"2025 GAA 2nm Mass Production"... Samsung Electronics Shows Confidence in Foundry Technology

Samsung Electronics Holds 2021 Foundry Forum
Confidence in World's First GAA Technology Development
Develops 17nm New Process from Legacy FinFET Process
Expands Applications in DDI and Image Sensors

[Asia Economy Reporter Suyeon Woo] Samsung Electronics held the 'Samsung Foundry Forum 2021,' showcasing next-generation foundry advanced process technology, online on the 6th (local time).

Under the theme 'Adding One More Dimension,' Samsung Electronics introduced new technology strategies including mass production plans for 3nm and 2nm processes based on GAA (Gate All Around) technology and the development of a new 17nm process with FinFET structure. In particular, they revealed a concrete roadmap to introduce the core GAA technology to the 3nm process for the first time in the world in the first half of next year and to apply it to the 2nm process by 2025 for mass production.

They also added plans to further advance process technology, line operations, and foundry services to strengthen competitiveness in the rapidly growing foundry market.

Choi Si-young, President of Samsung Electronics Foundry Business Division, said in his keynote speech, "We will expand production capacity through large-scale investments and continue differentiated technological innovation not only in advanced fine processes such as GAA but also in existing processes. Amid the rapid digital transformation caused by COVID-19, we will provide differentiated value so that customers' diverse ideas can be realized in chips."

This foundry forum demonstrated high industry interest, with more than 500 fabless customers and partners, totaling over 2,000 people, registering in advance?the highest number in the history of foundry forums.

Samsung Electronics: "Introducing GAA in 3nm in 2022 and 2nm in 2025"... Leading the Advanced Process Market

GAA technology is essential for continuing process miniaturization in terms of power efficiency, performance, and design flexibility. Samsung Electronics announced plans to introduce GAA technology to 3nm in the first half of 2022, mass-produce the second generation of 3nm in 2023, and mass-produce the GAA-based 2nm process in 2025, expressing confidence in leading next-generation transistor technology.

In particular, Samsung's proprietary GAA technology, the MBCFET™ (Multi Bridge Channel FET) structure applied to the 3nm process, is expected to improve performance by 30%, reduce power consumption by 50%, and decrease area by 35% compared to the FinFET-based 5nm process. Samsung stated that for the 3nm process, stable production yield has been secured and preparations for mass production are underway.

Development of 17nm FinFET New Process... Expanding Applications to Image Sensors and DDI

Samsung Electronics continues to improve FinFET technology to provide cost-efficient products with competitive advantages across application fields, and announced the new 17nm FinFET-based process at this forum. The 17nm new process is expected to improve performance by 39%, power efficiency by 49%, and reduce area by 43% compared to the 28nm process.

Notably, the 17nm new process can be applied to products such as image sensors and mobile display driver ICs, which mainly use planar transistor-based processes of 28nm or larger, demonstrating potential expansion into various applications.

Samsung Electronics is developing various options for the existing 14nm process, such as 3.3V high voltage and eMRAM (embedded MRAM) support, applicable to MCUs (Microcontroller Units), supporting diversification of FinFET process applications including IoT (Internet of Things) and wearable devices. At the same time, the 8nm RF (Radio Frequency) platform plans to secure leadership in the 5G semiconductor market for sub-6GHz ultra-high frequency (mmWave) products.

Samsung Electronics also plans to hold the SAFE Forum (Samsung Advanced Foundry Ecosystem) online in November to strengthen the ecosystem of foundry customers and partners.

This content was produced with the assistance of AI translation services.