bottom line: Scientists at the University of California, Berkeley have developed and demonstrated a new type of ultra-thin silicon nanowire that has heat-dissipating properties far superior to current technologies. The discovery could lead to smaller and faster microchips, but manufacturing could be a challenge.
Silicon in modern electronics is cheap, widespread, and a good conductor of electricity. This is not However, it is a good conductor of heat, which is not much of an issue considering that excess heat is the natural enemy of electronics. The problem is only exacerbated by tiny microchips containing billions of transistors.
Like Berkeley Lab explains, natural silicon consists of three main isotopes. Approximately 92 percent is silicon-28, while the remaining five percent and three percent are silicon-29 and silicon-30, respectively.
It has long been speculated that chips made from pure silicon-28 could conduct heat better and possibly lead to faster and denser electronics. This was tested in the mid-2000s, but single crystal samples showed only 10 percent better thermal conductivity. Simply put, creating isotopically pure silicon for such a small gain was not worth the money or effort, so the remaining silicon isotope material was put into storage at Berkeley Lab in case other scientists ever needed it.
A few years ago, just such a scenario presented itself.
Scientists at Berkeley were looking for ways to improve heat transfer in chips and wondered if pure silicon-28 nanowire would help. They contacted the owner of the stored material and were able to get enough for testing.
The first trial involved bulk silicon-28 crystals measuring 1 millimeter, and their results reflected a 10 percent improvement achieved several years ago. The team then used a process called chemical etching to create nanowires from natural silicon and silicon-28 that are only 90 nanometers (billionths of a meter) in diameter, about 1,000 times thinner than a strand of a human hair.
The scientists expected a larger gain over previous results, but were shocked to see that pure nanowires conduct heat 150 percent better than natural silicon nanowires. How was it possible?
Electron microscopy observation revealed a glassy layer of silicon dioxide on the surface of the silicon-28 nanowire. Computational modeling experiments also showed that the absence of silicon-29 and silicon-30 prevented phonons from reaching the surface where they could be slowed down.
Phonons are described as waves of atomic vibrations that carry heat through silicon. When they collide with silicon-29 or silicon-30, which have different atomic masses, the phonons get entangled and slow down, making heat transfer difficult. This is no longer a problem with pure silicon-28.
“Finding that two separate phonon-blocking mechanisms — surface and isotopes, previously thought to be independent of each other — now work synergistically to our advantage in thermal conduction is very surprising, but also very gratifying,” the study says. leader Junqiao Wu.
The team’s next goal is to determine if they can control, rather than just measure, thermal conductivity in pure silicon nanowires.
The full study is published in a peer-reviewed scientific journal. Physical Review Letters.