Compact and high-resolution spectrometer via Brillouin integrated circuits

Abstract

Optical spectrometers are indispensable tools across various fields, from chemical and biological sensing to astronomical observations and quantum technologies. However, the integration of spectrometers onto photonic chips has been hindered by the low spectral resolution or large device footprint with complex multiple channel operations. Here, we introduce a novel chip-integrated spectrometer by leveraging the acoustically-stimulated Brillouin scattering in a hybrid photonic-phononic chip. The Brillouin interaction provides a dynamic reflection grating with a high reflectivity up to 50% and a fast switching time on the microsecond scale, achieving an unprecedented spectral resolution of 0.56 nm over a 110 nm bandwidth using just a single 1 mm-long straight waveguide. This remarkable performance approaches the fundamental limit of resolution for a given device size, validating the potential of the hybrid photonic-phononic device for efficient and dynamically-reconfigurable spectral analysis, and thus opens up new avenues for advanced optical signal processing and sensing applications.

Figures (4)

• Figure 1: Schematic of on-chip spectrometer.a-b, Conventional on-chip spectrometers based on linear processes. a, A static planar grating enables dispersive optics to separate light with different wavelengths. b, Wavelength of the input light can be derived through a Fourier-transform approach by dynamically tuning a microring resonator or a Mach-Zehnder interferometer. c, The proposed spectrometer is based on nonlinear process that employs the phase-matching condition of acoustically-stimulated Brillouin scattering to achieve wavelength-dependent optical reflection. The input laser ($\omega_{1}$) and phonon drive ($\Omega$) propagate in opposite directions, and the reflected light ($\omega_{2}$) satisfies both momentum and energy conservations.
• Figure 2: Characterization of the on-chip spectrometer.a-b, Experimental setup for characterizing the spectrometer. The phononic mode in the waveguide is driven by a vector network analyzer (VNA) through an interdigital transducer (IDT, d). The optical input signal is reflected through the Brillouin process when the phase-matching condition is fulfilled. The reflected light is measured by the VNA through a heterodyne detection or directly detected by the photodiode (PD) under high signal-to-noise ratio conditions. The Brillouin interaction region is around $1\,\mathrm{mm}$. EDFA: erbium-doped fiber amplifier, PC: polarization controller. c and g, Displacement and electrical field distributions at the waveguide cross-section for the quasi-Love phononic mode and the fundamental transverse electric photonic mode, respectively. d and e, Scanning electron microscope pictures of the IDT and phononic-photonic mode multiplexing device. f, S11 parameter of IDT which exhibits a resonance frequency of 8.5 GHz. h, Measured reflected optical signal against the input RF frequency, with the input optical wavelength fixed at $1576\,\mathrm{nm}$.
• Figure 3: Broadband spectrum characterization.a, Separate spectral line response with different pump wavelengths, with an RF drive power is $20\,\mathrm{mW}$. Inset: the relationship between the reflected optical power and the input power; Dispersion relationship between the beat frequency of the output light and the input optical wavelength. The slope is $6.03\,\mathrm{MHz/nm}$. b, Top: RF response spectrum with two pump lasers at $1550.5\,\mathrm{nm}$ and $1550\,\mathrm{nm}$. The red line represents a double-Lorentz fitting for the measured data, which is the sum of two Lorentz distributions displayed in yellow. Bottom: Reconstructed optical spectrum. c, We use a standard commercial dense wavelength division multiplexing (DWDM) as a filter to generate the spectrum by tuning the spontaneous radiation of our EDFA without any laser input. The blue line represents the output signal from our device detected by a PD directly and the orange line comes from a commercial spectrometer.
• Figure 4: Spectrometer intrinsic resolution.a, A plot comparing the resolution, detection channel, and footprint scale for selected device demonstrations in the literature. The footprint scale includes the elements in the device that are active in resolving and detecting light, excluding the accessory components. Reference numbers are indicated within square brackets. b, Simulation results of the device resolution performance improved by the interaction length ($L$). The red line represents the result based on the current material and fabrication level with a phonon Q $\sim 3000$. The blue line shows the resolution with an improvement of phonon Q $\sim 20000$ in future experiments. The dots represent the simulation results with the impact from the phase-matching diffusion and the shadow part represents the standard deviation of the resolution. The standard deviation of the random diffusion value added in simulation is about $1500\,\mathrm{rad/m}$, which represents the standard derivation of waveguide width ($\Delta w$) around 4 nm and film thickness ($\Delta t$) around 0.5 nm. c, The heatmap represents the relationship between mismatching value $\Delta\beta$ and variation error on $\Delta w$ and $\Delta t$. The cyan line corresponds to the diffusion value in b.