On June 2, 2026, Microsoft researchers reported a characteristic parity lifetime of 22 ± 1 seconds in nanowire devices that use lead in place of aluminum. Microsoft presented the devices, described in a preprint posted that day, as its Majorana 2 chip.
Parity here refers to whether a superconducting nanowire contains an even or an odd number of electrons, the property in which Microsoft’s proposed qubits would store information. The parity lifetime is the typical time before the parity flips on its own.
Microsoft’s researchers presented only Z measurements in the preprint, Science News reported, although demonstrating a qubit requires both X and Z measurements. The two kinds of measurement read a qubit’s state in complementary ways.
Chetan Nayak, Microsoft’s technical fellow and corporate vice president of quantum hardware, wrote in a blog post the same day that Majorana 2 contains qubits “1,000x more reliable” than those in Majorana 1, the company’s previous processor. He put the qubit lifetimes in Majorana 1 at between 1 and 12 milliseconds. Nayak wrote that Microsoft had cut its timeline in half and now aims to reach a scalable, practical quantum computer by 2029.
What the researchers measured
The preprint lists 165 authors, with Morteza Aghaee first. The researchers wrote that they had improved the topological phase of their devices by replacing aluminum with “the higher-gap superconductor lead.”
In the topological phase, quantum states called Majorana zero modes are predicted to form at the ends of each nanowire and to resist local noise. An energy margin called the topological gap protects those states from disturbance. Nayak wrote that the topological gap of Majorana 2 is more than double that of Majorana 1.
The researchers extracted the 22-second figure, with its fit uncertainty, from 324 dwell intervals, the stretches between parity flips, recorded across repeated measurements. In the abstract, they gave the characteristic parity switching time as about 20 seconds and wrote that some instances reached “minute-scale” lengths. Nayak wrote of lifetimes “with a mean of 20 seconds, occasionally exceeding one minute,” in Majorana 2.
The researchers wrote that the flips followed the pattern they expected from “high-energy quasiparticle poisoning.” Quasiparticles are stray excitations in a superconductor that can enter a nanowire and change its parity.
The device has four tetrons, which Nayak described as a type of topological qubit made of two superconducting nanowires with Majorana zero modes at their ends. The researchers called the device “a prototype unit cell for multi-tetron devices” and wrote that the unit cell can be tiled into larger arrays.
They also reported a radio-frequency technique that resolves low-energy states at the ends of a wire and directly measures their energy splitting. The researchers listed tunneling between Majorana zero modes, the process that produces such a splitting, among the dominant sources of qubit error.
X measurements and peer review
Science News wrote that Microsoft had reported X and Z measurements for its previous chip but that many experts were not convinced by those X results. In February 2025, Nature published a Microsoft paper on the devices behind Majorana 1. The journal’s editors noted in the paper’s peer review file that the results did not represent evidence of Majorana zero modes, Physics World reported at the time.
In the June preprint, the authors wrote that the energy splittings they achieved “are expected to have beneficial implications for Pauli-X measurements.” They suggested that the switching time for X measurements could be “more than an order of magnitude longer” than in previous devices. Investigating these effects will be “an important direction for future work,” they wrote.
Reactions
“Nothing in this preprint resolves the fundamental issues,” Henry Legg, a physicist at the University of St Andrews in Scotland, told Science News. “Nothing in the presented data proves the existence of a topological qubit or Majoranas in these devices,” he said. Science News noted that the preprint had yet to be peer reviewed.
Kartiek Agarwal, a physicist at Argonne National Laboratory, told the magazine that results from the new measurement method support the case that the states are Majoranas. “This is fantastic progress,” Agarwal said.
Microsoft’s schedule
Nayak wrote that DARPA had advanced Microsoft as one of two companies to the final phase of its program to evaluate quantum systems. DARPA said in February 2025 that it had selected Microsoft and PsiQuantum for the final phase of its Underexplored Systems for Utility-Scale Quantum Computing program, one of two programs in its Quantum Benchmarking Initiative.
Nayak wrote that Microsoft intends to build a fault-tolerant prototype based on topological qubits within years. A fault-tolerant machine detects and corrects its own errors as it runs.
Update, November 13, 2026: On June 24, Nature published Legg’s challenge to the data behind Microsoft’s February 2025 paper, with a reply from Microsoft researchers. On September 22, Microsoft said it was providing topological qubits built on its Majorana 2 chip for independent testing in Maryland and that “DARPA will have full access” to the system on site. As of November 13, the preprint had no journal version.
