Real-Time Magnetic Field Sensing based on Microwave Frequency Modulated Photocurrent of Nitrogen-Vacancy Centers in Diamond

While photoelectric detection of magnetic resonance (PDMR) can be applied to miniaturize nitrogen-vacancy (NV) center-based quantum sensors, real demonstration of PDMR-based magnetic field sensing remains as a distinctive challenge. To tackle this challenge, in this article, we fabricate diamond samples with electrodes and microwave antenna on the surface, and realize PDMR by detecting photocurrent in picoampere range via various lock-in amplifying modes. We obtain a theoretical and experimental sensitivity 397 nT/Hz and 921 nT/Hz of magnetic field detection in DC-10 Hz range with a laser intensity and microwave frequency modulation mode, respectively, and demonstrate for the first time, a real-time tracking of alternating magnetic field with a standard deviation of 1.5 uT. Furthermore, we investigate systematically the dependence of the PDMR contrast, linewidth and the sensitivity on the laser and microwave power, and find a perfect agreement with a master equation-based theory. Thus, our results represent a critical step forward in transitioning PDMR from a spectroscopic technique to a practical sensing modality.

💡 Research Summary

In this work the authors demonstrate a practical magnetic‑field sensor based on photo‑electric detection of magnetic resonance (PDMR) of nitrogen‑vacancy (NV) centers in diamond, moving the technique from a purely spectroscopic tool to a real‑time sensing modality. A diamond sample grown by microwave‑plasma‑enhanced CVD is fabricated with interdigitated gold electrodes and a gold micro‑belt microwave antenna on its surface. A 532 nm laser is focused between the electrodes while a 50 V dc bias is applied, generating a pico‑ampere‑scale photocurrent that is amplified by a low‑noise transimpedance pre‑amplifier and demodulated with a lock‑in amplifier (LIA).

Two modulation schemes are explored. First, the laser intensity is square‑wave modulated at 321 Hz. This yields a conventional dip‑shaped PDMR spectrum with a contrast of 12.1 % and a linewidth of 19.5 MHz centered at 2.869 GHz. Using the shot‑noise limited sensitivity formula η = 4h · 3√3 · gμB · Γ · C · √R⁻¹, where R is the detected photocurrent rate, the authors estimate a magnetic‑field sensitivity of 397 nT · Hz⁻¹ᐟ².

Second, the microwave frequency is sinusoidally modulated with a 1 MHz deviation. This produces a dispersive (derivative‑like) PDMR response whose lock‑in voltage varies linearly with the microwave detuning near resonance. Fitting the dispersive curve gives a slope S ≈ 2.0 × 10⁻⁸ V · MHz⁻¹ and a linewidth of 15.3 MHz. Because the electron gyromagnetic ratio is γ = 2.8 MHz · G⁻¹, the magnetic field can be directly extracted as B = V/(S·γ). The measured voltage noise σ = 2.27 × 10⁻⁹ V together with an equivalent noise bandwidth of 10 Hz yields an experimental sensitivity of 898 nT · Hz⁻¹ᐟ², in good agreement with the theoretical prediction.

To validate the real‑time capability, the lock‑in voltage is recorded continuously under resonant microwave conditions while a weak alternating magnetic field is applied. By converting the voltage stream to magnetic field using the calibrated linear factor, the authors track the AC field with a standard deviation of 1.5 µT over a 1 Hz bandwidth, demonstrating true real‑time magnetic‑field monitoring.

The dependence of the PDMR performance on laser power and microwave power is systematically investigated. Photocurrent versus laser power follows J = aP + bP², with a = 6.6 pA · mW⁻¹ (linear background) and b = 0.079 pA · mW⁻² (quadratic NV‑related contribution). At 100 mW the quadratic term dominates, confirming that two‑photon ionization/recombination of NV⁻/NV⁰ is the main charge‑generation mechanism. Microwave power scans reveal an optimal power around 26 dBm, where contrast and linewidth are maximized without excessive power broadening.

All experimental observations are quantitatively reproduced by a master‑equation model that includes the NV electronic level structure, intersystem crossing rates, and charge‑state dynamics. The agreement validates the model as a design tool for further optimization, suggesting that higher NV concentrations, improved electrode geometry, and optimized biasing could push the sensitivity into the tens‑of‑nanotesla per root‑hertz regime.

In summary, the paper makes four key contributions: (1) fabrication of an integrated diamond device that enables pico‑ampere PDMR readout, (2) implementation of both laser‑intensity and microwave‑frequency lock‑in detection, (3) demonstration of a linear, dispersive PDMR signal that allows real‑time magnetic‑field extraction, and (4) a comprehensive theoretical‑experimental framework that accurately predicts device performance. These advances pave the way for compact, fully electrical NV‑based magnetometers that can be integrated into chip‑scale platforms without the need for bulky optics, opening new opportunities for portable quantum sensing applications.