Fang’s InGaP platform holds promise for more efficient quantum devices

1/21/2025 Colin Robertson

Written by Colin Robertson

In the race to develop more feasible quantum computing devices, roadblocks include needed improvements to scalability, optimization, and error correction, as well as the expensive requirement to cool devices down to cryogenic temperatures.

Illinois quantum optics researcher Kejie Fang is pioneering a new nonlinear integrated photonics platform that he believes will provide improved efficiencies, not only for quantum devices, but also for applications in classical telecommunications.

In the Holonyak Micro and Nanotechnology Lab, Fang, an associate professor of electrical and computer engineering and Yuen T. Lo Faculty Fellow, is exploring the use of indium gallium phosphide (InGaP) to fabricate more efficient nonlinear photonic devices.

Nonlinear optical materials, including InGaP and many other III-V semiconductor compounds, enable interactions of photons that are important for quantum information processing and a variety of other applications. Usually, photon-photon interactions will not occur in a vacuum. However, when very high-intensity light, like a laser beam, strikes a nonlinear optical material, photons may be observed interacting with each other.

For example, spontaneous parametric down-conversion (SPDC) is a nonlinear optical effect that can generate entangled photons, while second-harmonic generation (SHG) can be used to double the frequency of light. If quantum devices were fabricated from more efficient materials, these and other nonlinear effects could be generated using lower-powered lasers (that is, ones that use fewer photons). That would bring researchers closer to the goal of enabling interactions between individual photons without requiring the use of cryogenically cooled atoms to mediate those interactions.

In recent studies published in Optica and in Light: Science & Applications, Fang and his coauthors have shown that their InGaP waveguides have efficiencies that are one to two orders of magnitude higher than those of the state-of-the-art thin-film periodically poled lithium niobate (LiNbO3) waveguides for nonlinear optical processes, including SHG and SPDC.

In a more recent arXiv preprint, Fang, ECE Illinois graduate students Joshua Akin and Yunlei Zhao, and Illinois Physics faculty Elizabeth A. Goldschmidt and Paul G. Kwait became the first to demonstrate faithful quantum teleportation involving spectrally distinct photons by using a nonlinear Bell state analyzer.

Quantum teleportation makes use of entangled photons to communicate quantum information between distant quantum nodes without requiring the original photon to traverse the distance.

The outcome of a Bell state measurement determines the quantum state of the teleported photon. Conventional linear-optical Bell state measurements are limited by multiphoton noise. Fang’s nonlinear Bell state analyzer uses an InGaP nanophotonic cavity with efficient sum-frequency generation (SFG) to filter multiphoton emissions, resulting in observed fidelities of greater than 94% down to the single-photon level without post-selection.

“Our demonstration is the first one that demonstrated this kind of a nonlinear optical SFG-heralded quantum teleportation at the single-photon level,” said Fang. However, he indicated that there are still improvements to be made. “I think we still need to work on if we can improve the nonlinear efficiency by another order of magnitude. Then I would claim this will be a very useful technology to replace current state-of-the-art.”

Challenges include refinements to InGaP’s fabrication process. As a semiconductor material, thin-film single-crystalline InGaP can be grown on a conventional gallium arsenide (GaAs) substrate. However, since the two materials have similar refractive indexes, InGaP devices need to be separated from the GaAs substrate to contain light. Fang’s group has been using a transfer-free wet etching process to achieve separation. However, developing a wafer-bonding transfer process to move InGaP devices to a low-index substrate would improve scalability and enable more applications that require high power.

Second, even the smallest nonuniformities can have a big impact on thin-film InGaP devices, like waveguides, which are usually only about 100 nanometers thick. “One or two percent thickness in nonuniformity could cause a significant drop of the nonlinear efficiency if you make larger devices,” said Fang. “Strong nonlinearity breaks down because this nonlinear effect depends on the so-called phase-matching condition of the waveguide. But if the thickness changes, then this phase-matching condition no longer holds.” Fang’s group is currently working on improving the uniformity of InGaP devices.

Despite these challenges, Fang is excited about future research directions and is already pursuing follow-up work. In his next set of planned experiments, he hopes to demonstrate quantum entanglement swapping, an extension of quantum teleportation, at the single-photon level using a nonlinear Bell state analyzer.

Looking further out, Fang wants to investigate whether nonlinear InGaP devices can be used for single-photon detection at room temperature. Current detection techniques require cryogenic temperatures. “I think that's going to be revolutionary if we really realize it,” he said.

For classical telecommunications, Fang said that InGaP could be used for optical amplifier amplification, affording improved size, weight, flexibility, and power efficiency over conventional erbium-doped fiber amplifiers (EDFAs). “The EDFA is benchtop-size; our platform is on chip, a microdevice,” said Fang. “EDFAs are fixed-bandwidth because EDFAs basically use atomic transitions.... The photonic system is very flexible. You can engineer bandwidth, engineer frequency.” While EDFAs use watts of laser pump power, Fang believes that an InGaP device would allow for very high-gain amplification using only a few milliwatts.

Fang became interested in quantum optics during his final year of doctoral studies at Stanford University. After finishing his PhD thesis on classical nano-optics in 2013, he held a postdoctoral research position at Caltech, working on quantum optics and optomechanics. At Illinois, Fang received the National Science Foundation’s CAREER Award for research in optomechanics.

Discussions with ECE Illinois colleagues helped Fang realize InGaP’s potential as a nonlinear optical material. “Illinois has a legacy in microelectronics,” said Fang. “I was in photonics. I was not doing III-V [semiconductor material], but then I started my research here, talking with colleagues in the department. And I realized, basically, that the III-V material has the strongest optical nonlinearity.”

“One of the strengths of Illinois is it has a very diverse community. We have a large group of faculty, and we work in different fields,” said Fang. “I can talk to people in our ECE department who work on materials. And then I talk frequently with colleagues in Physics, who work on quantum optics.”

Fang says that his involvement as a researcher in two national quantum centers, Hybrid Quantum Architectures and Networks (HQAN) at Illinois and Next Generation Quantum Science and Engineering (Q-NEXT) at Argonne National Laboratory, has helped his work by enabling frequent useful discussions and collaboration. The recently announced Illinois Quantum and Microelectronics Park in Chicago should open additional possibilities. “It will attract more companies to Illinois and attract a more talented workforce to Illinois to work on quantum technology and quantum information science,” said Fang. “I think it's also beneficial for faculty who have ideas [for] commercialization of their research results, including my group.”

“I will not say InGaP is the ultimate nonlinear photonics material. It’s possible, I will say, [that] AI can design material, find the materials. Maybe one day we can find a better material which has a much stronger nonlinearity than InGaP,” said Fang.

“Eventually, maybe we can show nonlinearity so strong that we no longer need to use a single atomic system to induce photon-photon interaction. Maybe we can just use these kinds of integrated nonlinear photonic devices to induce photon-photon interactions. I think that’s one of my overarching goals or hopes. Maybe one day we can realize this kind of technology.”


Share this story

This story was published January 21, 2025.