KAIST Researchers Map the Physical Limits of Transistor Scaling

KAIST Researchers Map the Physical Limits of Transistor Scaling

KAIST Research Models Subatomic Transistor Scaling to Prevent Electron Leakage and Achieve Thresholds Below 4 Nanometers

A determination to come up with faster and more powerful processors is the motivation to always find a way to scale down physically so more transistors could be crammed onto a precious piece of silicon die. But this pursuit running to the edge of physical limitation. Academy of the University of Science and Technology of Korea, or KAIST for short, has carves out a computer simulation that can be used to chart out these Atomic frontier of physical limitation of transistor scaling.

KAIST Researchers Map the Physical Limits of Transistor Scaling

The biggest challenge in designing semiconductors on the subatomic level is electron leakage. The moment physical barriers in a transistor narrow too sharply, electrons escape through the gates. The leakage wastes power, heats up the transistor, and makes the whole processor less efficient. Researchers at KAIST modeled this by building an atomic level computer that tracks the journey of each individual electron through a matrix of experimental substances.

The simulation was primarily dealing with molybdenum disulfide. This is a super atomic material that is only one atom in thickness. Due to its structure and electrical properties molybdenum disulfide has been proposed as a replacement or addition to the current silicon material in the manufacturing of future microchips. The team used their model to examine the possibility of this material.

The KAIST based simulations’ results show that a transistors limits of operation are largely dictated by its internal layout and the materials specified. The calculations proved that under ideal layout parameters, a transistor consisting of for example molybdenum disulphide could operate electrical gates reliably and perform as expected at dimensions less than 4 nanometers. If the computer driven models can be doped through the manufacture process, then the research will represent a definitive guide to producing extremely small scale, energy effective processors in the next ten years.

About the author

mgtid
Owner of Technetbook | 10+ Years of Expertise in Technology | Seasoned Writer, Designer, and Programmer | Specialist in In-Depth Tech Reviews and Industry Insights | Passionate about Driving Innovation and Educating the Tech Community Technetbook

Join the conversation

Newsletter Subscription