What is a quantum computer and how does it work?

Quantum computing is heralded as humanity’s next great technological leap forward — one on a par with the agricultural or industrial revolutions. But what is it, how does it work… and what on Earth is a qubit?

Dilution refrigerator at the NQCC’s superconducting quantum computing lab, where ultra-cool environment of about -273 degrees C is maintained. Image: STFC, NQCC

The UK has opened its first quantum computing hub, the National Quantum Computing Centre (NQCC). But what is quantum computing, how is it different to ‘normal’ computing, and what might it be used for in the future?

There are currently only six operational quantum computers in Britain, but that doesn’t mean that quantum computing is a fringe industry. In fact, according some estimates, the market for quantum computing will reach £55 billion by 2035 and, for the financial, chemical, life science, and transport sectors alone, those computers will be worth some £1.5 trillion in economic impacts.

At the £140million NQCC, built at Harwell in Oxfordshire, UK, and operated by the Science and Technology Facilities Council (STFC), work is underway to construct 12 operational quantum computers. Eight will be built by private business and four by government scientists.

The hope is that quantum computing will represent the next big technological revolution and it will be used to tackle some of humanity’s most pressing existential problems.

Why do we need quantum computing?
To best understand just how radically different quantum computers are, it is probably worth recapping how a conventional computer works.

The processor in the computer you have at home (or in your phone, or smart watch etc.) handles information by breaking it down into ‘bits’ — which are either a one or a zero — basically a set of transistors that can be switched on and off (on being one and off being zero).
​
In the early days of computing, each ‘bit’ consisted of a vacuum tube, and a computer with 1,600 bits — such as the code-breaking Colossus machine (built in 1943) — might weigh a tonne and fill an entire really quite large room.

The Colossus mark 2 codebreaking computer built in 1943. Image: National Archives

Today, the silicon revolution means that vacuum tubes are no longer required and even an unremarkable mobile phone processor, smaller than your thumbnail, might have some 512 billion bits. Now that’s a lot of potential calculations.

The downside of conventional computers is that each calculation has to be carried out in sequence — each switch can only be a one or a zero at any one time and each must be changed in turn. This (by definition) binary way of doing thing means that one calculation must be completed before the next can begin.

Quantum computers get past this by using a little quantum physics ‘magic’ to enable them to carry out multiple calculations simultaneously. But how is this possible?

From bits to qubits
In place of the ‘bit’, the basic computational unit of a quantum computer is a ‘quantum bit’, or ‘qubit’. Instead of using anything so crude as a transistor, a qubit is made of some of the smallest building blocks of the Universe — individual atoms, electrons or photons.

There are actually many ways to build qubits — one of the machines built at the NQCC, for example, uses rubidium atoms held in place by laser beams — but all take advantage of one quantum mechanic’s strangest phenomena: quantum superposition.
​
Superposition is quality of quantum mechanics that means that a particle, or a photon, can exist in all of its possible states at the same time — it can be both spinning up and spinning down simultaneously and it will only settle which state to occupy when it is forced to ‘decide’ by being observed.

The Schrödinger’s Cat thought experiment was designed to illustrate a paradox of superposition in which a hypothetical cat, locked in a box with a vial of poison, may be considered both alive and dead simultaneously. It’s fate only determined when someone opens the box to observe it. Cartoon: Ben Gilliland

You can think of it as being like a coin toss. While a coin is spinning its way though the air, it could be said to exist as both heads and tails at the same time. It is only when the coin is ‘observed’ by landing on a tabletop, does it ‘decide’ which state to occupy.

The quantum ‘loophole’
It is the ‘Schrödinger’s cat’ ability of qubits to exist in all states simultaneously that enables quantum computers to break away from the sequential calculation limitations experienced by conventional bit-based computers. If a qubit can be both a one and zero at the same time, it can immediately have multiple outcomes beyond the limits of the binary bit.

Obviously a computer with a single bit, or even a single qubit, is not of much practical use and, as with conventional computers, the more you have the greater the number and greater the speed of calculations you can make.

For a conventional computer getting more bits is simply a matter or etching more bits into a silicon wafer, but if your computer is built of individual floating atoms, how do you get more and, more importantly, how do you get them to ‘talk’ to each other?

What an entangled web we weave
To get qubits to work together scientists must exploit another nugget of quantum weirdness: quantum entanglement. They can use this quirk to create networks of ‘entangled’ qubits with all of their superposition states connected together.

At its most simple, quantum entanglement means that two or more particles are connected though a sort of quantum wifi network.

This connection means that each of the qubits reacts to a change in another qubit’s state instantaneously no matter how far apart they are. This means that by just measuring the properties of one entangled qubit, you can instantly know the properties of its partners without having to look.

As a result, the more entangled qubits you can have in your quantum computer, the more (and faster) calculations that computer can make. And because each qubit exists in multiple states it can make multiple calculations simultaneously — add more entangled qubits and the number of possible calculations the computer can perform increases exponentially.

What does this exponentiality mean in practice? Well, for a quantum computer containing 1,180 qubits (still fewer than Colossus’ 1,600 bits), the number of entangled states on its processor would exceed the number of individual atoms in the known Universe.
​
This all means that, rather than having to run a calculation many times as you would with a boring old silicon processor, you can, in theory, run every conceivable calculation at the same time.

So what use is it?
Imagine you are a delivery company (perhaps named after one of the planet’s great rainforests) and you want to plot the best route from A to B in a city, but you also have to make 150 stops en-route to make deliveries.

A normal computer would have to look at each possible route individually and then compare the results to determine the most efficient outcome — and there may be tens of thousands of possible routes.

Because the quantum computer exists in many states at once, it can look at many different routes at once and determine the most efficient route almost instantaneously. Thus making complex calculations with many complex variables thousands of times faster.

This is a somewhat germane example, but the crucial takeaway is the ability to perform almost unlimited calculations at the same time will give quantum computers extraordinary real-time optimisation and pattern-finding abilities.

So it is expected that quantum computers would be a bank’s best friend and a fraudsters nightmare. They will be able to provide unimaginable insights into the sort complex natural systems that binary computers just can’t crack — such as how weather systems develop; how long-term climate change will unfold; and, on a smaller scale, how drugs interact with cancers… potentially saving your life and that of the planet upon which we depend for all our lives.

The flip side of this would be that a quantum computer could break almost any current data encryption system, which would give the planet’s spy agencies a bit of a headache and prompt an arms race of quantum-based cryptography verses the quantum hackers.

NQCC’s ion-trap quantum computing lab, where ions are trapped and employed as qubits. Image: STFC, NQCC

Leave me alone to think in peace
Unfortunately, scaling up your quantum computer is not as straight forward as just adding some more qubits and entangling them all together. In a normal computer, you can easily add more bits, but in quantum computers, the more qubits you have, the more unstable the system becomes and the harder it becomes to add more qubits to the system. This makes it incredibly difficult to scale up a quantum computer processor.

Also, superposition and entanglement are very unstable states and the slightest interference — such as vibrations or radiation — can cause these states to be disturbed or collapse. This is called quantum decoherence and when this happens the ‘broken’ qubit must be replaced.

This means that qubit-based processors need to be kept carefully isolated from the outside world by using vacuums, dampening vibrations and shielding from radiation.

It is this ‘scalability’ problem that centres like the National Quantum Computing Centre have been built to tackle but it is also the reason why you are unlikely to ever have a quantum computer on your desk or in your pocket — meaning that the future of computing is unlikely to be solely quantum and will probably be a combination of qubits and good old fashioned binary bits.

Story: Ben Gilliland