Skip to content

A defect in diamond with millisecond-scale spin relaxation time at room temperature

FieldValue
JournalOpen MIND
Published2026-03-07
AuthorsSounak Mukherjee, Anran Li, Johannes Eberle, Sean Karg, Zi-Huai Zhang
DOI10.48550/arxiv.2603.07265

The WAR5 defect in chemical vapor deposited diamond produced a spin relaxation time of 0.97(27) ms at room temperature. This is among the longest electron-spin relaxation times reported for a solid-state defect at room temperature.

Performance relative to known solid-state defects

Section titled “Performance relative to known solid-state defects”

Prior work cited in the paper gives NV centers a T1 of 6.67 ms and P1 centers about 2 ms at room temperature. The cited room-temperature T1 for silicon vacancies in silicon carbide was 500 µs. WAR5 exceeds that value but remains shorter than both NV and P1 centers.

Long T1 did not translate into equally long free-induction coherence. At room temperature, Hahn echo T2 was only 4.9(7) µs, far below the 2T1 limit. The authors attribute this coherence limitation to the sample’s P1 spin bath.

QuantityMeasurement conditionWAR5 resultReference point
Spin relaxation T14 K14.38(19) minLonger than NV in the same sample
Hahn echo T24 K246(7) µsNV in the same sample: 183(2) µs
CPMG T24 K, 1,024 π pulses6.49(34) msLimited by minimum pulse spacing
Hahn echo T2Room temperature4.9(7) µsWell below the 2T1 limit

The microscopic identity of WAR5 remains tentative. The authors assign it tentatively to the neutral oxygen vacancy, OV0, because its growth chemistry, spin Hamiltonian, and calculations support that model.

The tested sample was grown by chemical vapor deposition using high oxygen chemistry and nitrogen doping. Pulsed electron spin resonance operated in the X band at about 9.7 GHz. A helium flow cryostat provided temperature control, and the magnetic field was aligned along a ⟨111⟩ crystal axis.

At 4 K, 455 nm illumination for 10 s polarized the spins before measurement. The WAR5 transition appeared at 242.95 mT, separate from the three NV hyperfine peaks centered at 243.38 mT.

Fitting the orientation-dependent spectrum gave D = 2.8877(52) GHz and g = 2.0030(6) at 4 K. Calculations predict C3v symmetry, an S = 1 ground state, and D = 2.989 GHz for OV0. The measured Hamiltonian is consistent with this assignment, but the authors retain the tentative wording.

Conventional inversion recovery produced biexponential decay because spectral diffusion moved spins outside the ESR detection bandwidth. The authors therefore used saturation recovery, followed by Hahn echo detection, to measure T1.

The measured T1 values were 14.38(19) min at 4 K, 1.35(11) min at 77 K, and 0.97(27) ms at room temperature. The relaxation rate rose linearly to about 60 K, consistent with a direct single-phonon process. Above that region, an Orbach two-phonon process dominated the fitted relaxation rate.

At 54.3(8.7) meV and 137.8(10.9) meV, the fitted activation energies matched calculated spin-phonon coupling peaks at 58.8 meV and 143.0 meV. Oxygen vibrations dominated the lower phonon group near 69.2 meV, while carbon vibrations dominated the higher group near 149.8 meV.

Room-temperature relaxation was slightly faster than the model predicted. The authors note that this mismatch may indicate additional contributions, such as Raman processes.

Hahn echo measurements gave T2 = 246(7) µs at 4 K and 4.9(7) µs at room temperature. WAR5 had β = 1.22(4), while NV centers in the same sample had 183(2) µs and β = 1.44(3).

That low NV value also fell below typical cryogenic NV coherence, indicating a common noise source. Spin counting gave WAR5 concentration about 4 × 10^15 cm^-3, NV 1.7 × 10^16 cm^-3, and P1 5 × 10^17 cm^-3. The authors attribute the coherence limit to dense P1 electron spins.

A gCCE simulation of the spin bath produced β about 1.23, close to the measured 1.22(4). CPMG dynamical decoupling raised T2 to 6.49(34) ms at 4 K with 1,024 π pulses. The fitted power-law exponent was 0.507(6), consistent with square-root scaling with pulse count. The measurement was limited by the smallest usable pulse spacing.

A double Lorentzian spectral decomposition assigned one noise component to P1 spins and the other to natural-abundance carbon-13 nuclear spins. The authors suggest that samples with fewer background impurities could support much longer coherence times.

Optical spin polarization was demonstrated from 405 to 500 nm. At 10 K and 50 mW optical power, 455 nm produced 5.45% polarization, while 405 nm produced 5.16%. Thermal polarization saturated at 2.33% at 10 K.

The fitted buildup times were 55(4) s for 455 nm and 2.3(1) min for 405 nm. At the same temperature, thermal buildup required 16.3(1.8) min.

WAR5 polarization was strongest with the field aligned along ⟨111⟩ and was slower than NV polarization. Under the same optical configuration, NV centers reached 22.5% polarization with a 1.45 s characteristic time using 532 nm.

Broadband scans found a feature near 491 nm. The authors inferred that the WAR5 zero-phonon line lay between approximately 480 and 500 nm.

Zero-phonon-line search and electronic structure

Section titled “Zero-phonon-line search and electronic structure”

Photoluminescence spectra contained many lines from 450 to 550 nm, so the WAR5 zero-phonon line could not be isolated from background centers. Narrow linewidth scans found no ESR resonance from 489 to 494 nm or around 543 nm.

The 484 nm line was unique to the WAR5 sample among the compared oxygen-grown samples and had been assigned to an oxygen-related defect. It remains a candidate, not a confirmed assignment.

The 543 nm line also appeared in oxygen-grown and oxygen-implanted samples. The authors reject it as the WAR5 zero-phonon line and tentatively assign it to positively charged oxygen vacancies, OV+.

DFT predicts an OV0 zero-phonon line at 2.58 eV, or 480.6 nm, with a Huang-Rhys factor of 3.45. The transition dipole moment was 0.25 e·Å, and the calculated radiative rate was 3 × 10^6 s^-1.

The calculated vertical absorption energy was 2.88 eV, or 430.5 nm, with a sideband extending from 480.6 to 389.9 nm. This agrees with the observed 405 to 480 nm polarization response.

The proposed optical spin polarization model uses a singlet state 0.35 eV above the triplet ground state. It also includes a metastable triplet 0.45 eV above the ground state. Spin-flip paths through these states could create spin selectivity. The authors note that nonradiative competition may suppress the zero-phonon emission.

Three limits matter for device use. First, room-temperature Hahn echo coherence was only 4.9(7) µs despite the millisecond T1. Second, optical polarization saturated at 5.45% and required tens of seconds to minutes.

Third, the zero-phonon line and optical readout remain unresolved. Sample-specific charge dynamics and a wide linewidth distribution can hide narrow optical resonances. The dense P1 bath limits usable coherence despite the long intrinsic relaxation time.

The authors propose searching for other defects that incorporate light heteroatoms near the top of the periodic table. They argue that low spin-orbit coupling and small lattice distortion may support long-lived spins. For a practical sensor, lower P1 concentration, controlled illumination, and a confirmed zero-phonon line are the concrete next requirements.

For defect-based quantum sensing, the relevant material family is MPCVD-grown single-crystal diamond. The study requires high-oxygen-chemistry growth and nitrogen doping; the diamond materials catalog lists single-crystal and polycrystalline MPCVD material, including plates and wafers up to 125 mm. Those defect-specific growth and specimen requirements should be confirmed through engineering support.

View Original Abstract

Spin defects in diamond are promising platforms for quantum sensing. The longest electron spin relaxation times ($T_1$) at room temperature for solid-state defects are observed in nitrogen vacancy centers in diamond, which can reach 6.67 ms, and substitutional nitrogen (“P1 centers”) in diamond, which exhibit a $T_1$ of 2 ms. No other solid-state defect has exhibited millisecond-scale spin relaxation times at room temperature thus far. Here, we characterize the spin properties of the WAR5 defect in diamond with pulsed electron spin resonance. The observed $T_1$ is one of the longest for solid-state spin defects: 0.97(27) ms at room temperature and 14.38(19) min at 4 K. The observed coherence time ($T_2$) is 246(7) $μ$s, which can be extended to 6.49(34) ms at 4 K with dynamical decoupling. Furthermore, we demonstrate optical spin polarization with a range of wavelengths from 405 nm to 500 nm and propose potential zero-phonon line candidates.