Towards noble gas quantum optical magnetometry using direct ultraviolet detection

AI summary

Overview

Research area: Quantum sensing — specifically optical magnetometry using hyperpolarized noble gas (¹²⁹Xe) read out by two-photon excitation with an ultraviolet frequency comb laser.

Technical level: Advanced. The paper is a theoretical noise-analysis and sensitivity-modelling study; it assumes familiarity with atomic physics, Larmor precession, two-photon spectroscopy, and photodetector noise metrology.

Scope (one sentence): The paper calculates the magnetic-field sensitivity limits of an alkali-free ¹²⁹Xe magnetometer detected via Doppler-free two-photon ultraviolet excitation, and identifies the laser power, intensity noise, and beam size needed to beat state-of-the-art magnetometers.

What This Paper Is About

The authors ask whether a noble-gas magnetometer can be made competitive with the best existing magnetic sensors without ever mixing alkali vapour into the sensing gas. The idea is to spin-polarize ¹²⁹Xe beforehand, let it precess in a magnetic field, and read out that precession optically using a 256 nm ultraviolet laser that drives a two-photon transition; the resulting infrared fluorescence oscillates at the Larmor frequency and encodes the field strength. The paper's goal is to work out, from first principles and with realistic component values, what laser power, laser intensity noise, and beam size would be required to reach or exceed state-of-the-art sensitivity.

Key Contributions

1. First treatment of laser intensity noise in alkali-free ¹²⁹Xe magnetometry. The authors state that the impact of relative intensity noise (RIN) on this class of magnetometer had been overlooked in earlier proposals, and they derive its contribution to Larmor-frequency estimation error.

2. A full noise budget separating fundamental and technical limits. Photon shot noise and spin projection noise are treated as fundamental, while technical RIN and photodiode/detector noise are treated separately, so that experiments can be directed toward either raising laser power or lowering RIN depending on which regime they occupy.

3. A balanced-detection scheme with a low-power UV reference. A small fraction (less than 10 mW) of the UV power is sent to a balanced detector together with the collected infrared fluorescence, with the reference power chosen so the detector output is zero when the xenon is fully depolarized; common-mode RIN rejection is modelled with the detector's common-mode rejection ratio.

4. A concrete sensitivity prediction and a laser architecture to reach it. The analysis predicts 24 aT/√Hz in 10 minutes using a 10 W UV laser with RIN of −100 dBc/Hz, described as 6 times better than the current record, and proposes the fourth harmonic of a high-power ytterbium-doped fiber frequency comb as the UV source.

Main Findings

• Predicted sensitivity: 24 aT/√Hz is achievable in 10 minutes with 10 W of UV light at a RIN of −100 dBc/Hz; the paper describes this as 6 times better than the current record.
• Benchmark context: The paper cites 150 aT/√Hz for SQUIDs (which require cryogenic cooling), more than 1000 fT/√Hz for diamond NV centres (limited by millisecond-scale spin lifetimes), and 160 aT/√Hz for alkali atomic magnetometers in heated buffer gas.
• Power matters: Previous two-photon ¹²⁹Xe work used about 200 mW of useful UV power, sufficient only for spectroscopy at roughly 1 mbar with beam radii much less than 1 mm; the analysis considers laser powers from 1 W to 100 W, and higher power than used to date is required to reach or exceed state-of-the-art sensitivity.
• RIN matters: The technical RIN values considered span −80 to −120 dBc/Hz. For a 1 W UV laser, the shot-noise contribution to RIN would be −178 dBc/Hz, at least 10⁶ times below these technical values, so shot noise of the high-power excitation light is neglected.
• Nonlinear noise amplification: Because the two-photon excitation rate scales with the square of the excitation intensity, intensity noise in the infrared fluorescence is 4 times the RIN of the excitation light.
• Detected-light shot noise is not negligible: The detected UV power is estimated at 16 µW to 1.5 mW, corresponding to shot-noise RIN of −130 to −149 dBc/Hz, comparable to the residual technical noise after 45 dB common-mode rejection.
• Beam size must be optimized: Optimal beam radii are found to range from 0.1 to 10 mm, balancing spin projection noise against photon shot noise.
• Two operating regimes: Depending on parameters, either fundamental noise or technical noise dominates, which tells an experimentalist whether to increase laser power or reduce laser intensity noise.
• Detector dark noise is negligible: The electronic noise contribution is estimated to be six orders of magnitude lower than any other contribution and is neglected.
• Long spin lifetime is the central advantage: A spin dephasing time T₂ of 8000 s is assumed, following demonstrations of T₂ as long as 8000 s in a 4:1 xenon–nitrogen mix at about 50 mbar, with lower pressure correlated with longer T₂; room-temperature operation is assumed at 293 K.

Methodology in Plain English

The authors build an analytical model of a specific experiment rather than running a device. A UV frequency comb beam is split: a small portion goes to a balanced detector as a reference, and the rest is circularly polarized and focused into a xenon cell, then retroreflected so that counter-propagating beams cancel the Doppler shift. The circular polarization is chosen so that only one of the two ground-state magnetic sublevels (¹²⁹Xe has nuclear spin I = 1/2) can be excited, on a transition at 256 nm to the 5p⁵(²P₃/₂)6p²[5/2]₂, F′ = 3/2 state; the excited atoms decay in 38 ns, emitting infrared light that is collected onto the other side of the balanced detector.

In a magnetic field perpendicular to the quantization axis, the ground-state populations oscillate at the Larmor frequency ω_B = γ_Xe B, with γ_Xe = 2π × 11.78 MHz/T, so the infrared fluorescence is a decaying sinusoid whose oscillation frequency gives the field. Because the decay is slow, the frequency can be estimated precisely. The authors then take a standard formula for the frequency-estimation variance of a decaying sinusoid and feed it the noise spectral density of each noise source in turn: spin projection noise, shot noise on the reference UV and on the infrared light, technical RIN propagated through both the reference arm and the nonlinear two-photon excitation, detector noise-equivalent power, and dark noise. These variances are added in quadrature and converted to a sensitivity in T/√Hz using the gyromagnetic ratio, measurement time, and number of repeated measurements. All numerical inputs — polarization 0.90, ¹²⁹Xe isotopic fraction 0.80, 1 mbar pressure, 293 K, 66.7 ps pulses giving a 2 cm interaction length, two-photon absorption coefficient 83 cm⁴/J², natural linewidth 4.3 MHz, pressure broadening 18.6 MHz/mbar, 0.25 collection efficiency, 0.68 infrared and 0.45 UV quantum efficiencies — are taken from the literature or from the Thorlabs PDB220A2 balanced detector datasheet (0.5 A/W responsivity, 3.6 pW/√Hz NEP, 45 dB CMRR).

Why This Matters

The work argues that a noble-gas magnetometer can in principle combine the two things that usually trade off against each other: room-temperature operation and record-level sensitivity. Because the noble gas is separated from the alkali vapour used for hyperpolarization before detection, the sensing stage is alkali-free, and the closed electron shell of xenon leaves only nuclear spin, giving spin relaxation times measured in hours rather than milliseconds. If the predicted sensitivity is realized, it would surpass both cryogenic SQUIDs and SERF alkali magnetometers without cryogenics, and the noise budget tells experimenters exactly which knob — laser power or laser RIN — limits them.

Real-world applications the paper cites for sub-femtotesla/√Hz sensing:

• Biomedical imaging and sensing
• Low-field nuclear magnetic resonance spectroscopy and chemical analysis via NMR
• Navigation
• Fundamental physics tests, including a proposed search for quantum gravity effects and the search for the neutron electric dipole moment
• Environmental sensing and resource exploration

Industry relevance: The result sets concrete targets for laser and photonics suppliers — high-power UV frequency combs with RIN around −100 dBc/Hz, efficient fourth-harmonic generation from ytterbium fiber lasers, and balanced detectors with high common-mode rejection — and it points toward practical, non-cryogenic sensors for medical imaging, field-deployable NMR, and navigation hardware.

Future Directions

• Experimental demonstration: The paper notes that no magnetometry has yet been demonstrated with two-photon ¹²⁹Xe excitation, to the best of the authors' knowledge; realizing the predicted sensitivity is the obvious next step.
• Higher UV power and conversion efficiency: The proposal relies on frequency-quadrupling a 100 W-class ytterbium fiber comb with an expected overall conversion efficiency of a few percent, so improving deep-UV generation is a key engineering question.
• Lower-RIN laser sources and better common-mode rejection: Since technical RIN is a dominant limit in part of the parameter space, pushing RIN below the −100 dBc/Hz target or improving balanced-detector rejection beyond 45 dB would directly improve sensitivity.
• Systematic effects and parameter choices: The analysis assumes a specific pressure, isotope fraction, collection efficiency, and dephasing time; exploring how sensitivity scales with these — and validating them experimentally — remains open.

Target Audience

Quantum sensing and atomic magnetometry researchers, AMO physicists interested in noble-gas hyperpolarization and two-photon spectroscopy, and laser/photonics engineers working on high-power ultraviolet frequency combs and low-noise balanced detection. Graduate students entering optical magnetometry will also benefit from the explicit noise budget, though the derivations assume a solid foundation in atomic physics and noise analysis.

Authors’ abstract

We analyze a novel approach for quantum magnetic sensing via optical detection of nuclear spin precession in a noble gas. The detection is carried out with an ultraviolet frequency comb laser, so no alkali vapor is required to be mixed with the noble gas during the sensing stage, allowing for hours-long spin relaxation time at room temperature. Our analysis reveals that combining high laser power and low laser intensity noise is crucial to achieving and exceeding state-of-the-art magnetometric performance, including that of cryogenic magnetometers. We find operational regimes where either fundamental or technical noise dominate, which directs experimental effort towards increasing laser power or lowering laser intensity noise, respectively. We also show that optimizing beam size is key to achieving optimal sensitivity. Our calculations predict that a record-breaking sensitivity of $24~\mathrm{aT}/\sqrt{\mathrm{Hz}}$ is achievable in 10 minutes with a 10 W ultraviolet laser with a relative intensity noise of -100 dBc/Hz.