Chandeliers, ‘qubits’ and Schrödinger’s cat: Inside the bizarre world of quantum computing

It was 2009, and a group of scientists had crowded around a tiny monitor in a Sydney physics laboratory. The team, led by electrical engineer Andrea Morello, held their collective breath and peered at a single flat line on the screen. Then, in the blink of an eye, the line changed: it was a series of blips, like a heartbeat. They’d hit paydirt. “When you see that symbol, you’re in the game,” recalls physicist Charles Tahan.

Welcome to the quantum world, a subatomic realm – yes, smaller than an atom – where the laws of physics are, well, different. The experiment made Morello and his team the first in the world to show that a quantum computer could be made using not diamonds, not the superconductor niobium, but everyday silicon. “To show you could do that, 100 different things had to go right,” says Tahan.

Quantum computing is one of the key technologies defining this century. Its proponents say it will do for computers what the car did to the horse and buggy. It will upend encryption and molecular biology. In theory, it could even possibly find cures for cancer. Australia is among the world leaders in a quantum computing gold rush, with Prime Minister Anthony Albanese saying his government is “serious about building a strong quantum ecosystem” and investing nearly a billion dollars.

Yet despite eye-watering outlays worldwide, quantum computing is still in its infancy, part of a mind-bending world that physicists are only beginning to fully grasp. What is it? And how will it make a difference to our lives?

How does a quantum computer work (and why did Alice in Wonderland have the jump on it?)

It’s often remarked that it looks like a beautiful chandelier, or a wedding cake. And IBM’s quantum computer, one of the world’s largest, is certainly an intriguing object, an intricate dome of golden wires and graceful tubes suspended from a ceiling.

The part we can see, though, is really just the control and support system for a microchip (currently a version called Condor) buried deep inside. The glittering exoskeleton is there to keep its tiny heart as cold as possible – a hundred times colder than the coldest depths of space – and to feed information in and out, the process that makes it a “computer” and not just a science experiment.

The chip itself is smaller than a postage stamp, with several layers of circuitry and metal sandwiched in a tiny square, cooled to improve conductivity and to eliminate “noise”, the bugbear of many such devices. But even what we see here is not where the action happens. For that, we have to go deeper still, down into the chip and well beyond what’s visible through even the most powerful microscope, beyond the layers of individual atoms, into the atoms themselves.

‘The truth is the universe doesn’t run by the rules that we see every day – gravity, you bounce a ball and it falls. It runs on these other rules: quantum mechanics.’

In the subatomic realm of electrons, protons and photons, things happen very differently. Indeed, these tiny particles have no use for Newton’s laws of physics, nor even many of Einstein’s. “The truth,” says Tahan, “is the universe doesn’t run by the rules that we see every day – gravity, you bounce a ball and it falls. It runs on these other rules: quantum mechanics.”

Why these particles behave as they do in the quantum realm is a difficult concept to get your head around. Even Einstein resisted aspects of it, famously arguing that “God does not play dice with the Universe”. It was a misstep. “Because he didn’t believe in the completeness of quantum mechanics,” says Morello, “he lost a lot of productive years because he didn’t understand how it is possible that this thing is true.”

Many leading physicists investigated the concept from the early 1900s, including some familiar from popular culture today, notably Einstein and Neils Bohr (portrayed in Oppenheimer), Werner Heisenberg (whose name Walter White stole for his drug-chemist alter ego in Breaking Bad) and Viennese physicist Erwin Schrödinger, of the famous cat experiment (explained below).

Lewis Carroll, meanwhile, may have had the jump on quantum physics with his ideas in the Alice books, as James Renner observed in The Wall Street Journal in 2015. “We speak of going down the wormhole into a mysterious world that defies Newtonian physics, where once Alice went down the rabbit hole into a mysterious world that had much the same effect. Now we speak of quantum particles being in two places at once or even Schrödinger’s cat being alive and dead at the same time, long after the Cheshire Cat’s grin was magically displaced from its body.”

The quantum realm has only truly begun to reveal itself to us relatively recently. In 1964, the Irish-born physicist John Bell published one of the most important theorems, which, according to Professor Andrew Whitaker in Physics World, “opened up the possibility of experimental philosophy, the study of what are normally thought of as philosophical issues in experiments. Not only do these experiments probe the deepest and most profound aspects of quantum theory, they also provide information on the fundamental nature of the universe.”

At the time, Bell’s theorem could not be tested in a lab. That would come in the late 1960s and early 1970s when experimentalists such as John Clauser began to show that the weirdness Bell had described on paper actually occurred in the real world. Which, at the time, was considered quite rude. One eminent physicist “was highly offended by my impertinent effort and told me that it was tantamount to professing a disbelief in quantum physics,” Clauser later recalled.

He and other apostates continued their work because he found some of the predictions “sufficiently bizarre that I could not accept them without seeing experimental proof”. Working with leftover lab equipment – “physics department scrap” – he went on to provide experimental evidence that refuted even Einstein’s theories. “I was very sad to see that my own experiment had proven Einstein wrong,” he said. Clauser would go on to win the 2022 Nobel Prize for Physics alongside Alain Aspect and Anton Zeilinger for their groundbreaking experiments with so-called “entangled” particles.

Quantum physicists still despair at the difficulties of explaining their weird world to outsiders. “Many writers fall back on boilerplate that makes physicists howl in agony,” wrote computer scientist Scott Aaronson in The New York Times in 2019. Unfortunately, says Gerard Milburn, a quantum physicist from the University of Queensland, “The quantum world lies beyond the reach of natural language.”

But we’ll give it a go. Chiefly, particles in the quantum realm aren’t tied down to being in one place or state in the way bigger things are in the outside world, such as cats, which tend to be in one place at once. They obey different rules entirely. The first way to understand this oddness harks back to high-school science and how light can behave as both a particle and a wave. Photons – bursts of light – can act like ball bearings, with position and momentum. Yet light also behaves like a wave washing over a beach in a way that a ball bearing can’t. That’s a quantum physics party trick.

What is Schrödinger’s cat?

The most famous primer for quantum mechanics is the paradox of Schrödinger’s cat, a mind experiment devised by Schrödinger in 1935. A hypothetical cat goes into a box with a vial of poison gas, which has a 50-50 chance of killing it. You close the box and can’t see inside. So, without opening the box, you can’t know whether the poison has killed the cat or not. Therefore, posited Schrödinger, puss is neither dead nor alive but both – until you open the box.

In quantum mechanics, this state is called “superposition”. It’s impossible to describe properly without a background in higher-level mathematics, which we lack, but it generally means not knowing whether, say, an electron is over here, or over there, or – cat-like – in both places at the same time. Another way of looking at it is to visualise a coin spinning on its side. Until it stops and lands on one side or another, it’s not heads or tails but both.

At least, kind of.

“Every explanation we give in lay terms is in some way wrong,” says Stephen Bartlett, a theoretical quantum physicist and director of the University of Sydney Nano Institute. “People often say a quantum superposition is in the value zero and the value one at the same time. That’s sort of right, and sort of not right.” Says Morello: “This is where language is failing us, right? If you’re trying to say that in plain English, it sounds really weird. But if you actually think about, you know, the mathematics of it, and the actual physical embodiment of that, it actually makes perfect sense.” Suffice to say, a real cat cannot find itself in “superposition”: this is a state that can exist only in the subatomic world.

How is this useful (and what’s a qubit)?

In a conventional computer, information is encoded as voltages that pass around the system thanks to millions of tiny transistors – essentially, gates or switches that turn on and off. A quantum computer is similar except that instead of using transistors as switches it uses subatomic particles, such as electrons or photons, that can have many more states than simply on or off, according to their position, momentum, spin direction and polarisation. Once harnessed – and there are many ways of doing that – these particles-as-switches are called “qubits”, a play on the word “bit”, the smallest unit of data that a computer can process and store.

Information can be encoded on qubits in various ways: for his computers, Morello uses silicon and strong magnetic fields like those found in MRI machines. “The stuff that I build in my lab is actually a small silicon chip the size of a fingernail,” he says, “but the whole infrastructure around it is quite fancy. It’s a very powerful refrigerator that cools down the chip to [just] above absolute zero.

Critically, qubits can relate to other qubits in a state called “entanglement”. This enables quantum computers to perform multiple calculations simultaneously – which is their superpower. Schrödinger described entanglement as “the characteristic trait of quantum mechanics, the one that enforces its entire departure from classical lines of thought”. Sceptical Einstein called entanglement “spooky action at a distance”. Gerard Milburn says entanglement is great for building a computer “because if you want to have all these switches, having them strongly correlated is exactly what you want. And if you can control those correlations, you can do computation.”

It’s these characteristics that have scientists and engineers excited. Not because quantum computers will necessarily take over from the conventional miracles we so rely on today but because they can do some particular things that transistor-based machines either struggle with or fail at completely. Today’s supercomputer might try to simulate molecular behaviour with “brute force”, says IBM’s quantum department. “But as it moves past the simplest, most straightforward molecules available, the supercomputer stalls. Quantum algorithms take a new approach to these sorts of complex problems, creating multidimensional computational spaces.”

‘Calculations that are completely impossible … would rapidly become possible on a quantum computer.’

This might include modelling how a drug might affect a virus, how new materials in aeroplanes might react to stress, or the exact process necessary to create a perfect chemical reaction.

“Calculations that are completely impossible – not just on present-day conventional computers but any conceivable future conventional computer – would rapidly become possible on a quantum computer,” says Andrew Doherty, professor of physics at Sydney University. “These are things where the application of that computer would be transformative to people’s lives without every home, you know, having a laptop quantum computer.”

Whether quantum computing will also revolutionise artificial intelligence is less certain. AI works by drawing from a huge amount of data, which is not something that naturally fits on a quantum computer, says Morello. “The problem with quantum computers in general and, in particular, with applications to AI is that loading data into them and reading data out of them is slow and cumbersome and prone to errors. It’s a bottleneck.”

Why is it so hard to build quantum machines?

Creating the qubits is hard, whatever material you start with. Interfacing with them via tiny wires like the golden lattice in IBM’s machines is hard. Getting accurate information from them is hard: “noise”, in the form of vibration or spurious electromagnetic signals, makes them behave badly and give poor results – “decoherence”, in physics speak.

Much of Andrew Doherty’s team’s work these days is in figuring out how to reduce this noise to acceptable levels and compensating for the errors that noise produces. “Compared to our conventional computers, quantum technology [computers] are very noisy devices,” he says. “They go wrong one time in 1000 as opposed to one time in an astronomically huge number of uses of [normal computers].”

Teams around the world are working on multiple approaches both to building the machines and compensating for their eccentricities, from labs in Australia to the tech giants Google, Microsoft and IBM – its Quantum Lab relies on superconducting qubits made from aluminium and niobium – and the Chinese government.

Some versions, such as IBM’s and Morello’s, chill the atoms to keep them as stable as possible. In contrast, the Australian company Quantum Brilliance is working with synthetic diamonds because they work at room temperature. “Unlike other materials,” says co-founder Marcus Doherty (no relation to Andrew), “diamond’s incredible hardness means that even at room temperature, there isn’t enough thermal energy in the environment to create vibrations that destroy the quantum properties of qubits.”

‘People are working really hard to try to find different mathematical problems that they think maybe quantum computers can’t crack.’

Another key moment in the quantum timeline came in 1994 when US mathematician Peter Shor showed that a quantum computer could crack encryption standards safeguarding everything from national intelligence systems to bank accounts. “He devised an algorithm, in theory, that he would need to run on a quantum computer,” says Bartlett. “And if you could build a quantum computer of sufficient size, then yeah, all information security falls apart.” Cybersecurity experts call that scenario Q-Day. “At the moment, people are working really hard to try to find different mathematical problems that they think maybe quantum computers can’t crack,” says Bartlett.

In 2001, IBM made a seven-qubit device controlled through radio frequency pulses to demonstrate Shor’s Theorem. IBM’s prototype wasn’t given a huge task, managing to calculate that three times five was … wait for it ... 15. Yet this was considered a great success, even if it had shades of the famous supercomputer Deep Thought from Douglas Adams’ Hitchhiker’s Guide to the Galaxy, which ponders the question of “life, the universe and everything” for thousands of years before giving the answer: 42.(Adams also dreamed up the Infinite Improbability Drive for a spaceship called the Heart of Gold, which used quantum theory to allow it to cross through every point in the universe simultaneously, allowing interstellar travel without all that “tedious mucking about in hyperspace”.)

Then, in 2019, Google claimed its 54-qubit Sycamore processor had performed a calculation in 200 seconds that would have taken the world’s most powerful supercomputer 10,000 years – a breakthrough some breathlessly likened to the Wright brothers’ first flight in 1903. Others were more sceptical, saying it would be many more years before quantum computers would have enough qubits – and be stable enough – to do anything actually useful. IBM said it could simulate Google’s experiment in two-and-a-half days on a “normal” computer, albeit if it had access to the Oak Ridge Summit supercomputer, a beast occupying an area the size of two basketball courts. IBM’s shimmering Eagle, meanwhile, had just 127 qubits; its successors, Osprey and now Condor, have pushed the bar to over 1000 and it plans for 1386 qubits in its Kookaburra in 2025.

Yet most of these devices are still basically proofs of concept, says Morello, the first step to manufacturing and miniaturising, as we saw with today’s transistor-based computers, which started out as rooms full of valves and cables. In the post-World War II era, says Andrew Doherty, “computers took up office floors of buildings in Manhattan and people put paper cards into them to program them. Out came numbers that even my childhood pocket calculator could have calculated right.” Many hundreds of billions of dollars have gone into developing computer technology, in packaging it for convenience and getting the components to work reliably, he says. “That whole engineering thing has got to grow up around these quantum technologies.”

Already, Marcus Doherty’s Quantum Brilliance has installed a quantum computer at the Pawsey Supercomputing Centre in Perth and is about to export computers, at this stage for customers to experiment with quantum computing and work out how to make it useful. In the future, finding commercial applications for a quantum computer will require tailoring it, at least initially, says UNSW’s Maja Cassidy, a former quantum researcher for Microsoft. “People have to go in and understand a company’s problem, how they’re using classical computing and … work out what section might be enhanced by the quantum computer.”

Who is leading the quantum race?

China is the only other nation giving the United States a run for its money. Its leader, Xi Jinping, put accelerating quantum technology in his 2021 five-year plan. McKinsey & Company estimates China has spent at least $15.3 billion in public funds on quantum research. That partly came to fruition in April with the Chinese Academy of Science’s unveiling of a 504-qubit chip, still half the size of IBM’s biggest.

For its part, Australia got a head start with quantum research in the 1990s. Today, two major spinoffs from UNSW research are pursuing silicon-based quantum computing: Silicon Quantum Computing (SQC), led by Professor Michelle Simmons, and Diraq, led by Professor Andrew Dzurak. SQC’s approach involves embedding phosphorus atoms within a wafer of silicon. Dzurak’s approach involves trapping electrons on silicon transistors. Both have the advantage of silicon chips already being ubiquitous. “Today, in our phones and our laptops, the silicon chips that are in those have upwards of a billion transistors on them,” says Dzurak. “The benefit of the technology that we’ve developed is that we are also able to get millions, and ultimately billions, of quantum bits onto a single silicon chip.”

‘Computers as we know them today are an absolutely mind-blowing achievement. So any technology that has the ambition of being an alternative or being better than that has a phenomenally high bar to clear.’

Other leading players in Australia include Q-CTRL, which is working on controlling and enhancing quantum hardware; quantum security company QuintessenceLabs; and Doherty’s Quantum Brilliance, which used its synthetic diamonds in the world-first integration of a two-qubit quantum device into a supercomputer at room temperature in 2022.

Meanwhile, out of the University of Queensland came early gains in “photonic” quantum computing, which harnesses particles of light – photons – as qubits. In April, the federal and Queensland governments jointly set aside $940 million in share purchases, grants and loans for PsiQuantum, a Silicon Valley company whose founders include Australian professors Jeremy O’Brien and Terry Rudolph. Unlike ions or electrons, photons are not charged, Bartlett says, so are not sent haywire by electromagnetic forces – but they are hard to manipulate and measure.

Another approach involves “trapped ion” quantum computing, pursued by US company IonQ: physicists take an atom, flick off an electron so it is positively charged, then suspend the resulting ion within an electromagnetic field and manipulate the ion with lasers. But Bartlett says this approach will be hard to scale, given the technology to suspend millions of ion qubits doesn’t yet exist.

In April, the federal government also announced $18 million for the University of Sydney to establish Quantum Australia, a research consortium between several universities headed by Bartlett. “It used to be a small research community,” he says. “You absolutely needed to speak the common language of everyone else, which were the rules of physics. Now we are really seeing quite a bit of specialisation.”

Even if quantum computers develop practical uses, they will likely live in vast, super-cooled server farms where land and power are cheap. We’ll communicate with them through our everyday devices, much as we download and upload data from cloud servers, Netflix, TikTok, DropBox, whatever, today. The real challenge for quantum computing, says Morello, is that the alternative is “really, really good”. “Computers as we know them today are an absolutely mind-blowing achievement. So any technology that has the ambition of being an alternative or being better than that has a phenomenally high bar to clear.”

Get fascinating insights and explanations on the world’s most perplexing topics. Sign up for our weekly Explainer newsletter.