Integrated microcavity electric field sensors using Pound-Drever-Hall detection

Nat. Commun. 15, 1386 (2024) · DOI: 10.1038/s41467-024-45699-w

License: CC BY 4.0.

Abstract

Discerning weak electric fields has important implications for cosmology, quantum technology, and identifying power system failures. Photonic integration of electric field sensors is highly desired for practical considerations and offers opportunities to improve performance by enhancing microwave and lightwave interactions. Here, we demonstrate a high-Q microcavity electric field sensor (MEFS) by leveraging the silicon chip-based thin film lithium niobate photonic integrated circuits. Using the Pound-Drever-Hall detection scheme, our MEFS achieves a detection sensitivity of 5.2 $\mu$V/(m$\sqrt{\mathrm{Hz}}$), which surpasses previous lithium niobate electro-optical electric field sensors by nearly two orders of magnitude, and is comparable to atom-based quantum sensing approaches. Furthermore, our MEFS has a bandwidth that can be up to three orders of magnitude broader than quantum sensing approaches and measures fast electric field amplitude and phase variations in real-time. The ultra-sensitive MEFSs represent a significant step towards building electric field sensing networks and broaden the application spectrum of integrated microcavities.

Figures

20 panels with data across 4 figures. Each panel page shows the plot, its columns and its files; each data.csv begins with a header naming the paper, the panel, the source, the license and the provenance route.

Fig. 1

Illustrative figure, no extractable data. Shown in the paper PDF.

Fig. 2

• panel (b): Transmission of Device 1 (arb. units) against relative frequency (GHz): the measured transmission (blue dots), its Lorentzian fit (orange line) and the MZI interferogram (light blue line), which the print, like this panel, draws rescaled below the dip. data.csv
• panel (c): Transmission of Device 2 (arb. units) against relative frequency (GHz): the measured transmission (blue dots), its Lorentzian fit (orange line) and the MZI interferogram (light blue line), which the print, like this panel, draws rescaled below the dip. data.csv
• panel (d): PDH error signal (mV) against relative frequency (GHz) for Device 1 (red) and Device 2 (blue). The sketched sensing signal, electric field and slope arrows of the print are not reproduced. data.csv

Fig. 3

• panel (a-1): Temporal signal of Device 1 with a 100 MHz input: voltage (mV) against time (ns), measurement (blue dots) and sine fit (red line). Upper plot of panel (a). data.csv
• panel (a-2): Temporal signal of Device 2 with a 100 MHz input: voltage (mV) against time (ns), measurement (blue dots) and sine fit (red line). Lower plot of panel (a). data.csv
• panel (b-1): Sensing signal of Device 1 in the electric spectrum (100 MHz, RBW = 200 Hz): sensor response (dBm) against offset frequency (kHz) under 95 dB$\mu$V/m (blue) and 105 dB$\mu$V/m (red). The print labels the y axis 20 dB/div without numbers; here it is in dBm. Upper plot of panel (b). data.csv
• panel (b-2): Sensing signal of Device 2 in the electric spectrum (100 MHz, RBW = 200 Hz): sensor response (dBm) against offset frequency (kHz) under 95 dB$\mu$V/m (blue) and 105 dB$\mu$V/m (red). The print labels the y axis 20 dB/div without numbers; here it is in dBm. Lower plot of panel (b). data.csv
• panel (c): Linearity of the sensors: sensor response at 100 MHz (dBm) against applied electric field (dB$\mu$V/m) for Device 1 and Device 2 with the ECDL (red and blue circles) and Device 1 with the fiber laser (pink squares), with linear fits (black; dashed for the fiber laser) and the dashed noise-floor line. The broken y axis of the print is shown continuous. data.csv
• panel (c-inset): Inset of panel (c): intercept (dBm) of the log-log fits against $P_{\mathrm{in}}/(\kappa/2\pi)$ (mW/GHz) for Devices 1 to 4 (green dots, with error bars) and the slope-2 line (grey). data.csv
• panel (d): Frequency response of Device 1 (red dots) and Device 2 (blue dots): sensor response (dB) against frequency (GHz), with Lorentzian fits (black lines) and the dashed -3 dB line. The print labels the y axis 5 dB/div without numbers; here 0 dB is the top of the fits. data.csv
• panel (d-inset): Inset of panel (d): 3 dB bandwidth (MHz) against $\kappa/2\pi$ (MHz) for Devices 1 to 4 (green dots; error bars the 95% confidence bounds of the Lorentzian fit) and the slope-0.5 line (grey). data.csv

Fig. 4

• panel (a): Measurement of a double exponential wave: normalized output (arb. units) against time ($\mu$s), sensor output (blue) and applied signal (red dashed). data.csv
• panel (a-inset): Inset of panel (a): the sharp leading edge of the waveform, normalized output (arb. units) against leading time ($\mu$s), sensor output (blue) and applied signal (red dashed). data.csv
• panel (b): Measured decay time ($\mu$s) against set decay time ($\mu$s): sensor output (blue dots; error bars the standard deviation over multiple measurements) and applied signal (red line). data.csv
• panel (c): Measurement of a Gaussian pulse: normalized output (arb. units) against time ($\mu$s), trigger signal from the function generator (green), sensor output (blue) and applied signal (red dashed). data.csv
• panel (d): Measured delay change (ns) between the Gaussian pulse envelope peak and the trigger against set delay change (ns): sensor output (blue dots; error bars the standard deviation over multiple measurements) and applied signal (red line). data.csv
• panel (d-inset): Inset of panel (d): the 0 to 10 ns range enlarged, measured against set delay change (ns), sensor output (blue dots, with error bars) and applied signal (red line). The print puts its y axis on the right. data.csv
• panel (e): Measurement of a frequency modulated continuous wave (FMCW) signal: normalized output (arb. units) against time ($\mu$s), sensor output (blue) and applied signal (red). data.csv
• panel (f): Instantaneous frequency (MHz) retrieved from the FMCW signal against time ($\mu$s): sensor output (blue) and applied signal (red). data.csv

Fig. 5

• panel (a): Comparison of electric field sensors: bandwidth (Hz) against sensitivity (V/m/Hz$^{1/2}$), by type (electrostatic force, quantum sensing, induced charge, LN, previous TFLN, this work), each point labelled with its reference, and dashed lines of constant bandwidth over sensitivity. The shaded group regions of the print are not drawn. data.csv

Cite

Xinyu Ma et al. (11 authors). Integrated microcavity electric field sensors using Pound-Drever-Hall detection. Nat. Commun. 15, 1386 (2024). https://doi.org/10.1038/s41467-024-45699-w

When you use hosted data, cite the original paper and give the panel's URL so a reader can find the exact values you used.

All papers · About · Statistics · Submit data · API