Every time you swipe your phone, open your laptop, or tap your smartwatch, you’re using devices packed with microscopic transistors—the tiny switches that power modern life. But to make electronics even faster, smaller, and more energy-efficient, engineers are racing to go beyond the limits of silicon, the semiconductor that has fueled the tech revolution for decades.
A promising group of materials known as two-dimensional (2D) semiconductors—so thin they’re just a few atoms thick—has long teased a future filled with ultra-compact, high-performance electronics. These materials could revolutionize everything from smartphones to artificial intelligence hardware. But turning that potential into real, scalable devices has been harder than anyone imagined.
Now, a collaborative team of researchers from National Chung Hsing University, Kansai University, and National Cheng Kung University may have cracked part of the code. In a new study published in Nature Electronics, the team introduced an innovative strategy that could finally solve one of the most stubborn challenges in 2D electronics: integrating high-performance insulators with ultra-thin semiconductors in a reliable and scalable way.
Why 2D Semiconductors Are So Promising—And So Challenging
2D semiconductors, such as molybdenum disulfide (MoS₂), offer incredible advantages. They’re just a few atomic layers thick, which means they can be packed tightly to build smaller devices. They also exhibit exceptional electrical properties, including tunable conductivity and high electron mobility, making them ideal for next-generation field-effect transistors (FETs).
FETs act as digital switches, turning current on or off using an electric field. For these switches to work reliably, however, a critical piece must be in place: the gate dielectric. This insulating layer separates the gate electrode from the semiconductor channel and determines how well the transistor can control the flow of current.
To improve performance, the gate dielectric must have a high dielectric constant (κ)—a measure of how effectively it can store and manage electric energy. The higher the κ, the better the gate can control the channel with less power and less leakage. But most traditional high-κ materials are difficult to pair with atomically thin semiconductors. Their rough surfaces and chemical instability often damage the fragile 2D layers, causing performance to plummet.
The Innovation: Freestanding HZO Membranes
In the new study, lead researchers Che-Yi Lin, Bo-Cia Chen, and colleagues proposed a radical yet elegant solution: what if the dielectric layer didn’t have to be built directly on top of the 2D material at all?
Their idea centers around a material called hafnium zirconium oxide (Hf₀.₅Zr₀.₅O₂), commonly known as HZO. HZO is known for its high κ value and excellent ferroelectric properties—meaning it can maintain a stable electric polarization, a useful trait for memory applications. But growing it directly on 2D semiconductors had always caused problems—until now.
Instead of depositing HZO directly, the researchers created freestanding HZO membranes—thin, flexible films that can be independently formed and then transferred onto a device. These membranes range from 5 to 40 nanometers in thickness and can be placed onto 2D semiconductors like MoS₂ without damaging them. This decouples the manufacturing of the dielectric from the delicate 2D layers and opens the door to new fabrication strategies.
Outstanding Performance in Miniaturized Transistors
The performance of these new devices speaks volumes. A 20-nanometer HZO membrane exhibited a dielectric constant of 20.6 ± 0.5—a high value that surpasses many conventional gate insulators. Even more importantly, the leakage current—a measure of how much unwanted electricity slips through—remained incredibly low, under 2.6 × 10⁻⁶ A/cm², easily beating industry standards outlined by the International Technology Roadmap for Semiconductors.
When paired with MoS₂ in a field-effect transistor, the device demonstrated an on/off ratio of 10⁹ (a billion), showing remarkable control over the flow of electrons. The subthreshold swing, a metric of how sharply the device can turn on and off, was under 60 millivolts per decade, approaching the theoretical limit for silicon transistors.
In a single stroke, the team showed that 2D semiconductors can now be paired with powerful high-κ insulators in a way that’s both scalable and reliable.
Logic Circuits and the Future of Computing
But the innovation didn’t stop with individual transistors. To demonstrate the versatility of their approach, the team built a range of digital logic circuits, including inverters, logic gates, and even a 1-bit full adder—a key building block for arithmetic operations in computing.
They also created a MoS₂ transistor with an incredibly short channel length of just 13 nanometers, which still managed to maintain an on/off ratio of 10⁸ and a subthreshold swing of 70 mV/decade. These kinds of specs are critical for developing logic-in-memory systems, where data storage and computation occur in the same location—dramatically improving energy efficiency and speed.
What the researchers have built is not just a better transistor. It’s a blueprint for a new generation of low-power, high-speed, ultra-compact electronics that could eventually outperform today’s silicon-based devices.
The Road Ahead: Toward Scalable 2D Electronics
Of course, there is still work to be done. Manufacturing at industrial scales, ensuring long-term reliability, and integrating these components into full systems are challenges that remain. But the freestanding HZO approach overcomes one of the biggest roadblocks that has slowed down the adoption of 2D electronics.
It represents a shift in mindset: rather than trying to force traditional materials to work with new semiconductors, researchers are now reimagining how materials interact—leveraging flexibility, transfer techniques, and precision engineering to make the impossible achievable.
A Step Closer to the Post-Silicon Future
Ever since the invention of the transistor in 1947, the march of miniaturization has never stopped. Moore’s Law—predicting a doubling of transistor density every two years—has driven decades of innovation, but its pace is slowing as silicon approaches its physical limits.
2D materials, and breakthroughs like freestanding HZO membranes, offer a path forward. They hold the promise of pushing electronics into a new era where devices are not only smaller, but smarter, faster, and greener.
This study isn’t just about improving one component. It’s about reinventing the architecture of modern electronics. And in doing so, it brings us one step closer to the future we’ve only glimpsed in science fiction—where technology is seamless, efficient, and embedded in every part of our lives.
More information: Che-Yi Lin et al, Integration of freestanding hafnium zirconium oxide membranes into two-dimensional transistors as a high-κ ferroelectric dielectric, Nature Electronics (2025). DOI: 10.1038/s41928-025-01398-y.