From Photonics Media, this is All Things Photonics, a podcast shedding light on industry-shaping advancements in lasers, optics, imaging, metrology, and sensing. Each episode our hosts speak with luminaries, rising stars, and market leaders about the trends shaping the current and future landscape of photonics. From space and deep sea exploration to precision timekeeping and advanced manufacturing, this is the podcast for progress on all things photonic. This season of All Things Photonics is brought to you by PolyScience and Thorlabs.
Today's episode is brought to you by PolyScience. For over 60 years, the PolyScience line of award-winning temperature control equipment has been helping the photonics industry reach its goals safely and efficiently. Our products are made in America and delivered globally. Please visit polyscience.com to learn how we can help you. Changing the world of chillers is more than a tagline, it's a part of our history.
Today's episode is also brought to you by Thorlabs. Thorlabs believes in the positive impact of photonics and hopes today's discussion inspires you. We're a vertically integrated photonics designer and manufacturer dedicated to making your ideas successful. To join our team or to learn more about our stocked offering or custom solutions, visit thorlabs.com.
Last fall, the time was ripe for an in-depth examination of what recently imposed export control measures on gallium and germanium would mean for the optics supply chain. Both materials are needed to make alloys and compounds that are essential to the manufacture of semiconductor chips, optics, laser diodes, and other common components. China's directives mandated that its domestic exporters of these metals must now obtain a license to distribute them across international borders. The mandates raised both intrigue and concern. Photonics Spectra's coverage benefited from frequent dialogue with Sam Rubin, CEO of optics company LightPath Technologies. LightPath is a leading developer of Black Diamond, or BD, chalcogenide glasses, which can flourish as substitutions for germanium glass in a host of infrared and other applications. But one-for-one material swaps are complicated, and not just because of the distinctions that one material holds over another. Necessary qualifications, approvals, and even the familiar theme of moving away from what's common are all factors that are in play. For today's conversation, we welcome Sam Rubin to the podcast to discuss these considerations and many more. LightPath earlier this fall announced that it was selected to receive three and a half million to supply infrared optics for the F-35 combat aircraft program. Rubin, in our conversation, discusses the capabilities that a germanium-free optic offers an end user and how a legacy material may not always fit the bill. Our conversation with Sam Rubin is coming up today on All Things Photonics. But first, I speak with Leave and Penck, co-founder of design and simulation software company PlanOpSim. The company's solution is a streamlined single workflow design software for planar and meta optics, and Penck offers insights on the impetuses and requirements from industry and from research for such an offering. The story here is that meta-optical technologies are continuously expanding, and many more new application areas are requiring optical designers to move into uncharted territory. How a software design solution addresses that reality means fostering an understanding of new applications and their underlying factors. And with his unique qualifications to answer such a question, we ask Penck to shed light on something that may be overlooked in metasurfaces technology. All Things Photonics rolls ahead into November with Leave and Penck from PlanOpSim, followed by Sam Rubin from LightPath Technologies. Both are up next.
So I want to begin by talking about PlanOpSim, but I want to take sort of an indirect route to get there, and I want to ask a couple questions first, and ultimately have you describe the company through those answers. But just giving what the company does, I want to ask this first: if you're enabling planar optic designers and meta optic designers to optimize their designs, obviously you must have arrived as a business at a point where it was clear that this was a technology that needed a solution on the software side to take it to the next level. In just sort of in the technology arc of metasurfaces, when was that clear to you that this software solution was necessary?
I think the first time it became clear to me was now around six years ago in 2018. At that time I was involved in a very different technical field in organic LED displays and solar cells, and you were starting to see a lot of buzz about metasurfaces. At people at conferences and they were doing metasurface research but it was all very patchy. So I ended up straight up asking them like, isn't there some easy streamlined way to do this? And they said, well no, and that sort of got the cogs moving inside my head because in my background as an optical researcher I knew the different parts that had to be combined in order to make this work. So that was a trigger for me at first to start working on this and start making this software, and that ultimately led to founding the company and starting a business.
Every worthwhile technology, be it displays, organic materials, or metasurfaces, there's obviously some buzz, and then the buzz transitions to a reality where there's commercial success. You obviously perceived the buzz and heard about the buzz, and that was an impetus to start the company. Was there then a next level point where you realized, wow, this is really a technology that's taking off and changing the way that we do optics?
It didn't come as like a one sort of lightning flash moment. What we noticed more is how the type of conversations we were having with researchers, with customers, with R&D people was changing. Like at first the questions were really like, can we do this at all? What can it be used for? And as we started getting past that change, of course some people were earlier with this than others, but nowadays we see a lot more questions coming about manufacturability, about cost points at high volume manufacturing, about lifetime, about damage thresholds, these sort of questions that show that it's no longer an academic research project. People are seriously considering getting them into mass market applications and are getting more and more into the engineering aspect. And this is a great sign of maturity. So this sort of for us also showed that we're now in a different stage where we're no longer in the technology hype part but really moving to fixing out all the bugs, all the little diseases that you have with every new technology until it's really to a point where it can be applied at mass scale. So we're really maturing now with this technology, and I find that very exciting.
As a technology has matured, and I do promise we'll get into what PlanOpSim software is doing, I just have one more sort of table setting question. This transition, you know, obviously there's a natural transition from R&D and research to industry. Was that apparent at any point where you started noticing that customers were not just experimental, that they were really commercial?
It happened gradually. Some customers especially were very early with tackling these questions, and I think that's very wise because ultimately these are a lot of the big showstoppers that can happen to a new technology in this valley of that kind of phase, are those kind of questions about industrialization, robustness, cost points, and so on. So they were right to start tackling that early on. But there was no... I'm sorry, I forgot a little bit what the question was.
Just asking about if you noticed or perceived a shift from customers and their questions coming from the research side to more coming from the commercial side.
Yeah, we've seen that as well. So early on it was all research, and research doesn't stop. We still see a lot of research customers as well, but what started happening was the larger companies got a lot more interested, got a lot more serious in the type of questions they were asking. And now that's going on, and I still think there's a lot of potential, so I don't expect research to stop. This sort of draws back also on my experience in liquid crystals and organic LEDs, as long as there's a good application and a big market, there will be no end to the R&D. They will keep refining those solutions because that's what you need to do, especially in consumer electronics. New products need to be released all the time.
I want to ask now, I guess the million dollar question for the purposes of this interview, and feel free to divulge as many secrets as you'd like, but what is it that PlanOpSim's solution is delivering?
Essentially what we do with our software is we speed up the R&D cycle in optical design, and especially when it comes to these subwavelength nanostructures, the calculations and the design. And anyone that's worked with any of these softwares knows that it's painstakingly slow. You have to do very detailed solution and you can only do very small areas, and just the CPU calculations take weeks to months to get to a design. Individual calculations, it's very common to see people running things for days or several hours. So that just slows people down. And our core realization, this is not just about speed of computing, it's about applying the right methods. So what we try to do in everything we do in our software is to make that design cycle shorter. So that means faster calculations, taking the right approximations, but it also means streamlining the connection between design and manufacturing. For instance by outputting these files, doing these analysis that the manufacturing will need that maybe the designer is not concerned with in the first instance. Taking all of these shortcuts because in the end computer time is one thing, but the real costly aspect is a slowdown in time to market. That's why you need these fast R&D cycles, and say the naked hourly cost of qualified engineers, because these are always very highly skilled people that you need to do these jobs, so you don't find them on every corner of the street.
I think it's fair to say, and this is not the case with every company in the optics photonics space, but this is a unique company, right? When you talk about metasurface simulation software there aren't boundless companies doing this. It's a bona fide, I think fair to say, new type of company. That said, were there any, you think about other companies in the design software space, were there any lessons borrowed or any lessons applied from existing software solutions and then taking those and applying them to metasurface design and simulation?
Obviously I drew a lot on my experience in the display and solar cell simulation industry. So there I learned a lot. It basically transformed me from being a scientist and a researcher into understanding how an R&D company and an R&D software is supposed to work, what's really about that, how does it interact with users. From a business side of course, how's the company organized, how should you communicate, how should you sell this type of software. That's something I learned very much in those years. And what struck me still the most is a lot of software solutions are still very general. You can see that a lot of these things are trying to solve everything, and that's great of course if you can solve everything, but it naturally also means that you have a somewhat generic solution. What we try to do very differently in PlanOpSim is to really be specific on our technology of metasurfaces. Because running calculations is fine but without the physical understanding of what you're doing, which is where the real expertise is, and we try to get as much of that as well into the hands of the customers, because I like to tell the people that we shouldn't offer a software, we should offer expertise. And that's all condensed into the software. That's our philosophy here.
We will talk about LEDs, displays, liquid crystal technology in a few moments, because I think that is actually quite relevant to what PlanOpSim is doing and what you are doing. But I want to talk for the purposes of this question about meta-optical technology, planar optics technology. It's still widely considered a relatively new technology area, and that means the applications are new, there are new applications, evolving applications. What type of applications is PlanOpSim seeing now, and have those evolved from when the company really launched in earnest?
Yeah, they've changed a lot over the years. In the research phase, I would say in the hype phase, people always thought, ah, we will be able to do everything, we will replace all imaging systems with a single lens. It soon of course becomes obvious that there is no magic, there's no single lens that will replace a full five, ten lens imaging systems except in very rare cases maybe. So those sort of pie in the sky questions have gone away, and we've really discovered what this technology is good for. And that's what we see now in the first commercial applications as well. We see a lot of sensing because the technology is naturally very good in shaping the wavefront for a limited spectrum. So it's a natural fit with sensing, all types of specialty beams. And we see also that the market, the customers are more and more realizing that. We see fewer and fewer people thinking that they're going to cram a whole imaging camera into a single lens, which we might have encountered three, four years ago.
I think this is in some ways an obvious question, but bear with me. I just want to sort of explore the relationship between new and emergent applications and then the solution on the software side or the expertise that you're providing. Can you explain that relationship a little bit? Because again in some ways it's obvious but also these are new applications so we're going into territory that is somewhat uncharted.
Is your question about the relation between expertise and software? No? It's more about the... just so you encounter a new application or someone comes to you with a problem and they're not quite sure how to solve it because there isn't a template for it necessarily. How is that addressed on the software side? Yeah, actually we do that a lot with our team, because usually people come to us with a specific problem. Like, hey, I'm trying to make this. My first instinct as an engineer is always, I want to solve this, I want to figure out a way how to do this. And we sort of have to tell ourselves at some point, well, we shouldn't do the customer's job for them. But what we do essentially at that point is we take people through their ideas and how they could tackle that design, how they could start doing that. Like we usually will point them also to some of the literature, cases, or examples, or similar technologies. Like, hey, yeah, I think this can be done for this and this reason. Or in some cases it's also, well, and I'm sure imaging designers or illumination designers will be familiar with this, some customer comes with something that violates laws of physics, and then our job is actually to say, well, try what you like, you're not going to be able to do this.
Fair enough. And I do want to get back to your experience in displays and display materials. Just a couple questions about that. When you're working with OLEDs or liquid crystal displays, that has a much more established technology arc, right? The application is much more established. Then you go into something like metalenses or metasurfaces, and especially on the design side, that's much newer. Did your experience working in an established, a more established technology era benefit you as you made this transition in your career?
I'm not sure I agree if it's all that established. People always expect technology to move very fast, but the reality also for flat panel displays like liquid crystals or solar cells, that between like the R&D technical emergence of an application and a widespread adaptation there's often two, three decades. So in my time in OLED I saw that transition happening. When I was still part of it, people were still wondering, well, will it ever be in a real product? It was only after several years, I think they had been doing OLED displays for at least ten, fifteen years until Samsung put the first OLED into a smartphone. And in that respect the development of metasurfaces has actually been much faster than I expected. Of course you can argue when exactly was the first metasurface made and what qualifies as a metasurface and what not, but that's all let's say semantic discussion, doesn't really matter. But between let's say the first hype and now, products entering the market has been relatively short.
Liquid crystals are interesting because I think most people, I would say four out of five, maybe even nine out of ten, when they hear LCs they're inclined to add a D, right? They think liquid crystal displays. Suppose there are folks who think about spatial light modulation too, but liquid crystals themselves, just as a material, a lot of potential in things like sensing, and not just thermal sensing either. Pretty established capabilities. I mean, do you see any parallels between something like liquid crystals and metasurfaces where the sort of the application window is always open, just continuously expanding?
Yeah, I think it's definitely with any... this is like a toolbox, right? In the case of a liquid crystal you have the ability to change the optical system with voltage. I'm talking this is a very crude description of course, there's all kinds of engineering that you actually need to do. And with metasurfaces it's very similar, you have this toolbox of true the nanostructure manipulating the light, and that's a very wide open space. You can apply this in a lot of directions. And that sort of gives the, as they say in French, l'embarras du choix, the agony of choice, that there's infinite possibilities. That puts the burden on the designer then with, well, what do I actually want to do? I used to not have control, now I have control, and now I need to figure out what will that control bring me in a real world application. And that's what makes me convinced also that we're still in the infinite infancy of what's possible with a metasurface. We see a lot of people trying to replace existing solutions with a smaller version of it. Like you make a lens, it's a big block of glass, you take that functionality and you make it into a metasurface, and now it's a tiny piece. And that's good, that's great progress. What I think people have not really started tackling yet is, well, let's walk away from curved surfaces, what other crazy things can we do now that we have full control with it? And there's some work on this but I would say this type of applications, the community, the field as a whole has not really tackled this yet or exploited this potential completely.
