Millisecond Qubits Mark a Breakthrough in Quantum Tech

Breakthrough in Millisecond-Scale Superconducting Qubits

Quantum computers could revolutionize how we perform cryptography, calculate complex simulations like proteins’ 3D configuration, and probably have many more applications we are barely guessing today.

To function, they need as stable as possible “qubits”, the fundamental element of quantum computation. So far, only “trapped ion” quantum computers have managed to generate highly stable qubits. But this technology is inherently harder to scale up than superconducting qubits.

So while superconducting qubits might be the future of this technology, an improvement in the stability of their qubits’ coherence time is required.

This is just what a large team of researchers at Princeton University has just achieved. They created a type of superconducting qubits that can keep coherence for more than a millisecond, 3x longer than the best ever recorded.

They published their findings in Nature¹, under the title “Millisecond lifetimes and coherence times in 2D transmon qubits”.

The Qubit Coherence Limit

To perform quantum computation, a quantum computer needs to maintain “coherence”, a special quantum state that is extremely vulnerable to interference from the environment. In general, thermal noise and particle movement tend to destroy coherence in nanoseconds.

In special conditions, like ultra-cold conditions, a qubit’s lifetime can last longer. But still, long enough coherence is a major limitation to most quantum computers today, leading to calculation errors that not only diminish the total computing capacity, but also cannot easily be compensated with software upgrades.

So determining what material is able to maintain coherence for longer is a key step forward that needs to be made before reaching the commercial stage for the quantum computing industry.

“The real challenge, the thing that stops us from having useful quantum computers today, is that you build a qubit and the information just doesn’t last very long.

This is the next big jump forward.”

Andrew Houck, Princeton’s dean of engineering

How Researchers Extended Transmon Qubit Coherence

The researchers used the same type of superconducting qubits used by firms like Google or IBM in their own quantum computer, transmon qubits.

Transmon qubits have the advantage of being high-fidelity (single-qubit gate fidelities exceeding 99.9%), possible to produce at scale, and with high coherence times of 0.1 milliseconds.

This is promising, but the coherence time is still too low.

So when the Princeton researchers announced having managed to create a qubit lasting 1.68 ms on average, this is a massive improvement.

Source: Nature

This is a qubit duration 3x longer than the best ever created in a lab, and 15x stronger than the one used by private companies developing quantum computers.

Why Tantalum and Silicon Improve Quantum Coherence

Tantalum Boosting Coherence

To achieve this result, the researchers used two different improvements in the material used.

First, they used a metal called tantalum as a base layer to help the fragile circuits preserve energy. This is because tiny, hidden surface defects in the metal can trap and absorb energy as it moves.

It is especially problematic as more qubits are added to a chip, this type of error multiplies to the point of making it useless past a certain number.

Scanning transmission electron microscopy (STEM) was used to confirm the highly regular structure of cubic crystals of tantalum.

Source: Nature

Compared to metals like aluminum, tantalum has a lot fewer defects, and is highly resistant to harsh cleaning processes used to remove impurities.

“You can put tantalum in acid, and still the properties don’t change.”

Faranak Bahrami – Research at Princeton University

Growing tantalum directly on silicon was a challenge that took extensive effort to overcome.
Swipe to scroll →

Silicon Replacing Sapphire

Another source of energy loss leading to losing coherence is the sapphire substrate used in quantum chips.

Instead, the researchers used high-quality (high-resistivity) silicon, a common standard material of the traditional computing industry.

Together, these improvements in the materials used in this tantalum-on-silicon platform made the resulting single-qubit gates achieve 99.994% fidelity.

From Lab Breakthrough to Scalable Quantum Chips

The researchers went on to use their method to build a fully functioning quantum chip that outperforms all previous designs.

Because the error rate is multiplicative, this type of improvement scales exponentially with system size. As a result, the 10-15x improvement in error rate for individual qubits has a much larger effect on a multi-qubit computer.

Importantly, such a qubit is not an exotic new concept, but simply a “traditional” superconducting qubit using a different material, so they can easily be integrated into existing quantum computers and be used by existing quantum computing software.

“Swapping Princeton’s components into Google’s best quantum processor, called Willow, would enable it to work 1,000 times better.

The benefits of the Princeton qubit grow exponentially as system size grows, so adding more qubits would bring even greater benefit.”

Andrew Houck, Princeton’s dean of engineering

This means Princeton’s design could enable a hypothetical 1,000-qubit computer to work roughly 1 billion times better.

Even better, the use of tantalum and silicon means that the manufacturing method fits the ones already in use by the semiconductor industry, making mass production a much easier milestone to reach than an entirely new technology.

This research seems to indicate that silicon quantum chips, which we discussed previously, are likely the right direction for the quantum computing industry.

Together with better quantum light sources, hybrid quantum-photonic chips, and the possibility to carry quantum information together with normal telecom data flow, these steps toward much larger quantum computers show that the technology is quickly reaching commercial maturity.

Investing in Quantum Computing Innovation

1. Alphabet Inc.

Google is very active in quantum computing, mostly through its Google Quantum AI lab and Quantum AI campus in Santa Barbara.

Google’s quantum computer made history in 2019 when it claimed to have achieved “quantum supremacy” with its Sycamore machine. The machine performed a calculation in 200 seconds that would have taken a conventional supercomputer 10,000 years.

This is now dwarfed by its newest chip’s performance, called Willow. This is the very first quantum computing chip that has an error rate low enough that the more qubits you add, the less error you get. It makes it the very first scalable quantum chip design.

But maybe the greatest contribution of Google will be in software, an activity where it has an impressive track record, actually better than in hardware (Search, G Suite, Android, etc.).

Already, Google’s Quantum AI makes available a suite of software designed to assist scientists in developing quantum algorithms.

It also openly advocates for “researchers, engineers, and developers to join us on this journey by checking out our open source software and educational resources, including our new course on Coursera, where developers can learn the essentials of quantum error correction and help us create algorithms that can solve the problems of the future.”

Thanks to this open approach, Google is now leading in hardware as well as its cloud solutions. Google might be one of the companies setting the standards of quantum computing software and quantum programming, giving it a privileged position to direct the field’s future evolution.

Meanwhile, AI solutions, including Waymo’s self-driving car, might become the new revenue driver for Alphabet, which still holds a massively dominant position in the search & ads industries.

You can learn more about Google’s non-quantum-related activities, especially ads and AI, in our dedicated report from December 2024.

Latest Alphabet (GOOGL) Stock News and Developments

Study Referenced:

^{1. Bland, M.P., Bahrami, F., Martinez, J.G.C. et al. Millisecond lifetimes and coherence times in 2D transmon qubits. Nature 647, 343–348 (2025). https://doi.org/10.1038/s41586-025-09687-4}