11
$\begingroup$

I've been browsing The D-Wave 2000Q site when I bumped into this aspect of their quantum computers:

A Unique Processor Environment

Shielded to 50,000× less than Earth’s magnetic field

Why is that relevant? What would happen if it would be much less than 50.000x?

$\endgroup$
7
$\begingroup$

The DWave machine relies heavily on single-flux-quantum digital control for setting up qubit and coupler operating points, and for carrying out the annealing protocol. Any stray magnetic flux, if present while the chip is cooled through its superconducting transition, will be trapped inside the circuit and can cause it to fail.

You can calculate how much shielding you need by requiring the magnetic field inside the shield to be smaller than a flux quantum over the area of the chip. $B = \frac{\Phi_0}{A}$, where $\Phi_0 \sim 2 \cdot 10^{-15} ~ \text{Wb}$ is the flux quantum and $A$ is the area. If the area of the DWave chip is $(2 ~ \text{cm})^2$ (guessing) then $B \sim 5 ~ \text{pT}$. Earth’s field is about $0.25 ~ \mu \text{T}$ so you really want $\times 5 \cdot 10^6$ attenuation of the field. Shielding of 50,000 means that you will have on average about 100 flux quanta that can trap in the chip. Typically people add trapping sites on the chip to sequester the remaining flux in safe areas.

$\endgroup$
  • $\begingroup$ is there an easy way to perform such calculations? or can you provide references to works where this is done? $\endgroup$ – glS Mar 29 '18 at 11:27
3
$\begingroup$

It is relevant to reduce the quantum noise in the system. If the shield strength is more than 50,000x, better the quantum computing system is shielded from the earth's magnetic field and thus better quantum noise reduction., at least, theoretically.

EDIT: Superposition is the heart of quantum computing. Superposition state is susceptible to fluctuating external magnetic fields, thermal fluctuations, radiowave etc., The quantum processor should be in an space where the magnetic field is uniform and stable to avoid quantum noise introduced by the above mentioned factors. Thus, isolating the quantum computing system from its disturbing environment is mandatory.

Achieving a ideal quantum noise free environment is still a daunting task. However, progress made thus far has brought us to experimental realizations of quantum computers. Shielding more than 50,000x earth's magnetic field would reduce the quantum noise induced by earth's magnetic field.

$\endgroup$
  • 3
    $\begingroup$ I'd like to see some elaboration on how such magnetic fields could changes in qubits' entanglement-with-environment, etc. Secondly, I think you should expand on the "at least, theoretically" portion a bit. $\endgroup$ – Sanchayan Dutta Mar 29 '18 at 5:04
  • 1
    $\begingroup$ you should make references to specific quantum computing architectures to claim this. Magnetic fields are only a problem in some contexts (for example, photonic quantum processors would not be significantly disturbed by earth's magnetic fields). Similarly, the sentence "Superposition state is susceptible to fluctuating external magnetic fields, thermal fluctuations, radiowave etc.," does not make much sense without reference to a specific kind of system $\endgroup$ – glS Mar 29 '18 at 11:25
2
$\begingroup$

Flux Noise can be a major source of dephasing for superconducting qubits. If you look at the history of the field this makes complete sense. The ideas behind Superconducting Qubits can be traced to the SQUID, which itself was designed to be a very accurate magnetometer. So in general superconducting qubits tend to be quite sensitive to magnetic fields.

One challenge is to balance this senstivity to magnetic noise with the need to manipulate the qubits. Addressing this challenge is the subject of the Rigetti paper on the Charge- and Flux-Insensitive Tunable Superconducting Qubit .

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.