Teri Odom picture

Teri Odom. [Image: Northwestern University/Matthew Gildon]

For its May 2020 print article “The Laser at 60,” OPN interviewed a range of OSA Fellows to get their insights on some particularly interesting horizons in laser research today. We’re presenting a selection of those interviews online. Below is an edited version of our interview with Teri Odom, a professor of chemistry and materials science at Northwestern University, USA. Odom focuses on nanophotonic and plasmonic systems, where light and matter interact at subwavelength scales.

What makes a laser a nanolaser?

Part of it is that the operational principle—for example, the feedback volume and mechanism—is subwavelength in scale, meaning it can beat the diffraction limit. So, while typical lasers have cavity lengths that are on the order of a micron or larger, a nanolaser has at least one dimension—their cavity size or effective volume for the mechanism of feedback—on the nanometer scale.

How do those length scales compare, for example, with the kind of things that are being built for silicon photonics—the laser integrations in that sphere?

Silicon photonics will be limited to the diffraction limit—meaning the smallest ones are on the order of micron or so, maybe a little bit less than that. Even for vertical cavity surface emitting lasers (VCSELs), they still have cavity sizes that are larger than hundreds of nanometers.

How did you first get interested or involved in this nanolaser area?

In general, when I started in this area several years ago, we were mostly just interested in light–matter interactions at the nanoscale—what interesting phenomena can we discover, and if those phenomena are reproducible, then what is possible related to applications?

So, in the beginning, our goal wasn’t to say, Oh, let’s just make a nanolaser because it sounds cool. But it was more about, Oh, let’s see how excitation can be transferred between a gain medium such as dye molecules, or quantum dots, or any other type of emitter, with the localized (plasmonic) fields supported by metal nanoparticles. Noble-metal nanostructures are, of course, special since one of their intrinsic properties is extreme light localization.

That’s mostly how it started—how can we squeeze the light down into a very tiny volume? And then, if we have an emitter there, can the properties of the emitter be amplified or manipulated? That was the setup for our scope of work. And then, we just performed the normal type of characterization, where you would excite the dye, have its energy transfer to the plasmon, and see what happens. It turned out that the interesting characteristics we observed, especially at higher and higher pump powers, were reminiscent of macroscale lasers, with a very clear threshold and directional emission.

So those are the sorts of things you look for to see if something really is a nanolaser.

Actually, a number of years ago, Nature Photonics published a whole article on what must be included if you try to submit a laser paper to that journal. Apparently, everybody was just thinking, We see linewidth narrowing and threshold behavior at higher pump powers, and so nanolasers are everywhere. And that just wasn’t the case.

Now, these are just part of a checklist that’s required if you believe lasing is a central result of the manuscript. I am a fan of high and uniform standards—I think the early results in this area were like the “Wild West”; researchers had some really interesting systems, but the community didn’t really understand the fundamental mechanisms. Having these kinds of guidelines in place actually made my and my group’s lives easier since we now have a straightforward way to make the case for our very interesting lattice architectures that show coherent emission.

Late last year, you published a study in Nature Materials involving a nanolaser created with upconverting nanoparticles. Could you talk about how that system worked?

OK. First of all, upconversion nanoparticles have promise for bioimaging, because in these inorganic, transparent nanomaterials, excitation at longer wavelengths [which show greater penetration in tissue] produces shorter wavelength emission. That’s why the process is called upconversion—in most optical materials, excitation at shorter wavelengths results in longer-wavelength emission, so it’s an inverted process.

But what I thought was really interesting about that—related to whether we could make them into nanolasers with our nanoparticle array or lattice platform—was that in the plasmonics community, we’ve never been able to make continuous-wave lasers, and lasers with low threshold. Typically, high pump powers are needed in order to achieve lasing, partly because of the losses for some plasmonic designs based on metal films. And other times, the device can’t support the continuous-wave pump excitation because of heating and degradation.

But in this particular system, everything just came together really, really nicely. The host material of the upconversion nanoparticles is very similar to glass, which has a refractive index that’s about the same as the substrate supporting the plasmonic nanoparticle lattice. So, the entire solid-state device was very thin, and the design characteristics for creating high-quality nanocavities that could produce good feedback for lasing was possible.

We were very happy about this—the size of the laser device in the vertical direction was about 100 nanometers or so. And then, there was a high concentration of emitters packed in a single upconversion nanoparticle. In a 10-nanometer bead, hundreds of potential emitters could be used as gain for this system.

So, this was a really nice system that could produce continuous-wave lasing emission [or even pulsed emission to generate continuous-wave emission] because the lifetimes of these upconversion nanoparticles are very long. The whole system is just pretty cool.

And the other thing that’s interesting about this system is that the record low threshold under continuous-wave pumping at room temperature is about one hundred times lower than commercialized laser diodes. I think this metric is significant—even as we push on the fundamentals, we’re actually achieving technological benchmarks that indicate that these types of systems, which are fundamentally interesting in their own right, can potentially be used in applications. They perform well, head to head, with metrics that the community wants.

At the end of the paper on this nanolaser, you say that it offers prospects for previously unrealizable applications of coherent nanoscale light. Could you just go through a couple of those?

The one that’s received the most attention—partly because there’s just no current equivalent—is the ability to integrate this type of light source with tissue. Most of the development in upconversion nanoparticles has been for deep-tissue imaging, meaning you can use long-wavelength light to create excitation in the visible wherever the nanoparticles are, and the visible light can resolve biological structures. But, there hasn’t been a way to control the direction of emission from these particles or to select out a specific wavelength from the emission manifold.

And so, in this particular design, we can control the wavelength as well as the emission direction. And coherent light is much better for many types of excitation processes than incoherent light. So I think the ability to use them as light sources within tissue provides some opportunities, not just in terms of some therapeutic applications, but for imaging, and to resolve, for example, the structure or margins of tumors, or differences of refractive index of environments, which is often representative of different densities of tissue.

Our system also should be able to realize higher resolution. And, since the device is effectively all glass, you don’t have to work extra hard to make it biocompatible because it already is.

From another perspective, the ability to just have low-threshold, coherent light sources is needed for understanding other types of processes. How can this coherent light be applied to systems that you don’t want to disturb? As an example, often we want quantum systems to preserve their coherence at room temperature—and how can we access coherent light of a specific wavelength, and can we use it to design single quantum emitters or single photon sources? Quantum materials are often delicate, and so you don’t want to blast them with high-intensity laser pulses.

There’s a range of different possibilities for low-threshold coherent light sources—from imaging to potential quantum manipulation of emitters. It just depends on what the intended target is. The system we’ve designed is flexible, which is one of the best things about it.

A last question. This is the 60th anniversary of the first demonstration of a working laser, by Theodore Maiman. Do you have any thoughts on how far the laser has come since then? Anything about that progress that particularly amazes you?

Oh, yes. One of the things I do when I’m trying to introduce the concepts of a laser to students is to go back to Gordon Gould’s lab notebook, where the ideas are sort of sketched out—what would a cavity look like? What are the implications for amplification and emission? The origin story, I think, is quite fascinating.

And some older colleagues remember the time when the laser was called “a solution looking for a problem.” But now, of course, it’s ubiquitous, and for a range of applications—from scanners in the supermarket, to your smartphone identifying facial recognition, to even interferometry information from LIGO.

I think lasers—and especially nanolasers—are pretty neat. Even after 60 years, which is not that long, this device and concept have widened the space where many different disciplines can contribute both to new fundamental knowledge and potential applications.