IBM'S RESEARCH AND DEVELOPMENT department has announced "a major engineering breakthrough" in transistor technology that could transform the mobile device space as we know it, especially wearables.
IBM scientists demonstrated a new way to shrink transistor contacts in chips, thus speeding up the replacing of silicon transistors with carbon nanotubes which the firm has been working on for several years.
The company said that the breakthrough brings it closer to creating fully scaled carbon nanotube technology that will power future computing technologies while increasing performance and "opening a pathway to dramatically faster, smaller and more powerful chips".
Carbon nanotube chips have many benefits over traditional silicon. Transistors in silicon are approaching a point of physical limitation. They have been made smaller year after year, but shrinking the size of the transistor, including the channels and contacts, without compromising performance is becoming increasingly difficult.
Carbon nanotube chips could improve the capabilities of high-performance computers because they allow these contacts to be so small that they are virtually transparent.
This means that the size of the semiconductor can decrease dramatically, while the substrate of carbon nanotubes makes the chip more energy efficient and is a soft and flexible material that could allow new device form factors.
Shu-jen Han, IBM's manager of nanoscale science and technology, told us in an interview that wearable technology is one of the most exciting areas that this technology could transform owing to the unique property of the substrate, allowing new form factors with better performance and battery life.
However, the breakthrough isn't about the carbon nanotube material being a better replacement for silicon, but more of an engineering innovation that addresses part of the problem in successfully rolling out better performing and more efficient chips.
"We know what the issue has been, and the limits of the technology, for years. What we solved here is a device-level issue, a one-dimensional structure. We need to make a wafer of them, a high-quality wafer, which does not exist yet," Shu-jen told The INQUIRER.
The next stage for IBM's research group is to scale up the carbon nanotube technology to make reliable mass produced chips before they can make a difference to businesses and consumers.
Shu-jen said this could take five to 10 years, but could enable big data to be analysed faster and allow cloud data centres to deliver services more efficiently and economically. µ
The Turing cards look to be a major leap over last-gen Pascal GPUs
It has been a long year's wait
Chill without the Netflix
Some would say that's a lot for watching YouTube cat videos