The College of Engineering at Seoul National University announced that a research team led by Professor Lee Jong-ho of the Department of Electrical and Computer Engineering (former Minister of Science and ICT) has had three papers accepted at the 'IEEE IEDM (International Electron Devices Meeting) 2025', the world's most prestigious semiconductor technology conference.

The research presented covers key areas of next-generation semiconductor technology, including △ultra-low power memory devices △hardware-based security semiconductors △intelligent gas sensor systems, and is evaluated as the result of convergence research across materials, devices, and systems. The technologies featured in the papers, including △core device technology for low-power neuromorphic computing △new security technology utilizing the inherent characteristics of semiconductor devices △low-power high-performance sensor technology, are expected to become foundational technologies for future integrated AI systems.

In the first study, Professor Lee's team showcased innovations in ferroelectric memory, a type of non-volatile memory used for information storage. By introducing a pseudomorphic structure in the surrounding thin film of hafnium-zirconium composite oxide (HfZrO2), a ferroelectric material, they achieved world-leading polarization characteristics and low operating voltage simultaneously. This suggests the feasibility of implementing ultra-low power memory and a new AI computing method known as neuromorphic hardware.

The second study achieved the development of Physical Unclonable Function (PUF) technology using magnetic tunnel junction (MTJ), another next-generation non-volatile memory. The research team demonstrated the ability to stably generate and conceal an uncloneable hardware security key by utilizing the breakdown probability characteristics of the device. This security key, which maintains reliability even in high-temperature environments, is regarded as a high-security next-generation semiconductor solution.

The third study, conducted in collaboration with Professor Jung Kyu-won's team from the Department of Advanced Convergence, implemented the world's first mixed gas discrimination system integrating a gas sensor and circuit on a single chip. The research focused on implementing an ultra-small/ultra-low power system by applying an innovative 'Self-Cancellation' circuit structure to the low-power sensor. This system can selectively detect target gases within mixed gases without external signal processing, as demonstrated in a real-world application detecting the spoilage level of eggs in real-time.

All three technologies are easily commercializable due to their compatibility with existing CMOS semiconductor processes. For instance, ultra-low power memory can be applied to the development of low-power AI and neuromorphic chips, security semiconductor technology can be used for authentication and encryption in smartphones, automobiles, and IoT devices, and intelligent sensor technology can be utilized in systems for mobile devices, food, environment, medical, and industrial safety fields. The research achievements of Professor Lee's team, which lay the foundation for implementing low-power, high-security, and high-intelligence semiconductors, are expected to contribute significantly to strengthening national semiconductor competitiveness.

Professor Lee Jong-ho stated, "These studies are the result of a convergent approach encompassing semiconductor materials, devices, circuits, and systems, demonstrating that domestic technology can have global competitiveness in various application fields such as AI, security, and sensors." He added, "We plan to continue research to ensure that these achievements, which connect basic material research to demonstration based on commercial processes, lead to the practical application of next-generation semiconductor technology in the future."

Popular News