🔬 applied physics

Numerical Investigation of a 3D End-Firing Antenna Array Based on Two-Photon Polymerization on Thin-Film Lithium Niobate for Optical Beam Steering

This paper presents a numerically investigated 3D end-firing antenna array fabricated via two-photon polymerization on thin-film lithium niobate, which achieves high transmission efficiency and wide two-dimensional beam steering capabilities for solid-state optical phased arrays.

Original authors: David Trop, Boris Desiatov

Published 2026-07-21

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: David Trop, Boris Desiatov

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine trying to steer a giant, invisible flashlight beam without ever moving a single part of the flashlight itself. This is the magic of Optical Phased Arrays (OPAs). Think of an OPA not as a single mirror, but as a massive choir of tiny light emitters. If you ask every singer in the choir to start their note at exactly the same time, the sound (or light) shoots straight ahead. But if you tell the singers on the left to start a tiny fraction of a second earlier than those on the right, the whole beam of sound or light tilts to the right. By tweaking the timing of these tiny emitters, you can point a laser beam anywhere in the sky instantly, with no moving gears or heavy mirrors. This technology is the holy grail for things like self-driving car sensors (LiDAR) and super-fast internet, because it's small, solid, and incredibly fast. However, building these "choirs" of light is tricky. Usually, you have to choose between making them wide enough to see a lot of sky or making them fast enough to track a speeding car, but rarely both.

This paper introduces a clever new way to build these light choirs by mixing two very different worlds: a super-fast electronic material called Thin-Film Lithium Niobate (TFLN) and a 3D-printed polymer structure. The researchers, David Trop and Boris Desiatov, wanted to solve a specific headache: traditional light antennas are usually flat and stuck on the edge of a chip, which limits them to steering in just one direction (like a lighthouse sweeping left and right). To see up and down and left and right, you usually need complex, flat grids that are hard to make and lose a lot of light. The authors asked: "What if we could lift the antennas up into the air, like 3D towers, so we can pack them tightly in a grid?"

To do this, they used a technique called Two-Photon Polymerization (2PP). Imagine this as a high-tech 3D printer that can draw tiny, invisible bridges out of a special liquid plastic directly onto the surface of a silicon chip. The team designed a system where light travels through a flat waveguide on the chip, then gets handed off to these 3D-printed polymer "towers" that rise up into the air. These towers act as the emitters. Because they are lifted up, they can be arranged in a dense, two-dimensional grid (a 25 by 25 array) without getting in each other's way.

The results from their computer simulations are quite promising. They found that this hybrid design is incredibly efficient. In their digital tests, the system managed to send 89.5% of the light through the transition from the chip to the 3D antenna, which is a very high score for such a complex handoff. The system worked well across a broad range of colors (wavelengths) from 1.4 to 1.6 µm, which is the standard "telecom" range used for internet and sensing.

When they simulated steering the beam, the results were impressive. The array could sweep a wide area, covering a field of view of 24.9° × 22.8°. To put that in perspective, that's a wide enough angle to scan a large portion of the sky without moving the device. The beam itself was sharp, with a width of about 1.7°, meaning it could point precisely at a specific target. Crucially, because the system uses the Lithium Niobate material, it can change the beam's direction at lightning speed—potentially in nanoseconds—far faster than older methods that rely on heat.

The paper explicitly rules out the idea that you need to sacrifice speed for range or that you are stuck with flat, one-dimensional arrays. By lifting the antennas, they kept the "end-fire" advantage (where light shoots out the end of a tube efficiently) while gaining the ability to steer in two dimensions. However, it is important to note that these findings are currently based on rigorous numerical simulations. The authors have not yet built the full 25x25 array in a lab to prove these numbers in the real world, though they did successfully print a small 2x2 version to prove the 3D printing technique works. They suggest that this approach could be a major step toward making compact, high-speed LiDAR and optical communication systems, bridging the gap between tiny chip-based optics and the free space where we need to send light.

Technical Summary: Numerical Investigation of a 3D End-Firing Antenna Array Based on Two-Photon Polymerization on Thin-Film Lithium Niobate

Problem Statement
Optical phased arrays (OPAs) are critical for solid-state beam steering in applications such as LiDAR, optical communications, and adaptive optics. However, conventional integrated OPA designs face inherent trade-offs between bandwidth, steering range, and fabrication complexity. Traditional grating couplers facilitate 2D arrangements but are wavelength-sensitive and require large element pitches, limiting steering range. Conversely, end-fire antennas offer broadband operation and tight element spacing for superior beam quality but are historically confined to 1D arrays due to their planar, edge-emission geometry. Furthermore, existing material platforms like silicon photonics often rely on thermo-optic or free-carrier effects, which suffer from slow response times or high optical loss, respectively. There is a need for a platform that combines high-speed phase modulation, low loss, and the ability to realize true 2D end-fire antenna arrays.

Methodology
The authors present a novel hybrid architecture integrating a thin-film lithium niobate (TFLN) photonic platform with three-dimensional (3D) polymer waveguide antennas fabricated via two-photon polymerization (2PP). The study relies on rigorous numerical simulations, specifically full-wave three-dimensional finite-difference time-domain (FDTD) methods, to optimize and analyze the system.

The device design consists of four primary subsystems:

Waveguide Platform: A fully etched TFLN structure (300 nm layer) was selected over ridge waveguides to provide stronger optical mode confinement, facilitating efficient coupling to elevated structures.
Power Distribution: A network of Y-splitters distributes light from a single input to a 25 × 25 antenna array. This design was chosen for its inherent wavelength insensitivity and geometric simplicity across the 1.4–1.6 µm band.
Phase Control: Electro-optic (EO) phase shifters utilizing the Pockels effect ( $r_{33} \approx 31$ pm/V) are embedded in the TFLN waveguides. Simulations optimized the electrode gap (4–5 µm) to balance modulation efficiency ( $V_\pi L$ ) with optical propagation loss, enabling nanosecond-scale response times.
3D Antenna Architecture: The core innovation involves 3D polymer antennas (using IP-n62 polymer) printed directly atop the TFLN chip. The antenna geometry includes an inverse taper for coupling from the TFLN waveguide, an S-curved elevation section to lift the mode above the substrate, an adiabatic taper to reduce the cross-section, and a radiating aperture. The design was optimized for Transverse Electric (TE) polarization.

The simulation methodology employed a two-step approach for array analysis: first, simulating the single 3D antenna to extract the near-field profile; second, replicating this profile in a 2D array configuration with independent phase control to calculate far-field radiation patterns.

Key Contributions

Hybrid 3D Integration: The work introduces a method to overcome the 1D limitation of conventional end-fire antennas by elevating polymer structures above the chip surface using 2PP, enabling dense 2D array arrangements while preserving broadband end-fire characteristics.
Platform Synergy: This is the first demonstration (in this context) of integrating 3D polymer antenna architectures directly onto the TFLN platform. This combination leverages the exceptional electro-optic properties of lithium niobate for high-speed modulation, a capability not available in the silicon or bulk glass platforms used in prior 3D OPA demonstrations.
Optimized Antenna Design: The authors optimized the transition from the TFLN waveguide to the elevated polymer antenna, achieving high transmission efficiency while managing mode confinement and minimizing reflection.

Results
Numerical simulations yielded the following performance metrics:

Transmission Efficiency: The single-antenna design achieved a transmission efficiency of up to 89.5% across the 1.4–1.6 µm wavelength range, with less than 5% spectral variation. The primary loss mechanism was identified as the mode transition from the TFLN core to the polymer structure.
Beam Steering Performance: A 25 × 25 array with a 2 µm element pitch demonstrated a field of view (FOV) of 24.9° × 22.8°. The beamwidth was approximately 1.7° × 1.7°.
Steering Accuracy: The steering angle followed the theoretical relation $\theta = \sin^{-1}(\lambda \Delta\phi / 2\pi d)$ with deviations below 0.3°. The main-lobe intensity decreased by less than 3% across the steering range (0° to ±24.86°).
Broadband Operation: The system maintained performance across the telecom C-band (1.45–1.6 µm), with FOV varying linearly from 22.4° × 20.9° to 25.6° × 23.3°.
Grating Lobes: At the 2 µm pitch, grating lobes were pushed to ±50.70°, ensuring a grating-lobe-free steering range within the demonstrated FOV. In contrast, a 6 µm pitch resulted in grating lobes appearing within the visible range even at zero phase shift.
Loss Budget: The total insertion loss for the 25 × 25 array, including the Y-splitter network, was estimated at approximately 2.7 dB.

Significance and Claims
The authors claim that this work establishes a foundation for scalable 3D photonic phased arrays that bridge the gap between integrated optics and free-space beam manipulation. By combining the high-speed electro-optic modulation of TFLN with the geometric flexibility of 2PP-fabricated 3D antennas, the proposed architecture offers a solution for high-speed, low-loss beam steering systems.

The paper positions this design as particularly suitable for applications requiring rapid response, such as high-speed LiDAR and free-space optical communications, where the nanosecond-scale response of the EO phase shifters significantly outperforms thermo-optic alternatives. The authors note that while the current work is a numerical investigation, preliminary experimental trials have successfully fabricated 2 × 2 polymer antenna arrays on TFLN chips, validating the feasibility of the 2PP fabrication route. The study concludes that this hybrid approach provides a practical pathway toward compact, efficient, and reconfigurable optical beam steering systems for future photonic technologies.

More like this