CNT wrap-gate transistors could extend transistor performance scaling

Share via AddThis
Posted on May 30, 2013
(a) Cross-sectional diagram of the CNTFET illustrating how the CNT channel (cutaway diagram shown at right) is suspended across the silicon trench and contacted on either side by the source/drain. (b) and (c) show cross-sectional TEM images of nanotubes with a gate-all-around geometry, which protects the nanotubes from the influence of neighboring devices and stray charges. Credit: Franklin, et al. ©2013 American Chemical Society

(a) Cross-sectional diagram of the CNTFET illustrating how the CNT channel (cutaway diagram shown at right) is suspended across the silicon trench and contacted on either side by the source/drain. (b) and (c) show cross-sectional TEM images of nanotubes with a gate-all-around geometry, which protects the nanotubes from the influence of neighboring devices and stray charges. Credit: Franklin, et al. ©2013 American Chemical Society

Throughout the ’70s, ’80s and ’90s, transistor performance continually increased according to some simple scaling rules. These rules postulate that transistor size and supply voltage should decrease as power density remains constant, resulting in an overall increase in performance. However, physical limitations stopped supply voltage scaling in the early 2000s, so the simple scaling rules no longer apply. Now any increase in performance comes at the cost of an increase in power consumption, so that transistor performance has leveled off since the mid-2000s.

 

Now in a new study, researchers at the IBM T.J. Watson Research Center in Yorktown Heights, New York, have harnessed the potential of carbon nanotubes as a way to extend the scaling rules and achieve further performance improvements in transistors. One day, transistors made of carbon nanotubes may form the backbone of many of our electronic devices, including smart phones and tablets.

The scaling interruption that transistors encountered in the last decade was largely due to the physical characteristics of the transistors themselves, which are silicon metal-oxide-semiconductor field-effect transistors (MOSFETs). In the past few years, researchers have investigated the possibility of replacing MOSFETs with carbon nanotube field-effect transistors (CNTFETs). These transistors have already demonstrated many attractive characteristics, including good performance at low voltages with channel components of less than 10 nm in length—a scale that silicon MOSFETs cannot physically reach with good performance.

“There are two foremost reasons why CNTFETs provide benefits that MOSFETs cannot: 1) The CNTs are ultrathin-body semiconductors (~1 nm), which allows them to be integrated into aggressively scaled devices without losing control over the current in the channel; and 2) CNTFETs can operate at low supply voltages, meaning they can provide the level of electrical current needed to drive integrated circuits at less voltage than MOSFETs can ever achieve,” coauthor and IBM researcher Aaron D. Franklin toldPhys.org. “For nearly a decade, there has been little to no reduction in the supply voltage for MOSFET technologies—CNTFETs are one of the best options for changing that trend.”

Read more at: Phys.org