The new photonic chip creates a nested topological frequency comb

Scientists searching for compact and powerful sources of multi-color laser light have produced the first topological frequency comb. Their result, which relies on a tiny silicon nitride chip patterned with hundreds of microscopic rings, appears in the journal. Science.

Light from a typical laser emits a distinct, distinct color – or equivalently, a single frequency. A frequency comb is like a souped-up laser, but instead of emitting a single frequency of light, a frequency comb shines with many equally spaced pristine frequency spikes. The uniform spacing between spikes resembles the teeth of a comb, which gives the frequency comb its name.

The first frequency combs required bulky equipment to create. Recently, researchers have focused on miniaturizing them into integrated chip-based platforms. Despite great advances in miniaturizing the equipment needed to produce frequency combs, the basic ideas have not changed. Creating a useful frequency comb requires a stable light source and a way to scatter that light across the comb teeth, taking advantage of optical gain, loss, and other effects that appear when the light source is intensified.

In the new work, JQI Fellow Mohammad Hafezi, who is also Minta Martin Professor of Electrical and Computer Engineering and Physics at the University of Maryland (UMD), JQI Fellow Karthik Srinivasan, who is also a member of the National Institute of Standards and Technology. Tech and several colleagues have combined two lines of research into a new method for generating frequency combs.

One line is trying to minimize the generation of frequency combs by using microscopic resonant rings made of semiconductors. The second involves topological photonics, which uses patterns of repeating structures to create paths for light that are immune to small defects in the fabrication.

“The world of frequency combs in single-loop integrated systems is exploding,” says Chris Flower, a graduate student at JQI and the UMD Department of Physics and lead author of the new paper. Our idea was basically, can the same physics be realized in a special network of hundreds of coupled loops? This was a huge escalation in system complexity.

By designing a chip with hundreds of resonator rings arranged in a two-dimensional grid, Flower and his colleagues created a complex interference pattern that captures incoming laser light and circulates it around the edge of the chip while its material The chip has split it. At many frequencies

In the experiment, the researchers took snapshots of the light above the chip, showing that it was actually circulating around the edge. They also extracted some of the light to perform a high-resolution analysis of its frequencies, showing that the rotating light has a double-frequency comb structure. They found a comb with relatively wide teeth, nested inside each tooth and a smaller comb hidden.

Although this nested comb is currently only a proof of concept—its teeth aren’t perfectly uniform and are a little too noisy to be called pristine—the new device could eventually lead to smaller, more efficient frequency comb equipment. Used in atomic clocks, rangefinder detectors, quantum sensors, and many other tasks that require precise light measurement.

The specified spacing between spikes in an ideal frequency comb makes them an excellent tool for these measurements. Just as evenly spaced lines on a ruler provide a way to measure distance, the evenly spaced spikes of a frequency comb allow the measurement of unknown frequencies of light. Mixing one frequency comb with another light source produces a new signal that can reveal the frequencies in the second source.

Repetition begets repetition

Qualitatively at least, the repeating pattern of microscopic ring resonators on the new chip creates a pattern of frequency spikes that run around its edge.

Individually, the microrings form tiny cells that allow photons—quantum particles of light—to jump from ring to ring. The shape and size of the microloops are carefully chosen to create the right kind of interference between the different hopping paths, and the individual loops together form a superloop. Collectively, all the rings scatter the incoming light to multiple comb teeth and guide them along the edge of the grating.

The microrings and the larger superring provide two different time and length scales to the system, since light takes longer to travel around the larger superring than either of the smaller microrings. This eventually results in the production of two nested frequency combs: one is a coarse comb produced by smaller microrings, with widely spaced frequency spikes. Within each of these coarsely spaced spikes lives a finer comb produced by the ring cloud.

This comb-within-a-comb nesting structure, reminiscent of Russian nesting dolls, could be useful in applications that require precise measurement of two different frequencies that happen to be separated by a large gap, the authors say.

making things right

It took more than four years for the experiment to come together, a problem exacerbated by the fact that only one company in the world could make the chips the team designed.

Early chip prototypes had microrings that were too thick and too sharp bends. When the incoming light passes through these loops, it scatters in a variety of unwanted ways, defeating any hope of producing a frequency comb.

“The first generation of chips didn’t work at all because of this,” says Flower. Returning to the design, it reduced the width of the ring and rounded the corners, finally ending up on the third generation of chips, which will be delivered in mid-2022.

While iterating on the chip design, Flower and his colleagues realized that it was difficult to deliver enough laser power to the chip. For their chip to work, the intensity of the incoming light had to exceed a certain threshold—otherwise none of the frequency combs would form.

Typically, the team must obtain a commercial CW laser that provides a continuous beam of light. But those lasers overheat the chips, causing them to burn or swell and misalign with the light source. To deal with these thermal issues, the team needed to focus energy in bursts, so they turned to pulsed lasers, which deliver their energy in fractions of a second.

But this presented its own problems: off-the-shelf pulsed lasers had pulses that were too short and had too many frequencies. They tended to introduce a bunch of unwanted light—both at the edge of the chip and through the middle—instead of the specific edge-limited light that the chip was designed to scatter in a frequency comb. Given the time and cost involved in developing new chips, the team had to make sure they found a laser that balanced maximum power with longer, tunable pulses.

“I emailed basically every laser company,” Flower says. “I searched to find someone to build me a tunable laser with a long pulse length. Most people said [that] they [didn’t] Build it, and they are too busy to do custom lasers. But a company in France answered me and said, “We can do it.” Let’s talk.”

His persistence paid off, and after several shipments from France to install a more powerful cooling system for the new laser, the team finally sent the right kind of light into their chip and saw a nested frequency comb emerge.

The team says that while their experiment is specific to a chip made of silicon nitride, the design can easily be translated to other photonic materials that can create combs in different frequency bands. They also see their chip as introducing a new platform for the study of topological photonics, particularly in applications where there is a threshold between relatively predictable behavior and more complex effects – such as the generation of a frequency comb.