Short answer: Niobium nitride has a much higher superconducting transition temperature than aluminum, which makes qubits built from it less sensitive to some heat-related noise. A 2025 Nature Materials paper reports all-nitride transmons made by atomic layer deposition, with microsecond-scale relaxation times at temperatures above 300 millikelvin. A 2024 PRX Quantum paper reports niobium-trilayer transmons at frequencies up to 24 GHz that tolerate higher temperatures. Both are laboratory studies of small devices.
Why aluminum limits the operating temperature
The PRX Quantum abstract (Anferov and coauthors, published September 2024) explains the motivation. Current superconducting microwave qubits are cooled to extremely low temperatures to avoid decoherence. Higher operating temperatures would increase the cooling power available, which the authors say is desirable for scaling up qubit counts and for experiments that need more heat dissipation. To get there, two problems must be addressed: quasiparticle decoherence, which becomes significant for aluminum junctions above 160 millikelvin, and dephasing from thermal microwave photons, which is problematic above 50 millikelvin.
2024: niobium trilayer junctions at high frequency
The authors use low-loss niobium-trilayer junctions, which are less sensitive to quasiparticles because niobium has a higher superconducting transition temperature. They fabricate transmons at frequencies higher than previously studied, up to 24 GHz. They report that decoherence is unaffected by quasiparticles up to a temperature that the retrieved abstract text does not display clearly, and that the qubits can operate up to roughly a higher temperature while maintaining similar performance. The paper's title states "above 20 GHz operating over 200 mK". Because the numeric values in the abstract did not render, this post uses only the title's figures. The authors say the thermal resilience creates new options for scaling up quantum processors.
2025: all-nitride trilayers by atomic layer deposition
The Nature Materials abstract says that large-scale quantum processors benefit from qubits that can operate at elevated temperatures and be made with scalable, foundry-compatible processes. The authors report superconducting qubits based on NbN/AlN/NbN trilayers deposited entirely by atomic layer deposition (ALD). By varying the number of ALD cycles used for the AlN barrier, they achieve Josephson tunnelling through barriers of different thicknesses, with critical current density spanning seven orders of magnitude, which they present as showing uniformity and versatility of the process.
Because of the high critical temperature of NbN, transmons made from these trilayers show microsecond-scale relaxation times even above 300 millikelvin. The authors conclude that ALD is a viable low-temperature deposition technique for superconducting quantum circuits and that all-nitride ALD qubits are a promising platform for operation at elevated temperatures.
Limits
Microsecond-scale relaxation times are well below the millisecond values reported for the best aluminum or tantalum transmons at colder temperatures, so the appeal here is the temperature range and the fabrication route, not raw coherence. The abstracts report single devices and small test structures. A third source on this topic, a national research institute press page about an earlier all-nitride qubit, is not used here.
FAQ
Why does a higher transition temperature help?
It reduces sensitivity of the junction to quasiparticles, per the 2024 abstract.
What temperature was reported?
Above 300 millikelvin for the 2025 all-nitride transmons, and over 200 millikelvin per the 2024 title.
What is ALD?
Atomic layer deposition, a thin-film growth method used in industry, as the 2025 abstract describes it.
Are these coherence times competitive?
They are microsecond-scale. The authors stress temperature and scalability.
Are these in commercial processors?
The abstracts report no such use.
All field notes