LAKE WALES Fla.—Everybody already knows that semiconductors are quickly approaching the atomic-level at under 5 nanometers, but most proposed solutions are based on variations-on-a-theme, such as going to a different "semiconductor" like graphene. Why not scrap semiconductors, instead, and use tunneling field effect transistors (TFETs)? The answer is that most materials require cryogenic cooling to make TFETs, according to Professor Yoke Khin Yap at Michigan Tech. Yap, however, has found a room-temperature solution using quantum-dot studded nanotubes.
Professor Yoke Khin Yap at Michigan Tech explains how his transistors switch on and off without semiconductors, as well as there potentially superior performance. (Source: Michigan Tech)
Michigan Tech (Michigan Technological University, Houghton) is not all the way there yet but does have a room-temperature tunneling FET proof-of-concept using iron quantum-dots aligned on boron-nitride nanotubes. Yip claims this solution can not only replace semiconductors but will be flexible enough to create super-small wearable technologies that will perform at levels beyond our wildest imaginations for semiconductors today.
Yap's lab is working toward ultra-small flexible electronics that eliminate semiconductors in favor of the more flexible capabilities of metallic quantum dots and isoelectronic crystals. According to Yap, iron quantum dots (QDs) decorating boron-nitride nanotubes (BNNTs) can use tunneling from quantum-dot to quantum-dot at ultra-low turn-on voltages, to switch the way that transistors do, but using a flexible low-power substrate with absolutely zero leakage current.
Iron quantum dots (yellow, above and grey, below) stud a boron-nitride nanotube, creating transistors without semiconductors by using quantum tunneling between dots when a voltage threshold is exceeded, thus turning-on the transistors without semiconductors. The electron can tunnel between quantum dots in a row, or an find its own path around a nanotube with randomly placed dots.
(Source: Michigan Tech, Sue Hill)
"We already know that the turn-on voltage for a typical QD-BNNT channel can be below 0.1 volts," Yap told EE Times. "But for the current proof of concept work, it is higher at (about 15 volts) due to the long channel length on our STM-TEM [scanning tunneling microscope-tunneling electron microscope] holder."
Using the STM-TEM Yap's team has observed the quantum tunneling in action on bendable nanotubes, which they say will be perfect for flexible wearables. In addition to having zero-leakage when "off," when "on" the channel current encounters zero resistance. Unlike an ordinary transistor channel, the electrons are merely tunneling from quantum dot to quantum dot, making devices based on them ultra-low power since almost no energy is lost as heat when the transistors is on or off.
Using a transmission electron microscope (TEM) the researchers observed quantum tunneling on bendable nanotubes--ac mat >>>Mmaterial could make wearable tech better. Credit: Mi‘‘]
(Source: Michigan Tech)
Yap's team has also observed the same action using gold quantum dots on BNNTs, using the same insulating boron-nitride nanotubes as the flexible tunneling channels. However, the iron quantum dots exert an averaging effect on the tunneling pathways--for randomly studded nanotubes--making their action more consistent and unaffected by bending.
"Iron has a higher melting point than gold and is more convenient to make a uniform distribution of quantum dots on BNNTs by annealing. We still can anneal gold quantum dots for the same purpose, but the current work chose to use iron for better distribution," Yap told us.
The purpose of the current work is to demonstrate that Fe QDs-BNNTs is the flexible channels that behave the same as Au QDs-BNNTs in a two-terminal configuration. These Fe QDs-BNNTs can be put down as T-FETs in the future as we demonstrated for Au QDs-BNNTs.
There is no gate here. The paper reads: "our back gate configuration is not ideal for such an investigation where only moderate gate effect was demonstrated25. Therefore we plan to study these topics with an alternative gate configuration for both Fe and Au QDs-BNNTs in the future." I wonder how the authors justify using "transistor" as the title of a paper reporting a 2-terminal device and how a reputable journal is ok with that.
Here's his justification for being vague: "There are many other factors that determine the mobility, including the properties gate oxide, distance between QDs, etc. I am trying not to speculate a number here before we have the experimental evidence. Stay tune :)"
Thanks Colin. Sadly they didn't mention even an order of magnitude estimate on on future carrier mobility , because they are orders of magnitude far from reasonable mobility , so just being "much higher" doesn't mean a lot .
Alex, here is the authors answer to your questioin about disall mobility: "One unique feature of this type of TFET channel is that switching is more efficient at very short channel length due to the smaller self capacitance and tunneling resistance. This is in contrast to the short channel effect in semiconducting FETs. The low mobility and high turn-on voltages in the proof of concept work here are related to the long channel length tested in the STM-TEM holder. We believe that the mobility will be much higher in actual transistors with gate electrodes and shorter channel lengths (10-100nm)."
Interesting find Colin, sound like a great transistor structure . But looking into the article , it has carrier mobility of 3.2*10^-7 which sounds very bad(if i'm undersnading it right) , probably because they choose larger transistors . But how well do they think this will scale down to real transistors ?
Semiconductors are one the way out, if you believe in quantum-tunneling transistors on nanotubes like Michigan Tech. These TFETs are different than others you may have been researching and are still years away from commercialization, but they dream the dream of zero leakage and sub-0.1 volt operation.