Building a quantum computer: experimental techniques#

Author: Alexandra Semposki

Maintainer: Alexandra Semposki

DiVincenzo criteria#

When building a real quantum computer, there are several requirements that need to be addressed to achieve one that is as reliable as possible. David DiVincenzo, while working on this task, came up with a set of these conditions now called DiVincenzo criteria [1]:

DiVincenzo criteria

  1. A physical system that is scalable and possesses well-characterized qubits.

  2. Qubits that can be initialized to a fiducial state (\(|000000...000\rangle\)).

  3. A long decoherence time with respect to the gate operation time.

  4. A universal set of quantum gates.

  5. A qubit-specific measurement capability.

There are also two more criteria that directly relate to quantum communication:

DiVincenzo criteria (cont.)

  1. The ability to interchange between stationary vs flying qubits.

  2. The ability to transmit flying qubits between specific locations reliably.

Let’s break down stationary and flying qubits to better understand the last two rules above.

Stationary vs. flying qubits#

Flying qubits gives the impression of qubits flying around one’s head like a bird—of course, this is not the case, but it paints an interesting picture that actually has some merit. Flying qubits are qubits that send information across a distance without corrupting their messages. This distance must be a macroscopic distance, and these qubits are only given this one task, to send information from one node of the system to another [2].

Stationary qubits, on the other hand, must be able to not only store information, but perform calculations with it [2]. They have a much more widespread use than flying qubits. They must also have good entanglement capabilities.

So, returning to rules 6 and 7 above, Rule 6 references the ability of a quantum computer to switch between flying and stationary qubits, and Rule 7 refers to the ability of these flying qubits to transfer their information in a stable way.

Approaches to fabricating qubits#

Below are a few approaches described in some detail, but there are a few more that we will not get into. These include NV-centers in diamond lattices and NMR. Now, let’s dig into the main experimental implementations of qubits!

Superconducting qubits (IBM, Rigetti)#

Superconductor fabrication has long been one of the most impactful technologies of the 20th century. Superconductors are conductors that possess infinite conductivity, a concept that seems quite far-fetched but is real! It happens when conduction electrons form Cooper pairs at very low temperatures in a material, and therefore no longer need to obey the Pauli Exclusion Principle [3]. This leads to infinite conductivity, and hence superconductivity.

How do we build this in real life? We can construct a circuit with superconducting wires, cool it to very low temperatures (mK level) and voila! we have a small quantum system! This system exhibits quantum properties and discrete energy levels at a macroscopic level [3]. However, we will need these energy levels to possess nonuniform spacing, or our excitement of electrons to other energy levels will not be as controlled as we would want.

We can begin by constructing a circuit like so:

image

Here we have an inductor to store the magnetic field, a capacitor to store electric charge, and superconducting wires. However, this is not enough to create the nonuniform energy levels—to do that, we’ll need to insert what is called a Josephson junction. This is shown below, in place of the inductor:

image

The Josephson junction serves as a capacitor with a non-linear inductance. These are small pieces of insulating material placed between our superconducting materials—hence enabling quantum tunnelling [3]. This also creates the effect of nonuniform energy levels, so our circuit will now work as a robust quantum device! We do still need to have some capacitance control so that our qubit works as part of a quantum computer, however—this is so that the qubit has external controls [3]. To do this, we add a gate capacitor, and make sure that our capacitance is large enough that the energy levels stay separated. This is called entering the transmon regime, and hence our qubit is referred to as a transmon [3]. These devices are built into a quantum computer.

Superconducting QUantum Interference Device (SQUID) : > SQUID devices are instances of superconducting qubits that possess two Josephson junctions directly across from one another. These devices are extremely sensitive to magnetic fields and can detect even extremely weak fields very precisely. : An example is SQUID device

Trapped ions (Ion Q, Honeywell)#

Photonic qubits (Xanadu)#

Quantum dots in microwave cavities (Si-technology)#

References#

[1] Jha, Raghav G. “Notes on Quantum Computation and Information”. arXiv: 2301.09679 [hep-ph]. (Link here).

[2] https://quantumcomputing.stackexchange.com/questions/8900/what-is-a-flying-qubit

[3] Alvaro Ballon, “Quantum computing with superconducting qubits”. https://pennylane.ai/qml/demos/tutorial_sc_qubits.html