Combining one and two photon polymerization for accelerated high performance (3 + 1)D photonic integration.

Area of Science:

• Optoelectronics and Photonics
• Materials Science and Engineering
• Additive Manufacturing

Background:

• High-performance electronic integration faces limitations due to 2D lithography and thermal dissipation challenges.
• Three-dimensional (3D) photonic integration is crucial for scalable, application-specific integrated circuits, particularly for neural networks.
• Direct laser writing, specifically two-photon polymerization (TPP), is a promising CMOS-compatible technique for high-resolution 3D photonic integration.

Purpose of the Study:

• To develop a novel, accelerated fabrication process for 3D photonic integration by combining one-photon polymerization (OPP) and TPP.
• To enhance optical confinement and reduce fabrication time for 3D photonic circuits.
• To demonstrate the performance and stability of waveguides fabricated using the new method.

Main Methods:

• Introduced a hybrid (3 + 1)D "flash"-TPP technique combining femtosecond TPP for waveguide cores and support structures with UV-initiated OPP for passive volumes.
• Optimized voxel spacing during TPP for smooth interfaces (cores) and rapid printing (supports).
• Utilized UV blanket irradiation for rapid polymerization of the entire chip's passive volume, decoupling fabrication time from volume size.

Main Results:

• Successfully fabricated vertical single-mode waveguides with lengths up to 6 mm and numerical apertures (NA) of 0.16.
• Achieved low injection losses (-0.26 dB) and propagation losses (-1.36 dB/mm at 660 nm), comparable to silicon photonics.
• Demonstrated long-term optical performance stability (>3000 h post-printing) and operational stability (600 h continuous use).

Conclusions:

• The combined OPP and TPP approach significantly accelerates 3D photonic circuit fabrication while improving optical confinement.
• The fabricated waveguides exhibit high performance metrics and excellent long-term stability, suitable for advanced photonic applications.
• This method offers a scalable and efficient pathway for realizing complex 3D integrated photonic devices for emerging technologies like neural networks.