If geeks love it, we’re on it

Intel feeds transistors the juice

Intel feeds transistors the juice

Intel announced on Saturday that it has made a breakthrough in the quest to make faster transistors use less power by looking to compound transistor materials.

The old'n'busted MOS transistor.

The old'n'busted MOS transistor.

Intel explained that current transistors are Metal-Oxide-Semiconductor (MOS) devices with a controlling gate electrode fabricated with an insulating oxide layer. A thinner oxide layer allows the semiconductor to exhibit greater control over the electrical flow, but often at the expense of higher leakage as electrons tunnel through the insulator. The firm has combated the latter effect by implementing a high-k layer which allows for a hundred-fold reduction in leakage and a 33% increase in switching speeds as compared to the traditional SiO2 oxide layer. This process came to bear in the Core 2 Duo and sees continued use today, but soon chips will be too small to continue with high-k MOS transistors.

The new hotness QWFET transistor.

The new hotness QWFET transistor.

Intel has been researching the possibility of replacing the silicon channel (shown right) in a transistor with a compound material such as indium-gallium-arsenide (InGaAs). This new type of transistor is called a Quantum Well Field Effect Transistor (QWFET), and it sandwiches an extremely thin (~10nm) electrical channel between two high-resistance barriers. Intel’s Mike Mayberry explained that QWFETs using compound materials like InGaAs (called III-V devices, referring to the periodic table) exhibit higher performance at smaller sizes, as it is easier for electrons to navigate their designated channel.

While InGaAs-based or other compound III-V transistors hold great promise, there is still considerable work to be done before they are an economical solution.

“If we can make all of the material integration challenges happen and also make dense devices then III-V technology could replace silicon technology starting around the middle of the next decade. There is still much work to be done to achieve this but stay tuned for further progress,” Mayberry said.

Comments

  1. Butters
    Butters Interesting newsbit.
    ..a breakthrough in the quest to make faster Twisted[sic] Transistors use less power...

    I wonder what "less power" actually is. I know it would be hard to give an accurate figure at the moment, but I'm curious at what their projections would be.

    Also, not a biggie, but I might as well point it out. Paragraph 3, last sentence, "referring to period table", I'm assuming should be periodic table.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!