Single Atom Quantum Computing In Silicon

Using existing Silicon fabrication facilities, it is possible to dope a high purity silicon chip with a single phosphorus donor atom and manipulate the atom using a varying magnetic field to manipulate the quantum spin state of the atom to form a quantum bit or qubit.

The nucleus of the phosphorus atom can store a single qubit for long periods of time in the way it spins. A magnetic field could easily address this qubit using well-known techniques from nuclear magnetic resonance spectroscopy. This allows single-qubit manipulations but not two-qubit operations, because nuclear spins do not interact significantly of each other.

For that, we must transfer the spin quantum number of the nucleus to an electron orbiting the phosphorus atom, which would interact much more easily with an electron orbiting a nearby phosphorus atom. Two-qubit operations would then be possible by manipulating the two electrons with high frequency AC electric fields.

The big advantage of this type of quantum computer, sometimes called the Kane quantum computer after physicist Bruce Kane who suggested the device back in the late 1990's, is that it is scalable. Since each atom could be addressed individually using standard electronic circuitry, it is straightforward to increase the size of the computer by adding more atoms and their associated electronics and then to connect it to a conventional computer.

The disadvantages of course is that the atoms must be placed at precise locations in the Silicon, using a scanning tunneling microscope. The manipulation of the phosphorus atom spin itself is also problematic as this requires powerful magnetic fields which reduces scalability.

But the big unsolved challenge has been to find a way to address the spin of an individual electron orbiting a phosphorus atom and to read out its value.

To do this requires scientists to implant a single phosphorus atom in a silicon nanostructure and place it in a powerful magnetic field at a temperature close to absolute zero, cooling the chip using liquid helium. This makes it possible to flip the state of an electron orbiting the phosphorus atom by irradiating it with microwaves.

The final step, a significant challenge in itself, is to read out the state of the electron using a process known as spin-to-charge conversion.

The end result is a device that can store and manipulate a qubit and has the potential to perform two-qubit logic operations with atoms nearby; in other words the fundamental building block of a scalable quantum computer.


However, some stiff competition has emerged in the 15 years since Kane published his original design.

In particular, physicists have found a straightforward way to store and process quantum information in nitrogen vacancy defects in diamond, which offer the best possibility to make a functional quantum computer as this structure can produced quantum gate operations that can work at room temperature.

Then there is D-Wave Systems, which already manufactures a scalable quantum computer working in an entirely different way that it has famously sold to companies such as Lockheed Martin and Google.

The big advantage of the Australian design is its compatibility with the existing silicon-based chip-making industry. In theory, it will be straightforward to incorporate this technology into future chips.

Currently, the Australian Kane quantum computer has the highest performance capabilities of any solid state qubit.

Due to the ease of reproducing the diamond NV- centers, their ease of operation without using liquid helium to cool the chip as well as their speed using optics and electronics it seems that diamond based quantum computers are providing the biggest competition to the Kane quantum computer in the race to develop a functioning, gate quantum computer.

Ref: arxiv.org/abs/1305.4481: A single-Atom Electron Spin Qubit in Silicon