top of page
Writer's pictureLatitude Design Systems

Millimeter-Long Antennas for Precise Environmental Scanning with Optical Phased Arrays

Introduction

The increasing demand for high-precision optical sensors, such as those used in Light Detection and Ranging (LiDAR) systems for autonomous vehicles, has driven the development of advanced beam-scanning technologies. Silicon photonics platforms have enabled the integration of non-mechanical beam-steering solutions through optical phased arrays (OPAs). OPAs rely on an array of optical emitters, or antennas, that couple light from the chip into free space. By controlling the emission from these antennas, precise scanning of the environment can be achieved.

Diffraction Gratings as Optical Antennas

One well-established approach to controlling the coupling of light from a chip is through the use of diffraction gratings. These periodic structures modulate the refractive index, allowing the emission from an incident wave to be directed in specific directions governed by the Bragg condition. The theory of diffraction dictates that the divergence of the diffracted beam is inversely proportional to the aperture size, or effective length, of the emitter. The radiation profile divergence (FWHM) of a single antenna can be expressed as:

Δ𝜃FWHM = 0.886𝜆 / (Lcos(𝜃))

Where L is the length of the grating, 𝜃 is the diffraction angle (in radians), and 𝜆 is the operating wavelength.

Scheme of designed antennas
Fig. 1 – Scheme of designed antennas (a) BIC-effect (Bound state In the Continuum) corrugated antenna, (b) Dual-layer Si-SiN antenna

Precise obstacle detection requires focused beams, necessitating weak, long diffraction gratings based on low index contrast. To address this need, two architectures of millimeter-long single waveguide grating antennas (SWGAs) were designed and fabricated to achieve a divergence lower than 0.08° at 1550 nm wavelength.

BIC-Effect Corrugated Antenna

The first design, shown in Fig. 1(a), relies on laterally corrugated waveguides, taking advantage of the Bound state In the Continuum (BIC) effect. These corrugations create periodic lateral narrowing, which can be fabricated in a single lithography step. However, traditional SWGAs suffer from strong diffraction strength, limiting the propagation of input light to short distances and resulting in high far-field divergence.

FDTD simulations predict an optimal decay length (Ld) of 250 μm for wc = 580 nm and 70 nm corrugation depth
Fig. 2 – (a) Simulated decay lengths at different widths of a 70nm-corrugated waveguide (insets: transversal cut of the E-field profiles) (b) Simulated and measured decay lengths as a function of the corrugation depth at the BIC-effect ([4] previous implemented standard corrugated SWGA and *lithography simulation with rounding of corrugations)

The BIC-effect principle enables the realization of longer SWGAs with low divergence by exploiting a critical waveguide width (wc) at which the extraction strength is minimized due to destructive intrawaveguide interferences between laterally diffracted fields. FDTD simulations predict an optimal decay length (Ld) of 250 μm for wc = 580 nm and 70 nm corrugation depth, as shown in Fig. 2(b*). However, the vertical directivity remains around 50/50 up-down, leading to potential blind spots due to destructive interference between upward emission and downward power reflected by the substrate.

Dual-Layer Antenna

To improve upward directivity, the vertical symmetry of the design must be broken. This is achieved by introducing a second index modulation layer, as depicted in Fig. 1(b). A 360 nm-thick SiN guide layer is placed above the Si guide layer, and a fully-etched SiN grating is periodically structured atop the standard Si waveguide. The low index contrast offered by this dual-layer configuration enables the realization of very weak and long antennas, with simulated decay lengths of several millimeters (Fig. 3(a)) and directivity up to 80% upward.

Simulated and measured decay lengths as a function of the Si waveguide width
Fig. 3 – (a) Simulated and measured decay lengths as a function of the Si waveguide width, (b) transversal and (c) longitudinal cross-sections of the fabricated antennas (TEM images)
Fabrication and Characterization

Both BIC-effect and dual-layer SWGAs were fabricated on a 300 mm SOI industrial platform, with periods calculated to address an 11.5° emission angle in air. The extracted decay lengths for BIC-effect SWGAs were longer than predicted, likely due to process-induced rounding of the corrugations. For dual-layer SWGAs, the measured decay lengths demonstrated light extraction over several millimeters, consistent with simulations.

Farfield of single 1mm-long
Fig. 4 – Farfield of single 1mm-long (a) BIC-effect (540nm-large) and (b) dual-layer (350nm-large) SWGAs and extracted average longitudinal divergences.

As expected, the far-field radiation pattern of the BIC-effect SWGA exhibited blind spots due to the 50% upward directivity, while the Si-SiN SWGAs offered a nearly homogeneous emission over the full field of view. The measured longitudinal divergences for 1 mm-long SWGAs were very low, with 0.18° for the 540 nm-large BIC antenna and 0.11° for the 350 nm-large dual-layer antenna. However, these divergences corresponded to emission lengths shorter than the measured decay lengths, likely due to the non-apodized nature of the gratings and potential loss of coherence along the very long antennas.

Conclusion and Perspectives

The results confirm the potential of BIC-effect and dual-layer Si-SiN SWGAs to function as millimeter-long antennas, enabling precise scanning of the environment for OPA-based sensing applications. Future work will involve integrating these antennas into active OPAs with carrier-depletion phase control and designing apodized gratings to further optimize the performance and reduce sensitivity to minor fabrication deviations.

Reference

[1] L.-E. Bataille, S. Guerber, S. Crémer, B. Dagens, and F. Boeuf, "Millimeter-long Antennas for Optical-Phased Arrays," STMicroelectronics, Crolles, France; CEA Leti, Grenoble, France; Université Paris-Saclay, CNRS, Centre de Nanosciences et de Nanotechnologies, Palaiseau, France, 2024.

Comentarios


bottom of page