Microsoft and the mystery of the disappearing physicist
- 08 September, 2017 13:06
Nobody knows what happened to Ettore Majorana on March 25, 1938. Or after.
Having first hypothesised the existence of the strange particles called Majorana fermions the year before, the young Italian boarded a boat from Palermo to Naples and was never seen again.
Fittingly perhaps for a physics pioneer whose work would become fundamental to one of the most convincing approaches to building a quantum computer – it was as if he was there and then not there. A 1 and a 0, at once.
Some believe he committed suicide. He suffered long periods of depression, once locking himself away and refusing to publish any work, considering it too banal, for four years.
Does the letter of resignation from his teaching post just before his trip suggest that fate? Then why had he withdrawn all his money a few days before?
Some believe he foresaw the creation of nuclear weapons, which sparked a “mystical crisis” – a man matching his description was reported soon after at a monastery in southern Italy.
What of the Naples beggar who helped local students with their maths homework? Or the sightings in Argentina and Venezuela? Could he have been kidnapped by the Nazis and forced to develop an atomic bomb?
The peculiar quasiparticles he put his name to have also remained an enigma. But in recent years evidence has been mounting for their existence. This week a paper by researchers in Sydney and the Netherlands provides the most convincing supporting data to date.
It’s so convincing, Microsoft is betting its quantum computing future, and millions of dollars in research funding, on finding Majorana’s mysterious particles.
“This was really the missing link,” Dr Maja Cassidy of Microsoft’s Station Q lab in Sydney and co-author of the paper told Computerworld. “It was the one thing we could never rebut against peoples’ arguments against Majorana fermions. And now we have the evidence.”
Mind the gap
Building on the 1928 work of quantum pioneer Paul Dirac, in 1937 Majorana predicted a new class of particle. A particle that is its own antiparticle: the Majorana fermion. (The theory is complex but well explained here)
But Majorana’s theory remained only that for decades.
“For about 80 years nothing happened because no-one knew how to create the material that they would exist in,” Cassidy explains.
Then in 2012, at Microsoft’s Station Q quantum computing lab at Delft University of Technology in the Netherlands, Mourik et al observed tell-tale signs of Majorana fermions in hybrid superconductor-semiconductor nanowire devices.
“But as with observing anything, there can be other explanations. There were many who said so,” explains Cassidy, a post-doctorate at Delft at the time who joined Station Q in Sydney last year.
“One of the bits of evidence that was really required for confirming the existence of Majoranas was this ‘helical gap’. The helical gap was the missing piece of the puzzle,” she says.
In the paper published in Nature this week – Conductance through a helical state in an Indium antimonide nanowire – Cassidy and her colleagues provide evidence they have found that missing piece.
In essence, it proves that electrons on a one-dimensional semiconducting nanowire will have a quantum spin opposite to its momentum in a finite magnetic field.
“Now that we’ve seen this it really strengthens the evidence for these Majoranas,” she says. “It’s a big tick on the checklist. That's very satisfying.”
The existence of these Majorana fermions in the natural world is tantalising for those, like Microsoft, pursuing the topological approach to building a quantum computer. Their potential robustness against decoherence, which dogs other approaches, could prove hugely advantageous when creating arrays of qubits.
Topological qubits, Microsoft says, are better able to withstand heat and electrical noise, which allows them to remain in a stable quantum state for longer and offers a more viable way to make a scalable, usable quantum computer.
Microsoft’s topological approach is at odds with its competitors – such as the silicon-based route taken by the UNSW-based CQC2T or the superconducting loop approach of Google and IBM. With the existence of Majorana fermions still unproven, the company’s significant ramp up in research (opening one of eight new labs at the University of Sydney in July) is something of a gamble.
“Everything in the world involves taking risks. You’ve got to compare the other options available – are they really suitable for building a useful quantum computer?” Cassidy says.
“A lot of people might be able to build a toy model of a quantum computer, with tens of qubits for example. But getting to a system you can actually use for solving real problems and which will be commercially viable, that comes in the hundreds of thousands of qubits. And scaling those systems is a lot harder than scaling a topological computer.”
Microsoft considers what's been dubbed the ‘new space race’ to build a quantum computer more of a marathon.
“It’s the tortoise and the hare fable,” Cassidy says. “The hare may be speeding up right now but the tortoise approach may be more solid in the long run.”