News
News Categories

Moore’s Law is over? Researchers just made a working 1nm transistor

By Koh Wanzi - on 7 Oct 2016, 12:03pm

Moore’s Law is over? Researchers just made a working 1nm transistor

A schematic diagram showing the channel and carbon nanotube gate. (Image Source: Sujay Desai/UC Berkeley)

Physics can be such a bugger to deal with. It’s no secret that chipmakers are quickly running up against the limits of current silicon technology – beyond 7nm, quantum tunneling makes it effectively impossible for transistors to exist in an off state with no electrons flowing. As a result, it’s actually news that researchers at Lawrence Berkeley National Laboratory have managed to build a functional 1nm transistor gate, which they say is the smallest working transistor to date.

Moore’s Law has governed the computing industry for years, but its progress has slowed in recent. Intel’s 10nm Cannonlake processors have been delayed till either 2017 or 2018, and the company has been stuck on 14nm for three generations of chips now. More tellingly, it also killed off its tick-tock upgrade cycle in favor of a more measured process-upgrade-optimization model that would give it more time to shift to smaller process nodes.

One oft talked about solution is to explore using various silicon alternatives, and that’s exactly what the Berkeley lab team has done. They used carbon nanotubes and molybdenum disulfide, with the hollow carbon nanotube serving as the gate to control the flow of electrons between the source and drain. Molybdenum disulfide has a crystal lattice structure just like silicon, but electrons flowing through it are heavier as compared to when they flow through the latter, so their flow can be controlled using smaller transistor gates below 5nm. 

A transmission electron microscope image of a cross section of the transistor. (Image Source: Qingxiao Wang/UT Dallas) 

Of course, this research is still in its very early stages, and the team is far from creating a working chip with their technology. A 14nm die today has over a billion transistors on it, and the researchers have also yet to come up with a feasible method to mass produce these new 1nm transistors.

Instead, their efforts serve as a valuable proof of concept. New materials can indeed enable smaller transistor sizes, and may very well be a viable avenue to explore as we continue to seek more efficient and powerful chips. Other methods being researched include quantum computers, which have the potential to be exponentially more powerful than anything we have today.

Source: Lawrence Berkeley National Laboratory via The Verge

Join HWZ's Telegram channel here and catch all the latest tech news!
Our articles may contain affiliate links. If you buy through these links, we may earn a small commission.