Atomic magnetometry based on the ground-state Hanle effect in an elliptically polarized light wave

TL;DR

This work demonstrates a SERF-free, zero-field atomic magnetometer based on the ground-state Hanle effect in Cs using a single elliptically polarized light beam scanned in the Voigt geometry. By detecting magneto-optical resonances through changes in ellipticity with a differential polarimeter, laser-intensity noise is substantially suppressed without large detuning. A Bloch-equation–based theory yields a Lorentzian GSHE response $M_z = \frac{\Delta(R_c - R_p)}{\Delta^2 + \Omega^2} M_0$ with $\Delta = \Gamma + R_c + R_p$ and $\Omega = \gamma B_x$, and experiments in a compact $\approx0.125$ cm$^3$ Cs cell at $T \approx 85^ obreak extdegree$C achieve a magnetic-field sensitivity of about $180$ fT/√Hz (limited by technical noise) and a 200 Hz bandwidth, with a photon-shot-noise limit near $5$ fT/√Hz. The approach offers notable advantages over SERF sensors, including lower operating temperature and relaxed magnetic shielding, and holds promise for biomedical sensing such as magnetocardiography (MCG).

Abstract

We investigate the ground-state Hanle effect in alkali-metal vapor irradiating by a resonant elliptically polarized light wave. The magneto-optical resonances are observed as a change in the ellipticity parameter of the light wave polarization when scanning the transverse magnetic field near zero. We use a miniature ($\approx\,$$0.125$ cm$^3$) glass cesium vapor cell heated to a relatively low temperature of $\approx\,$$85^\circ$C. Under the current experimental conditions, the sensitivity of magnetic field measurements is limited by a technical noise, reaching $180$ fT/$\surd$Hz in a $200$ Hz bandwidth. The ultimate photon-shot-noise-limited sensitivity of the method is estimated to be $\approx\,$$5$ fT/$\surd$Hz. The proposed scheme is promising for the development of a zero-field atomic magnetometer with reduced heat dissipation of the sensor head and relaxed requirements for magnetic shielding compared to counterparts operating in the spin-exchange relaxation-free regime. These features are of particular value for biomedical applications.

Figures (6)

• Figure 1: Transformation of the light wave polarization upon passing through the sensor head: $\lambda/4$, quarter-wave plate; Cs, cesium vapor cell; WP, Wollaston prism; BPD, balanced photodetector; Ch1, Ch2, Diff, photodetector outputs corresponding to channel 1, channel 2, and differential (balanced) channel, respectively. ${\bf E}_{\it in}$ is the linearly polarized input wave with the wave vector ${\bf k}_{\it in}$.
• Figure 2: The transmitted intensity of the pump wave (upper curve) and the probe wave (lower curve) after passing through the vapor cell as a function of the transverse magnetic field (Larmor frequency in units of $\Gamma$). Parameters used for the calculation as follows: $I_p(0)$$\,=\,$$I_c(0)/3$, $R_c$$\,= \,$$3R_p$$\,=\,$$3\Gamma$, $\varkappa L$$\,=\,$$0.2$. The curves are normalized to $I_c(0)$.
• Figure 3: (a) Schematic of the experimental apparatus: ECDL, external-cavity diode laser; $\lambda/2$, $\lambda/4$, half- and quarter-wave plates, respectively; PM fiber, polarization-maintaining fiber; PBS, polarizing beam splitter; Cs, cesium vapor cell containing buffer gas; WP, Wollaston prism; BPD, balanced photodetector. (b) Measured absorption profile of the vapor cell at $T$$\,\approx\,$$80\,^\circ$C and $P$$\,\approx\,$$1$ mW.
• Figure 4: (a) MOR signals recorded from the three BPD channels at $P$$\,\approx\,$$300$$\mu$W (the photodetector bandwidth is $\approx\,$$1$ MHz). (b), (c) The half-width at half maximum and amplitude of the resonance as functions of the optical power at the input of the vapor cell. (d) Signal noise from the balanced BPD channel (magenta) measured on the resonance slope at $P$$\,\approx\,$$300$$\mu$W. The contribution of the technical noise of the photodetector is shown as black data, while the photon-shot-noise limit is shown as a horizontal solid line. (e), (f) Signal-to-noise ratio and sensor sensitivity as functions of optical power, calculated according to Eq. (\ref{['Sensitivity']}). In all measurements $T$$\,\approx\,$$85$$\,^\circ$C and $\epsilon$$\,\approx\,$$10^\circ$. Solid lines in panels (b), (c), (e), and (f) are drawn as a guide to the eye.
• Figure 5: (a) Resonance slope as a function of the modulation frequency at a fixed scan amplitude. (b) Resonance slope as a function of the modulation frequency at a fixed modulation index, i.e., the ratio of amplitude to modulation frequency. $T$$\,\approx\,$$85\,^\circ$C, $P$$\,\approx\,$$500$$\mu$W, $\epsilon$$\,\approx\,$$10^\circ$. Solid lines are drawn as a guide to the eye.