Harvard Kicks Photonics Up a Notch With First Integrated Laser on TFLN Chip
Hoping to keep pushing photonic technology further, researchers at Harvard University have successfully integrated laser onto thin-film lithium niobate (TFLN) chips.
Researchers are constantly working to develop more power and cost-efficient optical sources to support applications such as telecommunications networks and optical interconnects. To this end, one technology that has recently gained popularity is TFLN photonics.
Unfortunately, a glaring roadblock toward TFLN has been an inability to integrate laser sources directly onto the chip—resulting in increased costs, stability, and a failure to scale. Looking to solve this, last week, researchers at Harvard University announced the successful design of the world’s first laser-integrated TFLN chip.
Illustration of Havard researcher's approach to integrating a laser onto a TFLN chip. Image used courtesy of Shams-Ansari et al
This article will discuss TFLN technology, its benefits, challenges, and the new research coming out of Harvard.
Lithium Niobate Technology
When supporting optical communication networks, an "ideal" transmitter offers a large bandwidth, small driving amplitude, high output power, low insertion loss, and low cost. A simultaneous combination of these feats can be challenging, but researchers have recently found an increased interest in TFLN to meet these needs.
A high-level view of LN integrated photonics. Image used courtesy of Zhu et al
Lithium niobate (LN) has been a considerably popular material in photonics for decades thanks to a combination of unique electrical, nonlinear, and acousto-optic properties. These properties include a broad transparency window, a high refractive index, and a high Curie temperature.
Thanks to these properties, LN has been a foundational building block of high-speed optical communications, including optical wavelength conversion, waveguides, photon-pair generation, and electro-optic modulators.
TFLN hopes to take LN to the next level, as the thin films provide higher intensity per unit of power in waveguides, which results in a stronger nonlinear optical effect and shorter interaction length. For this reason, TFLN is an up-and-coming technology that could enable even smaller and more power-efficient optical transmitters in the future.
Challenges with Laser Integration
Although LN has had tremendous application, and TFLN further improves the potential, the technology has been hindered by difficulties with material integration and processing.
Traditionally, integrated LN devices have required low-index-contrast waveguides, which have the pitfall of a large footprint and reduced nonlinear efficiencies. The result of this is that LN and TFLN technologies need the coupling of multiple discrete components
(i.e., laser and modulator) to achieve a complete solution.
Thus, LN technology lags behind competing for integrated photonics, which may have worse material properties but can be integrated and scaled for power efficiency, size, cost, and complexity.
The ultimate solution for optical communications would be a fully integrated solution nowhere high power, low noise lasers exist on the same chip as TFLN photonic platform.
First Laser-integrated TFLN Chip
Despite the challenges with TFLN technology, last week, researchers from Harvard University made news when they announced the successful development of a fully integrated laser-TFLN chip.
In their paper published in Optica, the researchers described their successful method, which merged distributed feedback (DFB) laser technology with TFLN photonics.
The researchers coupled the DFB and TFLN waveguides through a flip-chip bonding process with this approach. The result was that the DFB lasers could sit in trenches in the LN and deliver up to 60 mW of power to the waveguides.
Combining the laser with a 50 GHz electro-optic modulator in the LN, the researchers effectively built a high-power transmitter.
The laser-integrated TFLN fabrication process. Image used courtesy of Shams-Ansari et al
Importantly, the researchers claim that their approach opens the door to large-scale, low-cost arrays of these transmitters by using an industry-friendly process. These arrays would serve as the building blocks of larger optical systems, going beyond communications to encompass range detection solutions like LiDAR.
Moving forward, the team plans to work towards improving the laser's power and scalability to fit even more applications.