requirements. In some cases, battery life can be extended from days to weeks or even months.
For large-scale computing systems—especially those used in artificial intelligence—the benefits can be even greater. Artificial intelligence workloads are extremely energy intensive, requiring large data centers that consume significant electricity. More efficient memory can reduce this footprint, enabling faster processing with less power.
Importantly, hafnium oxide is now compatible with current semiconductor manufacturing processes. This means that the transition from laboratory research to commercial application can be relatively fastcompared to entirely new material systems.
Rethinking borders
For Majima, the work is as much about challenging assumptions as it is about advancing technology.
Scientific and engineering frontiers, he suggests, are often treated as fixed boundaries. But instead, they may reflect gaps in understanding—gaps that can be bridged by rethinking the problem.
His team’s success illustrates this principle. Rather than admit that miniaturization would degrade performance, they asked whether shrinking even further might produce a different result.
The answer, it turns out, is yes.
A low-power future for computing?
As digital technologies continue to expand—from personal devices to global AI infrastructure—the demand for energy-efficient computing is becoming increasingly urgent. The environmental and economic costs of powering these systems are increasing rapidly.
Advances in memory design, such as ferroelectric tunnel junctions based on hafnium oxide, offer a glimpse of another future—one where computing power increases without a corresponding increase in energy use.
There is still work to be done before such devices become widespread. Scaling up production, ensuring durability and integrating with existing architectures will present challenges.
But the principle is established: sometimes, the way forward lies not in working around limitations, but in redefining them entirely.
