Scientists from the University of California, Berkeley, have developed and demonstrated a new type of ultra-thin silicon nanowires that dissipate heat far better than currently used technology. The discovery could lead to smaller, faster Microchip, but manufacturing could be a problem.
Silicon in modern electronics is a cheap, abundant and good conductor of electricity. However, it is not a good conductor of heat. And that problem is only amplified in tiny chips packed with billions of transistors.
As Berkeley Lab notes, natural silicon is made up of three main isotopes. About 92% consists of the isotope silicon-28, with the remaining 5% and 3% being silicon-29 and silicon-30, respectively.
Chips made from pure silicon-28 have long been thought to conduct heat better and potentially produce faster and denser electronics. This was tested in the mid-2000s, but the thermal conductivity of the single crystal samples improved by only 10%. Simply put, it's not worth making isotopically pure silicon for such a small profit, so the leftover silicon isotope material is stored at the Berkeley Lab in case other scientists might someday use it.
This happened a few years ago.
The Berkeley scientists are trying to come up with ways to improve heat transfer in chips, and they also want to know if nanowires made of pure silicon-28 might help. They got in touch with the owners of the stored materials and got enough testing materials.
The first test involved bulk silicon-28 crystals in the size of 1 mm, and the results reflected a 10% improvement achieved several years ago. The team then used a process called electrolysis free etching to make natural silicon and silicon-28 nanowires that are just 90 nanometers (billionths of a meter) in diameter, or about 1,000 times thinner than a strand of human hair.
The scientists expected incremental results over previous ones, but they were shocked to see that pure nanowires conducted heat 150 percent better than natural silicon nanowires.
Looking under an electron microscope, the researchers saw a glassy layer of silica on the surface of the silica-28 nanowires. Computational simulations further indicate that in the absence of silicon-29 and silicon-30, phonons cannot escape to the surface, where they are slowed down.
Phonons are described as waves of atomic vibration that carry heat in silicon. When you encounter silicon-29 or silicon-30 with different atomic masses, phonons get confused and slow down, blocking heat transfer. For pure silicon-28, this is no longer a problem.
"To discover that two independent phonon blocking mechanisms -- surface and isotope, previously thought to be independent of each other -- now work in tandem, allowing us to benefit in heat transfer, was very surprising, but also very pleasing," said study leader Junqiao Wu.
The team says their next goal is to determine whether they can control, rather than simply measure, the heat conduction of pure silicon nanowires.