Majorana 1: A quantum phenomenon

A view of Microsoft’s Majorana 1 chip.
| Photo Credit: Microsoft

In a head-turning announcement earlier this week, Microsoft unveiled a new quantum chip called Majorana 1. According to its press release, it consists of four qubits made of a “new state of matter”.

Qubit is short for ‘quantum bit’, the fundamental unit of operations in a quantum computer, similar to classical bits in conventional computers. A single qubit represents more information than a single bit. A quantum computer heightens the advantages of this ability using quantum phenomena like superposition and entanglement to solve complex mathematical problems in fewer steps than perhaps the most powerful conventional computer.

Quantum computers differ depending on their choice of qubit. For example, Google’s Willow chip uses small circuits that mimic the properties of atoms in a quantum state. Microsoft has said its new chip uses Majorana particles, elusive subatomic entities supposed to appear in materials called topological superconductors. These particles have unusual properties that physicists believe could be used to build qubits that are less error-prone than the designs Google and IBM use.

“Topological qubits can win if, and only if, they turn out to be so much more reliable that they leapfrog the earlier approaches…,” University of Texas at Austin professor Scott Aaronson wrote on his blog. “Whether that will happen is still an open question, to put it extremely mildly.”

A physics primer

If you cool some water vapour, it will become water and then ice. If you keep lowering the temperature until nearly absolute zero, the system will have minimal thermal energy, allowing quantum states of matter to show. In the 1970s, Michael Kosterlitz and David Thouless found that the surface of superfluid helium sometimes developed microscopic vortices that moved in pairs. When they raised the temperature, the vortices decoupled and moved freely. It was a new kind of quantum phase transition: the object’s topological attributes changed in response to changes in energy.

Topology is the study of the properties an object retains when it is deformed continuously, e.g. when it’s stretched but not ruptured. Topological materials are materials with topologically protected states: the ‘protection’ keeps the state stable against impurities or small disturbances. For example, a vortex in a bucket of water will fade as it loses kinetic energy. But vortices in superfluid helium are topologically protected. With the right internal conditions, they will keep spinning. Only quantum physics allows this.

Similarly, topological superconductors are materials whose ability to conduct an electric current without resistance is topologically protected. The signature of a topological superconductor is its ability to host Majorana particles. These materials are not found in nature. Instead, scientists engineer them in the lab to have the combination of internal properties that allows Majorana particles to exist. Microsoft has reportedly used indium arsenide with aluminium.

A fundamental tension

But since the announcement, independent experts have asked whether the new chip really has Majorana particles. Many previous attempts, including some sponsored by Microsoft, that claimed to have found these particles were later overturned by further scrutiny.

Microsoft’s press release also claimed that a paper simultaneously published in Nature provided “peer-reviewed confirmation” that its team had created Majorana particles for use as qubits — but the paper, authored by the same team, stated that the work therein didn’t test whether the particles were really Majorana. Microsoft researcher Chetan Nayak later clarified on Mr. Aaronson’s blog that the team submitted the paper to Nature in March 2024 and that team members have “continued to make progress in the intervening year”.

The contradiction still subtracted from confidence in Microsoft’s claims because it had leaped from a paper describing older work to a press release making new claims. This is a familiar problem: when the private sector does cutting-edge research, its need to protect IP may prevent it from sharing all the information necessary to confirm the validity of a claim, at the expense of doing right by the research community whose approval it seeks.

For now, scientific opinion about the alleged prowess of Majorana 1 is uneven but also mostly sceptical. Microsoft thus has a lot to gain, or lose, by showing the chip at work, but its actions may have left it with little choice.