Using powerful computational tools, Ping identifies a promising oxide qubit material

In a paper published June 3, 2026, in the journal PRX Quantum, MS&E Associate Professor Yuan Ping, along with UW-Madison graduate students Shimin Zhang, Erik Perez and Kejun Li, used a suite of sophisticated predictive tools developed by their lab to identify an ideal candidate material for deep-level defect qubits. This material could help expand the material foundation for solid-state quantum information science.

Qubits, or quantum bits, are the fundamental units of information in quantum computing and sensing. These particles must be stored and encoded using advanced physical systems that protect them from outside interference. One strategy is to embed quantum particles in defects or vacancy centers in materials, or spots where an atom has been knocked out. Vacancy centers in a small handful of materials including diamond, silicon carbide and hexagonal boron nitride are all being investigated as qubit hosts.

Another recently-identified, promising host material is zinc oxide, which has several advantages. It can be fabricated with exceptional purity, is magnetically “quiet,” and can be strained to tune its quantum state.

Zinc oxide already hosts shallow-donor qubits, but those rely on weakly bound electrons that work only at cryogenic temperatures and emit in the ultraviolet spectrum, which is not ideal for fiber-based quantum networks. What zinc oxide has lacked is a deep-level defect: a tightly localized, spin-triplet system that could operate at higher temperatures and emit usable visible or near-infrared light.

To produce vacancies and serve as a qubit host, however, zinc oxide needs to be combined with another element to create a complex. Synthesizing and analyzing all the possible variations of these oxide complexes experimentally would be costly and tedious. That’s why Ping and her team approached the problem computationally, applying a suite of first principles tools they’ve developed over the years to the problem, which particularly address the strong electron correlation difficulty in zinc oxide.

They assessed a wide array of candidate materials, looking for a zinc oxide complex that protects and supports long-lasting quantum states. They also assessed the complexes for the ability to support high-fidelity, single-shot readout, an advanced property which allows quantum state information to be read in a single run, instead of being averaging over multiple experimental runs, which is the current standard.

“You have to go through a list of properties, first doing a very broad search for candidates with the right spin state,” says Ping. “We then do thermodynamic stability, optical properties, spectroscopic properties, and spin coherence. Then we have to measure the kinetic processes related to single shot readout.”

The final result of the predictive model indicates that a zinc oxide complex including molybdenum-vacancies is the ideal candidate to produce the desired deep-level defect qubits in zinc oxide. The complex is a spin triplet in the ground state, has visible-range optical transitions with high quantum yield, and exhibits an unusually small Huang–Rhys factor of about 5, compared with 10–30 for other known zinc oxide defects.

In plainer terms, the defect’s atomic structure barely rearranges when it absorbs or emits a photon, yielding a sharp zero-phonon line and high-quality optical addressing. Crucially, strong spin–orbit coupling combined with the absence of Jahn–Teller distortion produces spin-selective intersystem crossings that should support high-fidelity single-shot readout across a wide range of magnetic fields and at elevated temperatures — not only in cryostats.

Experimental researchers have synthesized the material and will soon begin to measure its properties to validate the prediction. While the molybdenum-vacancy complex might lead to new qubits and advances in quantum technologies, Ping says the computational process demonstrated in the paper is just as important, if not more so, than the result.

“The ideal situation is to produce a predictive theory where you give inputs first, then let experimentalists validate them,” says Ping, who points out that currently the opposite it usually the case; theory and prediction are often used to explain and refine experimental results. “The goal is really to have a model you can use to guide the experiment, instead of just interpreting it.”

The work also flags an underappreciated bottleneck for any oxide-qubit roadmap: paramagnetic impurities — not nuclear spins — are likely to set the coherence ceiling once isotopically purified zinc oxide becomes available. Putting a quantitative threshold on that limit gives material growers a concrete target.

If the screening framework generalizes, the same logic could be applied to other oxide and nitride hosts, accelerating the search for qubits that are easier to integrate, more compatible with existing semiconductor processes, or simply better behaved than the systems studied today.

Other authors include Taejoon Park and Hosung Seo of Sungkyunkwan University, Suwon, Korea; Xingyi Wang and Kai-Mei C. Fu of the University of Washington, Seattle, Washington; Masoud Mansouri of the Universidad Autónoma de Madrid, Spain; and Yanyong Wang, Jorge D Vega Bazantes, Ruiqi Zhang, and Jianwei Sun of Tulane University, New Orleans, Louisiana.

The project is supported by the U.S. Department of Defense through the Air Force Office of Scientific Research (AFOSR) under its Center of Excellence on Fundamental Information Research (CFIRE) program under grant FA9550-23-1-0418— a DOD investment in the basic science underpinning future quantum information technologies.

It used resources of the Scientific Data and Computing center, a component of the Computational Science Initiative, at Brookhaven National Laboratory under Contract No. DE-SC0012704, the National Energy Research Scientific Computing Center (NERSC) a U.S. Department of Energy Office of Science User Facility operated under Contract No. DE-AC02-05CH11231. This work used the TACC Stampede3 system at the University of Texas at Austin through allocation PHY240212 from the Advanced Cyberinfrastructure Coordination Ecosystem: Services and Support (ACCESS) program.

Ping also acknowledges the support by the National Science Foundation under grant no. DMR-2143233 for the support of computational technique development in this work.