The Amazing Vanishing Transistor Act

Radical changes are in the offing for transistors as their dimensions shrink to a few tens of nanometers

11 min read

A decade from now you won't recognize a transistor even if it's walking toward you up the street, assuming you could see it, of course. The gate length--the marker for gauging how small that CMOS transistor is--will be roughly one-fifth the size of the smallest in production today, only 10 nm instead of today's 50 nm. To get to that size and ensure that the transistor still operates will require many changes:

  • To improve performance, silicon will be mixed with a semiconductor like germanium to produce a more spacious, strained crystalline structure that lets electric charge carriers move faster.
  • To reduce the leakage of current that drives up power consumption, gate oxides will be made of materials with more than eight times the dielectric constant (k) of today's silicon dioxide.
  • For better control of the transistor's on and off states, gates will be of metal, instead of polysilicon.
  • For better control and (again) to reduce power consumption, gates themselves will be doubled up so that two will do the job a single gate does now.

Among these techniques, strained silicon is the only one to have been commercialized so far. The rest are still at various stages of R and D. High-k dielectrics and metal gates could be next on the market as soon as they can be integrated into the manufacturing process. As for the double-gate devices, the jury is still out. Most researchers believe that they will be necessary when gate lengths shrink to 10 nm. But some think that they could be used earlier in portable applications, such as cellphones and handheld devices, to reduce the number of chips and power dissipation or to add capabilities.

Keep reading...Show less

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions

Sleep Can Keep AI From Catastrophic Forgetting

New data-replay strategy prevents AI amnesia

3 min read
silhouette of head laying down with abstract colorful towers inside
iStock

Neural networks can achieve superhuman performance in many tasks, but these AI systems can suddenly and completely forget what they have learned if asked to absorb new memories. Now a new study reveals a novel way for neural networks to undergo sleep-like phases and help prevent such amnesia.

A major challenge that artificial neural networks face is "catastrophic forgetting." When they learn a new task, they have an unfortunate tendency to abruptly and entirely forget what they previously learned. Essentially, they overwrite past data with new knowledge.

Keep Reading ↓Show less

Robot Gift Guide 2022

Your yearly selection of awesome robot gifts

7 min read
A collage of 9 photos of robots, including quadrupeds robots, wheeled robots, and drones.
IEEE Spectrum (Robots: Companies)

It’s been a couple of years, but the IEEE Spectrum Robot Gift Guide is back for 2022! We’ve got all kinds of new robots, and right now is an excellent time to buy one (or a dozen), since many of them are on sale this week. We’ve tried to focus on consumer robots that are actually available (or that you can at least order), but depending on when you’re reading this guide, the prices we have here may not be up to date, and we’re not taking shipping into account.

And if these robots aren’t enough for you, many of our picks from years past are still available: check out our guides from 2019, 2018, 2017, 2016, 2015, 2014, 2013, and 2012. And as always, if you have suggestions that you’d like to share, post a comment to help the rest of us find the perfect robot gift.

Keep Reading ↓Show less

Fourth Generation Digitizers With Easy-to-Use API

Learn about the latest generation high-performance data acquisition boards from Teledyne

1 min read

In this webinar, we explain the design principles and operation of our fourth-generation digitizers with a focus on the application programming interface (API).

Register now for this free webinar!

Keep Reading ↓Show less