SK hynix Inc. R&D`s Global Revolutionary Technology Center (RTC) will present two presentations on memory for AI applications at the 2025 VLSI Technology and Circuits Symposium (VLSI 2025), held in Kyoto, Japan, from June 8 to 12, 2025. We anticipate great interest in this topic and look forward to meaningful discussions and collaborations with you at the 2025 VLSI conference site or online connections.

 “Achieving Outstanding Endurance (> 107) in Large-Array Two-Deck 16 nm SOM through Process, Structure, and Design Strategies for Emerging SCM Applications” Sanghyun Ban.

We successfully demonstrated a two-deck structure selector-only memory (SOM) based on a 16 nm half-pitch, achieving endurance exceeding 107 cycles at raw bit error rate (RBER) 200 ppm through process, structure, and design strategies. Our findings reveal that SOM endurance is not only influenced by process factors such as dual functional material (DFM) and encapsulation but also strongly depends on the spike charge. This underscores the importance of both spike charge suppression and process optimizations to enhance SOM performance.

“224 TOPS/W-level Analog Computation in Memory Cell using Hybrid Ferroelectric Tunnel Junction Having Enhanced On-State Conductance” Won-Tae Koo.

Here, we first demonstrate hybrid ferroelectric tunnel junctions (H-FTJs) that combine ferroelectric and resistive switching for analog computation in memory. The key of H-FTJs lies in the modulation of the effective tunneling thickness by the control of oxygen vacancy-based unconnected filaments in HfZrO2. The H-FTJs exhibited high on-state conductance (1.6×103 S/cm2) and on/off ratios (32,000) compared to recent studies on HfZrO2-based FTJs. We also fully integrated one-transistor-one-(H-FTJ) cross-bar arrays, and confirmed their analog multiply-accumulate operations (accuracy 91.9%) with high energy efficiency (224.4 TOPS/W) for inference tasks.

Meanwhile, a short course by Soogil Kim, Vice president of Global RTC, is scheduled for Monday, June 9th, at 1:45 pm. Soogil Kim will present on the topic 'Current Landscape and Future Outlook of Emerging Memory Technologies' covering introduction of the research and development directions that emerging memory research should pursue, considering recent changes in computing environments and memory technology.