Australians Create 1-Atom Silicon Quantum Computing Bit (quBit)
World’s first ever single-atom silicon quantum bit opens the door to new quantum computers based on traditional materials
Using a silicon circuit with a phosphorous donor atom, researchers at the University of New South Wales have achieved a critical step forward in the path to quantum computers.
Past efforts have been limited to making nanostructures to read spin, or to making quBits that stored it. This work represents a bold step forward as it’s reportedly the first single-atom silicon quBit system endowed with both the ability to read and write the bit coherently.
The bit showed excellent coherence, lasting for 200 micro-seconds (0.2 milliseconds), or about 400,000 clock cycles on a 2 GHz CPU. The paper suggests that with tighter control of the phosphorous doping, coherence could be extended to last for seconds.
The quBit system was manufactured using standard complementary metal-oxide-semiconductor (CMOS) techniques. Nanostructures were grown on silicon-dioxide.
The quantum bit read/write nanostructure [Image Source: Nature/UNSW]
Electron spin was first set by 1 Tesla magnetic field — around the intensity of the magnetic field at the surface of a Neodymium magnet (by contrast the Earth’s magnetic field is a near 31 microTeslas at the Equator).
The electron temperature was then brought down to 300 milliKelvin (note to readers: milliKelvin coolers are pretty expensive, as one might think — but work is being done to bring down prices).
Visualization of the electrons set in a certain spin. [Image Source: Nature/UNSW]
Reads were accomplished given a technique called single-shot projective measurement.
The long term goal of this kind of research is to create nanostructure quBits that use traditional processes, with some extra tweaks (cooling, high power magnetic sync) to encode information in the spins of electrons, and then use other nanostructures to read that stored information coherently.
QuBits could then be applied to one of two purposes.
First, quBits could be used with traditional transistor circuitry to provide dense storage, as one quBit (with the right read/write nanostructure equipment) could store multiple states (0, 1, 2, 3, …) (via different electron spins) versus a traditional bit, which only can have two states (0, 1).
Second, quBits could be coupled together to produce a quantum computer capable of solving certain types of problems like integer factorization far faster than traditional computers.
The new work still has some issues before its ready for prime-time — relatively short coherence, the need for intense cooling — but it also is promising given its use of traditional materials and manufacturing techniques
The work was presented in the prestigious peer-review journal Nature. The first author was Jarryd Pla and the senior author was Professor Andrea Morello. The work was funded by the US Army Research Office and the Australian Research Council Centre of Excellence for Quantum Computation and Communication Technology.
Read more: www.dailytech.com