Graphene at Scale: How Paragraf Is Realising the Wonder Material’s Potential | Dr Simon Thomas

Dr Simon Thomas never expected to be a founder. After 12 years inside TSMC, Samsung and LG, he was ready for a break from the pressure of the world’s biggest fabs. Then Professor Sir Colin Humphreys set him a challenge: could he make graphene at wafer scale – the same problem Samsung had spent $2.2 billion trying, and failing, to crack?

Five weeks of experiments later, Simon saw the first promising result. A short time after that, Paragraf was born, and by December 2024 the company had switched on the world’s first graphene electronics foundry.

In this episode of Lab to Market Leadership, Simon explains how a decade‑plus in Asian fabs gave him a process‑engineering lens no one else brought to graphene, how a chance train journey with Colin Humphreys became the turning point, and why copper‑free, transfer‑free graphene was essential to escape the industry’s 'trough of disillusionment.'

He also talks candidly about the hardest part of scaling a Deep Tech company: people. From hiring the first team to recognising when someone’s journey with the company is over, and from open‑door collaboration to faster, tougher CEO calls, this is essential listening for technical founders learning to lead as CEOs.

Learn more about Lab to Market Leadership: https://www.deeptechleaders.com

Follow us on LinkedIn: https://www.linkedin.com/company/deeptechleaders

Podcast Production: Beauxhaus

Transcript

0:00

The goal now is to be successful. The goal is to get the value that this company can deliver and the benefits. One of the biggest reasons we have some great people here is we are creating huge benefits for the world, for the people of the world. And you've got to think about that as well. I've got to make this hard decision because actually this molecular sensor that we want to get into testing potassium could save people's lives from kidney disease. So I don't want to get stuck in the decision now because somebody wants to make the PCB pink and not, not green. No, it's green. Done. Go. No, but I like pink. I don't care. Um, being a bit flippant about it, but it's, it's… You have to realize that particularly in the, in the strategic leadership positions, speed of decision is critical because if it's not fast, the time you waste here gets magnified in the time that gets wasted in, in the tasks that are being done to create the product or the technology, so. Welcome to the Lab to Market Leadership Podcast. Too many advanced science and engineering companies fail to deliver their innovations from the lab to the market. We're on a mission to change that. My name is Chris Reichhelm, and I'm the founder and CEO of Deep Tech Leaders. Each week, we speak with some of the world's leading entrepreneurs, investors, corporates, and policy makers about what it takes to succeed on the lab to market journey. Join us Welcome back, everyone. This is the second in our two-part series on founding CEOs, and in particular, founding CEO archetypes. From our observations as deep tech leaders, we recognize five typical founding CEO archetypes. If you wanna know more about them, head to our website, deeptechleaders.com. Check out Insights. You'll find a white paper and presentations all about this. But briefly, there are five founding CEO archetypes, including the scientific steward, the systems builder, the visionary narrator, the repeat founder, and the domain insider. Those are the five typical founding CEO archetypes that we identify in companies. Now, why is this important? Why does it matter what type of founding CEO archetype you have? That's what we want to find out. Does it make a difference? And we would, uh, we would hypothesize that it probably does. Not that one would naturally be superior to another. I don't think it works that way. But there may be certain archetypes that do much better through certain phases and stages of the lab to market journey than others, and that's what this is going to explore. In the first part of the series, we met with Dr. Jan Goetz, who is the co-founder and CEO of IQM, Europe's leading quantum computing company. Jan embodies the scientific steward, from researcher through to founding CEO, and all of the, uh, challenges and opportunities that his life in IQM has thrown at him since founding the company. Today, we're talking with a different type of founding CEO archetype. We're talking to a systems builder. Now, systems builders tend to be a little bit of engineers. You know, they're engineers in their background. They tend to come from very applied engineering areas. Think aerospace, uh, complex systems integration, industrial automation, and semiconductors, okay? Semiconductors are a big one. Our guest today c- spent twenty years in the semiconductor industry before founding his company, Paragraf. Paragraf does graphene electronics, and the core insight of the company, I believe, came from his understanding on the way that semiconductors are actually produced. So what kind of lab to market journey has Paragraf had? That's what we're gonna explore. We're gonna explore a lot more with Simon. This is gonna be a wonderful episode. I hope you enjoy. Let's get into it Simon Thomas, thank you so much for joining me. Wonderful to be here, Chris. Thank you for the opportunity to have a chat with you. Did you always wanna be a founder? Was that a dream? No, never. In fact, it was kind of came out of, came out of the blue, to be honest. I think my, uh, my career development path has been not what I expected at all, and to say the, to say the least. And I think that I, I never really would have thought of starting a company if it wasn't for the fact that I'd managed to work with, with Professor Sir Colin Humphreys, and we'd made this breakthrough and, and actually there was a bit of an ecosystem in, in Cambridge that was quite supportive of taking what had been invented at the university and turning it into something that could be highly valuable. But no, nowhere in my life did I ever think I'd be founding a company, so. Hmm. What was the, what was the catalyst for you? What was that trigger or event or whatever it is that kinda gave you that push and said, "You know what? Maybe"? Well, I think the development work that we did at the University of Cambridge, um, on graphene was trying to solve a problem. It was trying to take what already existed in the world, which was the material graphene, but turn it into a solution that could be, uh, integrated with the rest of the semicon world so that it could be used commercially, and you could get that material into application spaces that people could harness, if you like. And so we started from that viewpoint of let's make sure whatever we do when we're creating this, this material and synthesizing this material, let's make sure whatever we do can be utilized in the real world. When we'd achieved the goal, there was a lot of papers, a lot of publishers, a lot of patents on graphene already. So the choice was basically, do we publish some more papers through the university academic press and say, "Yes, we've done something similar to you. Here's another graphene paper." Or did we want to take the material we had and the invention we had and m- potentially impact in the real world environment? And so I, I think if there hadn't been so much history behind graphene already, if there hadn't been so many, you know, 15 years of people publishing papers, we may have thought, "Well, we're one of the first to do this. Let's go to a nature paper. Let's get the, you know, let's get the, um, uh, the, the kind of academic route going where people can start." But we'd had… You could look round. Graphene had one of the fastest publishing rates of any material that had ever been invented. On top of that, uh, as I mentioned previously, there's a, there's an environment here in, in the University of Cambridge, particularly with Cambridge Enterprise, of looking at whether value can be derived from university research. And surrounding that, there's a network of people that have done it before, and you come across them all the time. And so there's kind of an ecosystem of I wanna say entrepreneurial ecosystem that, that you can feel, you can tangibly feel when you talk to the people that have launched companies through Cambridge University or have actually been really successful with companies in Cambridge. So you kind of have this other, I don't know, this other person sitting on your shoulder saying, "Well, we, we could do the academic stuff or we could do the real world stuff," and you have to make a choice. And fortunately for me, I met, I met somebody called Andrew Lynn, um, who, uh, was CEO of an… A founder CEO of another company, and after having quite a long chat with him, it became very apparent to me of what we could achieve through the invention we made if we could push it a bit further. And by push it a bit further, I mean get it into people's hands, get it out in the real world, get people to understand what was now possible. Because Paragraf's vision in order to get graphene into electronics is reliant on other people understanding what they want to do with the graphene. So if you keep publishing academically, you're probably not gonna get those partners that you want to help you drive forward. And therefore really the inception of the idea of creating Paragraf was a collection of those things. The, the research impact, the environment we were sitting in, and then having the fortune to meet people like Andrew that could really paint the picture of what's achievable through a spin-out or founding a company. Yeah. Yeah. But also I'm getting, I'm guessing tied up in all of that too was the confidence that must have come from Andrew and others that you know what, and you could do this. Yes, absolutely. Absolutely. I think- I was, I was very fortunate to be supported by quite a few people when I looked at spinning a company up. Um, but I'm sure we'll come onto it later. My, my background isn't… I didn't do a PhD and then postdoc and spin a company. I spent many years in industry previously, so I had some acumen on that side in how a business runs and- Yeah… essentially what would be required to turn the idea into the business. So I think that'll come in- Can you talk a little bit about that? Can you kind of dive into the pre-Paragraf days? You know, what skills and experience had you accumulated up until the point where you started to engage, uh, with, uh, Professor Sir Colin Humphreys? Testing my memory now, I think. Sorry. No, it's, um, so I, my, my, my original degree was in aerospace engineering, and so I did that in the University of Liverpool. I then went on to do, because of that link to aerospace engineering as part of my degree, I, I started to look at, uh, aerospace alloys for a project. That then turned into a material science PhD, only the aerospace alloys turned into semiconductors. Um, so I spent, spent, uh, my PhD researching semiconductors, uh, particularly gallium nitride, which meant I ended up getting a career as a process engineer for an equipment company, mainly working in, in Asia, so Taiwan, China, South Korea, Japan, where I got to see the inside of some of the world's largest fabs. I got to understand how the foundries work. I got to understand how the semiconductor ecosystem works. So not just the materials process, which was kind of my background, but also how you go from materials through to development of devices, through to end product, and seeing them, seeing that process happen in what's the gold standard of the world, basically. I've been to TSMC- Mm … Samsung, LG. And so I built up that knowledge background, um, over a period of about 12 years. When I left that career, um, because of my background in gallium nitride, it was Professor Humphreys that realized that I could potentially be useful to his research group who were working on gallium nitride on silicon. And so over a period of about a few weeks, Colin persuaded me that research was just as relaxing as, as having some time off, because at that point in time, I'd, I'd decided to, to take a year out. I just wanted to have some… I just wanted to have a break. It'd been a pretty intensive 12 years. Hmm. But he persuaded me and, um, I, you know, I-- you don't often get the opportunity to work with somebody like Colin, and I couldn't pass that up, so. Yeah. I joined him at the university and where I was looking at bringing my knowledge from, from the field from working those foundries into his processes for creating gallium nitride on, on what was, uh, then a big target, which was creating it on silicon large scale- Hmm. Hmm … of which there were a few process issues. As it turned out, when I arrived, Colin had a project with the Nobel Prize winners for graphene in Manchester. And so Colin wanted to create a hybrid device that had a better contact layer than what we were currently doing with, uh, indium tin oxide, a material that is becoming quite scarc- scarce in the world. Um, so he wanted to remove that and put graphene in its place and create this conductive graphene layer. So working with the Nobel Prize winners, Colin made wafers of, large wafers, eight-inch wafers of graphene, and then we were gonna combine that with the materials that came from Manchester. However, the materials in Manchester were tiny little flakes of graphene. And so creating, again, real world impact commercial scale was not possible by combining these two projects together And I remember that on the train home on the way back from Manchester, Colin and I had a chat, and he said, "Do you think there's any way that you could make graphene in wafer scale?" Now bearing in mind, people have been trying to do this for 10 years. Samsung had put $2.2 billion into trying to do this. So my initial answer was, "Probably doubt it." But it was one of those serendipitous moments. Nobody had tried to use the tool sets that I was using in my compound semiconductor world in the graphene world. And so after a little bit of thought about it, thought we'd do a few experiments at the university, seeing if my process knowledge from all that I'd learnt in the foundries and the fabs that where I'd been working over in Asia could be applied to creating a single layer of carbon atoms on a surface. Doesn't, didn't sound like it was going to, but lo and behold, it worked. Yeah. Which was kind of, you know, as they say, bringing two ecosystems and two worlds together often results in innovation, deep innovation. Yeah. And that's clearly what happened here. What, what, um, how quickly did it happen? From kind of idea on the train and this frustration and okay, you know, we're playing with tiny amounts and tiny sizes of graphene here, we need to get something bigger, to the actually being a… And then the idea and then being able to manifest it So the idea Design experiment, obviously, or a build, some kind of prototype So the first results that looked promising took about five weeks. Um, so we managed to create-- wasn't real graphene at that point. It was more of an amorphous carbon on the surface, but it was starting to show from the analysis we were doing some of the characteristics of graphene. So I, I knew that the theory had a possibility as soon as we saw that result. So from kind of two or three days worth of thinking about it, trying to set the reactor up in a certain way, and then yeah, five weeks later, we'd, we'd managed to get that first result I then understood what needed to be done to get that material better and better. And so for the next six months, we refined the process, refined the material. Um, but not really. That first five weeks was the, was the critical part. If, if I'd have got to the end of that, those experiments of which I'd designed, you know, a set of experiments, um, and we hadn't seen that result, then we- I probably would've stopped. But your, your background, all of that experience you had for the prior 12 years out in Asia working in production on compound semis, was it that that gave you the insight of what to do next? But the kind of data that you had seen through the experiment. So my role- And do you think, and if I may, do you think, you know, would any other member of, of, uh, of Colin's research group been able to figure out what to do next with that data? Or do you think a typical researcher would have been able to figure out what to, what you could do with that data from that point? Not unless they've come from my world, I don't think. I spent a long time developing materials for customers and that's process tweaks, understanding. Through my, through my role, I became highly educated in chemistry, highly educated in physics, and the applied sciences of thermodynamics. I'd obviously learnt about this stuff in my degree, but not really done it hands-on, and it's that hands-on experience where you understand that theory isn't the same as the real world, and there's, uh, always a delta between what you believe and what you actually can do. The process that you use to create compound semiconductors is actually a really complicated process, and even to date, people still don't understand the reactions and mechanisms that occur in creating some of the crystal structures that we have and take for granted in some of the devices we have today. So I, I don't think unless you had a background in the particular growth processes that I did, I, I think it would be very, very difficult. And also experience. I mean, without going into too much into the IP- Hmm … the way I approached thinking about how to create the material would only have been possible through understanding how the crystal structure of gallium nitride worked, and in particular, the problems from creating gallium nitride and how you could use one of those problems as an advantage Again, I'm, I'm, I'm walking a very fine line between actually, uh- And is that… no, no, this is really… no, this is great. But again, let me ask, is this, uh, is this a function or is this derivative from your scientific knowledge or from your practical knowledge? I think it's a function of both. Yeah. I think you, you have to have the science thought path, if you like, in order to try and assess what's going on. But then as I mentioned, that delta between scientific sorts and putting- Yeah … a load of equations on a page and then turning them into what actually comes out at the other end of a, a, a physical process. Yeah. My, my role, um, in that, in that foundry process when I was working in Asia was, was really that, that, that kind of, that kind of career of how do we take what we understand in theory and how do we convert it into material a customer wants, because there's a gap in between. And while theory is nice and, and the output that you get is nice, what you want to do is define a controllable process. And what that means is of all of the inputs that go into the chamber to make a material, you've got to understand which one of, which one of those levers you can pull with those inputs gives you the output that you're gonna get for a material performance, if you like. And so you have to have both of those. If you don't, then you're gonna be missing a, a, a very, very large understanding of what you're trying to achieve. And I think that's kind of what happened in the earlier days of graphene, where people using metal catalysts to create graphene, where that process very, very clear of how a sublimation process works. So the dissolving of the graphene precursor into the, into the copper, and then the ejection of it as the copper cools. Theory's perfect. It, it's really, it's a really nice process. But what you eventually get at the end is graphene that's got copper in it. You shouldn't. Theory states you shouldn't. All of the carbon should be ejected into the surface, but that's not what happens in real life. No. And that's why I believe that process did not work, because that gap, that delta between science and real was not assessed beforehand. And- And it, and it sounds like your, your experience and insight, and again, you know, between the science and, you know, the delta be- uh, between the theory and the practical, that saved you a tremendous amount of time, I'm guessing. Even though it took you another six months to kind of build on that original hypothesis and original experiment, I'm guessing, I'm assuming that, uh, knowing what to do from that point saved you an awful lot of time. Absolutely. So the, as we call them, the design of experiment frameworks that you use in this type of process, I'd been using them for 12 years, and it was, "Okay, I'm not creating a compound semiconductor. I'm not creating gallium nitride or indium gallium arsenide." and creating a single layer of atoms, but the approaches you take, whether, you know, the different process flows, the different temperatures, vacuum levels, the, uh, the chemical precursors you use, all of those things, uh, it's a very well-known process to me of how you get those to work within the parameter space. So yeah, absolutely. Without having that background, could, uh, could you have achieved what, uh, what we were doing? Probably, but you'd probably still be experimenting now Yeah. Yeah, or you would have had to go through the, "Well, we think we might have something, but we've got to demonstrate it, so now we're gonna incorporate and spin out." And then, "Oh, we have to hire someone who knows about these things, so let's go hire an engineering lead or let's go hire someone else to, you know, to, to help us scale this and figure out what to do." Yeah. Yeah, absolutely. But I, I think you'd almost need the engineering person at the start because the process design itself right from the beginning would require you to understand which levers you need to pull, for example, to- Yeah not, not remove material from the surface of the wafer and deposit material. So it's a para- it's a parasitic process. I wonder if that… You know, I've had other material scientists or material science leaders talk about this before and, you know, I've asked them, "If you could do it all again and make that journey smoother or more likely to end in success, what would you do differently?" And a lot have said, "I'd have every, I'd have all the key functions up front to fi- because you need to understand- Yeah … the dynamic between all of this stuff at the early stages to avoid making really consequential negative, uh, mistakes along the way." Yeah. Um, mater- as you know, materials businesses and materials dev- uh, developments are some of the most difficult to do. Yeah. And the, the, the UK landscape is littered with materials companies- Yeah … that didn't quite make it, and that all comes back to that point I was making, the difference between theory and real world. Something in the real world always gets in the way. You know, you don't- Yeah … account for the fact there's other molecules around. You don't account for the fact that the stoichiometry of the material when it cools is different from what it is in theory. The phase graphs- Yeah … that you have of these perfect phase graphs, you know, you don't perfectly follow the line because something in the at- atmosphere changes, and so there's a lot of complicated parameters, um, to creating new materials. But you're absolutely right. If you could have all that information up front, then perfect. But- Yeah. When you- Then, then you wouldn't need an inventor, would you? So… Well, well, yeah, but you still need that spark. You still need, you still need the, the, you know, the fundamental question or hypothesis or whatever it is. But you still- Mm … need that spark. But if, um… You know, it sounds like… But this, you know, you described it earlier as serendipity. You were playing with gallium nitride to start, and then you were looking for that initial contact layer. Mm-hmm."Let's play with the contact layer and see if we can improve that," and then that suddenly became something else- Yeah which then, you know, you know, created the innovation for Paragraf. Yeah. It doesn't sound like it was initially the core focus, "Hey, let's do graphene on electronics and see how that goes." No, no, it wasn't actually. The, the work I was joining Sir Colin to do was actually a new methodology to create gallium nitride on diamond substrates. Right. Um, that… So there was the silicon project, and then there was this diamond pro- the, the kind of interesting thing to me was the new substrate of diamond, which was in the news everywhere at that point in time. Yeah. Um, but then, as you say, a, a trip to see- Yeah … Manchester University, and then the conversation started. And then there's a, there's, there's a, there's a question, right? Exactly. Is this po- is this possible? How, uh, when you guys… How quickly after that did you guys incorporate? After that kind of, "Hey, we've got something with carbon," with this kind of amorphous carbon that you described, and then later on you could see, "Well, actually, there's enough here that's relevant to graphene. I think we might be able to do something," and then you had another six months of experiments or a number six months of development work. How, how quickly after that did you incorporate and then start thinking about it as a business? So after the six months of work, um, probably during that six months, we decided, we made that decision to patent as opposed to publish. Yep. We had enough information to understand that, that the, the graphene process was going to work. So we went down the route of creating a patent. I think it was the August or September of that year that we filed the patent in the UK, and then we, we put an entity in place, which was actually called, um, 2D Technologies. I almost forgot what the name of the company was. So yeah, we started off with… We used to be called 2D Technologies and, but that was really just to, to have an entity that the patent could sit with. Mm-hmm. Um, probably another, another six months after that, we started to look at projects that we could work with other entities with to try and prove that the graphene had an application space. So to understand what I mean by that, graphene had a pretty bad reputation in 2015, 2016. There'd been a lot of investment in companies. There'd been a lot of hype around this copper-enabled graphene growth technique. There was even a statement made that everyone knows how to make graphene now, so electronics are coming. And so hundreds of millions of dollars got put into companies that subsequently were not successful. So when we came out with the graphene layer that was pristine on a substrate,

the first question that was asked was: 27:22

"Well, what's different about your graphene compared to what's failed in the past?" And you can explain it. You can articulate it. Of course you can. And you can do, um, some academic experiments to show, for example, uh, an element map. There's an element map of a wafer. There's no copper there, right? See? So there's no copper. We're good. But I think someone once said that the, the, the trough of disillusionment for graphene is the deepest trough that anyone's ever seen. And so we had to climb out of that. So we were- Wow … scaling up the side of the other side, if you like. And to do that, we, we decided that we would create some of the first devices made of that graphene. And so we spent six months doing that after, after we'd developed the process. The process continued to get developed in the background, but we also created some of those first prototype devices, if you like. Mm-hmm. Once we got some performance parameters out of, out of them so we could see that not only was the graphene incorporated properly but there was that shining light that this device is gonna be very good if we could make it outside of a university environment, that's when we started to think about the real company. Yeah. So even though we had 2D technologies in the background sitting there holding the patent, we actually in that time when we made the device, our first device prototype… well, not even prototype, proof of concept We published another patent at that point in time. Yep. Because what we started to learn was the techniques to con- to actually create the device. How you put metals onto graphene, how you put dielectrics onto graphene, that was all, uh, new IP as well. Okay. Okay. Did-- You described earlier about the-- having to-- about the deep trough associated, you know, the deepest trough of disillusionment because of graphene and all the hype that had come on. I remember those days. The, it-- When you demonstrated that there-- that your graphene on this substrate was copper free, did the market underst-- you know, did they appreciate what that meant? In other words, did the market understand the questions or the challenges that this had at the time? Or were they just kinda looking at it blankly saying, "Anything with graphene is kind of crap. We're not gonna look at it." Definitely the latter. Um… Okay. Uh, you've gotta remember companies also invested in taking graphene in and working on graphene. So you go back to that fact I was saying before about how much Samsung spent on trying to get graphene to work. And so it was a-- that was like a big figure in the world. Everyone knew it. Over two billion dollars had been spent by Samsung to create graphene. And it's that age-old, you know, when we start talking to investors,

it's that age-old question: 30:26

"Well, Samsung couldn't do it. Why do you think you can do it?" It's… Maybe we're trying something different. Uh, but having a badge like that… And of course, don't forget Intel spent nearly a billion dollars on trying to do this as well. Yeah. Wow. So getting over that huge hurdle- So, so it was just kind of almost at the, you know, before you get to the front door, at the lobby, they're just kinda saying like, "Forget it. We're not doing this." So you also had to kind of educate them on what the results you were seeing meant. Absolutely. And how-- Okay. Yeah. I can give you an anecdote if you like, 'cause it's- Yeah, please … uh, an interesting one. I won't name the company, but it's a very large company, one of the largest in this country, so it's one of two. Um, so, uh, we went to South Korea, um, as part of meeting with some of our customers, and during that time, we got the opportunity to meet with one of these large foundries, a meeting with the CTO of the company of all people, and we sat down and I had a wafer of graphene with me. We introduced each other, whole group of, of, of what I believe was high up scientists in the company. The CTO looked at me across the table and said, "Before we get started, I just want to let you know that you're a liar." Gosh. I've never- Oh, dear … never been in a meeting where the CTO of a large global corporation just looks at me and says, "You're a liar." Um, so it was a, it was- Whether it was tactics to put me off, I, I don't know. But what he actually meant is This company had been working on graphene for a long time, and he himself, as part of one of the projects he owned, was trying to create graphene. And so he knew the challenges to making wafer scale graphene, and he knew the challenges to making pristine single monolayer graphene across a substrate. And he could not see a way forward to do it. So when we sat there and explained to them, here's the wafer, this is what we've done, this is the layers that we've got. A one-hour meeting with all of these top several executives turned into a two-and-a-half-hour meeting because by the-- when we started explaining where we've got to, why we'd been able to do what we were doing, and we made them aware that we were fully cognizant of the issues that they'd had, then it became a very different conversation. So it turned out to be a good meeting. It was a shocking meeting to start with, but in the end it turned out to be It, it was an education for them and for us because they learned that actually something was coming, but I learned really, really starkly just how little the world trusted graphene anymore. Hmm. And the ability to create it. So It was an int- it was an interesting point in my career. How many, how many people were in Paragraf or 2D Technologies at that time? So it was Paragraf at that point. Um, we pretty much changed our name when we launched the company. Yeah. Uh, when we spun out. So we had probably about forty people at that point. Okay. All right. So you had critical mass. You definitely had critical mass at that point. Yeah, because we had our R&D site, which, um- Yeah … we needed to hire quite a lot of scientists to start with. Yeah. Um- But just being able to hold that discussion, to have that discussion, and to convince them then and there that we know the, we understand the reasons for the skepticism slash cynicism regarding our claims, um, uh, a bold opening line, but we appreciate the frustration, and obviously you've been trying to make this for a while, but here's… You know. And so, you know, if I think about you as a CEO for the company at that point and what you bring to it, you're un- y- you're able to have… You're able to build a lot of trust in that meeting because you understand the processes, a lot of the processes that they will have gone through, and the frustration. You will have understood the practicalities of, uh, and the frustration in them, in, in practically trying to do this, as opposed to someone who's just existed in a lab filing, doing research. You know, a classic almost scientific steward, which is one of these other, which is a classic founding CEO archetype. But y- that experience allows you to build trust that much faster. Obviously, you also have, "Here's this wafer of graphene that we produced too, and it works, and here are all our findings," and, you know, we can… Not, I don't know if you'd wanna tell them how to do it, but, you know, you can say, "We've worked around it. We figured something out here." And then I'm, I- I'm guessing that brings you so much further, so much faster at that point in your journey. Uh, absolutely. I think the combination, going back to what we were discussing earlier, the combination of the ability to be able to explain the science but also explain the real world, i.e., how- Yeah … a process line works, what the problems are with the techniques they were using. I think that was one of the biggest, if you like, uh, dynamics of the conversation. I was able to tell them how their process lines work and why what they were doing previously doesn't, um, why what we're doing now does, and that's the practical hands-on Part of my career in compound semi that allowed me to have those conversations. You know, walking into a foundry and having the ability to, to articulate why that foundry needs to do certain things in order to create devices, in order to create technologies is a massive advantage because you're, you're immediately on the same level playing field with them. If you go in with just a scientific theory, then you're, you're not going to be seen as somebody that understands the practicalities of what's trying to be achieved. If you go in with just the foundry knowledge, then you're also sitting with the person that ran that project on graphene, so understands the science. So you kind of… I think the combination of the, the educations is, is, is absolutely critical. Yeah. How, um, where were you on your journey, you know, kind of at that point? You're forty people. You'd been capitalized at that point as well. Mm-hmm. You know, you know, can you kind of walk us through your lab to market journey and almost kind of bring us up to today as well? You know- Mm … it sounds like some parts went quite quickly, um, or as quickly as they can in material science let's say- Mm-hmm'cause materials is a tough area. Um, and, and kind of where you've gotten to today as well. Sure, sure. Yeah. So we spun out of the university, uh, in twenty eighteen, early twenty eighteen, and we immediately acquired an R&D site just north of Cambridge University, and that was where we did the tech transfer. And so we, we got some seed funding and were able to put some machines in and showed that you could take what was done in university and transfer it to the real world. Mm-hmm. That didn't take us very long. I think that took us about eighteen months. So the second round came quite quick after that, and the second round was to start putting equipment in so we could start making devices. So if you like, the first stage was about the material. Second stage became about the product that we could make for, for customers. So turning the graphene into something that we could put a vanguard technology into somebody's hands and go, "Look, yeah, we don't just make graphene, but look, we can make devices out of it." That was our, our series A round that, uh, allowed us then to expand that, that R&D facility properly and get our first MVP of a magnetic sensor into the hands of people like CERN and, um, Magnetic Labs and places like that, that just to prove that we could, we could make some. Then when we got to series B, we were, we-- the narrative then became from just not making just prototypes, but turning into real product, real reproducible product. So our series B round of about sixty million dollars. It's when our financing started to move away from UK shores. The lead was the US investor at that point in time Um, which is when we started to think about this site here, which is our manufacturing site. So we have two sites in the UK, the R&D site and the manufacturing site. So we raised some capital to do that and a- another follow-on round, a series C round, to get the capacity into this, this fa- facility here. And I can say quite pleasingly, in December last year, our foundry came online, the world's first graphene electronics foundry. Can't find one anywhere else, uh, any- anywhere else, and we're getting large volume of chips going through right now. Our first volumes are actually due in the next two weeks because it's not just the foundry that you need to do stuff. You need to send it out for, uh- Yeah … packaging and things like that. So we're- Yeah. Yeah … we're looking forward to seeing thousands and thousands of devices on the shelves in the next two weeks. Yep. Beyond that, we're now- But we'll do final assembly will hap- will happen elsewhere? Yes. Yeah. Okay. We, we- Yeah … we started off as a, an IDM, an integrated device manufacturer. Yeah. So every step we did, but the backend steps, the, the far backend steps of packaging and test and singulation are not value-adding, so we wanna get rid of those, if you like- Yes those resource, resource drains from the business. So we just go into that, that new horizon now of, of being a- Yeah … a deep tech materials- Yeah … device manufacturer- Yeah … uh, in the UK, which is challenging landscape in its own right when electricity's so expensive. Um, but- Yes, and not getting cheaper anytime soon. No. No, not at all. Um, what are the… You know, what about the, you know, what about the commercialization journey? You mentioned earlier obviously convincing people that graphene had a place and we could do something with it was a real challenge. Is it still a real challenge, or are those ba- are those, are those obstacles starting to be removed now that you've got real evidence? The major part of our commercial activities, business development if you like, has been education. Going to customers, not just educating about something new, but educating about something new that is not as bad as the thing that came before, and that's a difficult task. It… The ability to go to a customer with a new widget and say, "Look, this is a new thing. This is how it performs," is one thing. But having to go there and qualify that what you're doing now is not the same as what it was in the past- Mm … that's been the real challenge for us. You know, the, the ability to be able to get some trust back in those customers. We've, we've approached several customers that have tried graphene. They've put graphene into their devices, or they've had a graphene company make something for them, and they've been impressed by the performance but very, very disappointed in the fact that you can't get one device the same as the next, and that has a lot to do with the contaminated graphene that existed. It's not uniform. It's, it's highly defective. It's got contamination in it. So we've had to go through that journey of really, really educating people that this is different. It's not what you have seen previously. But those companies have spent money in the past. They've pumped money into something. They don't want to pump it into the same thing again and get the same result, so you have to go that extra mile to get across that implementation gap- Yeah if you like. Yep. And to some extent, we're getting there, but we're still not there. There is still a lot of skepticism about whether this material will prove to be all that people believe it will be. A simple example of that is the automotive industry. We don't really want to go into automotive industry, but there's some high-end applications in automotive. First question we get now is, "Can you give me the 10-year life cycle data sheet, please?" Uh, well, no, it's only been around for four or five years. We can't do that. And so you, you can't go into those spaces because there's n- y- if something fails in a car or it's problematic, and you want 10, 15-year lifespan at least- Yep in automotive. So we have that issue still that it's great if this, this thing that people, particularly investors say, "It's great you've got this differentiation. It's fantastic. You know, you're ahead of the world. You can… You, you've got this huge value proposition When it comes to commercialisation, it's not good at all because if you're the only one singing out about something, then there's an immediate response from partners and customers of, "Well, if you're the only one, does this thing really work? Because surely if you were, if it was so great, there'd be many others singing the same thing alongside you." Yeah. So- And from their perspective, they've got supply chain risk as well. Exactly. Sole supplier issues if no one else can do this but you can. You know, it's a classic sole supplier issue from a supply chain perspective. Yeah. What about, you know, typically how much will they invest? How much will your prospective partners or customers invest in a qualification cycle for something like this? It's different. I think we, we have products in different market spaces now. So on our magnetic sensor, we're in aerospace, we're in space, we're in medical diagnostics. And- To a large extent, it's two parameters. It's how regulated that industry is played off against how much they have a distrust for graphene. Mm-hmm. So if I just expand on that, if we're going into the aerospace industry, you can't have a failure, sp- particularly space. You can't have a failure at all but if graphene is showing that if you put something on a satellite, for example, you can use a 10th of the energy, then all of a sudden there's a kind of… There's a playoff at that satellite company of do we want to put the effort in to save 90% of the energy that goes into our current sensors or, or do we want to make sure that Paragraf have fully qualified this in, you know, space regulated and where, where do we come in as the customer to, to contribute to that? Whereas at the other end of the spectrum in automotive, basically it's a component. If you can't prove that it works, we're not gonna give you any money for it. Um, so I think it's understanding those market spaces, particularly for, for deep tech and, uh, and new technology manufactured products, if you like. Choosing your bets is a real, is a real challenge. Yeah. Uh, so we've had an aerospace company do full extremely expensive radiation testing for us that would've cost us hundreds of thousands of pounds, but they've taken the cost themselves. But we go over to the EV market where they say, "Well, we're… You, you design a chip for us, we'll put it into us, and we'll test it for you, but that's it. You've got to do all the other work and pay for it yourself." Hmm. And so- Yeah … you're seeing, seeing the same kind of pushback in even in, uh, the biosensing markets. People have been trying to make biosensors out of graphene for a long time, and some of the large diagnostics companies have tried them. In the past, they've paid those other graphene companies for the work. Now they won't do it because it failed, and so we have to kind of find the right customers that maybe have not used graphene before and want to get, want to get beyond their competitors, if you like- Yeah … um, in order to be able to leverage some, some value. Yeah. Yeah. Yeah. Yeah. I can, I can see that challenge. Hmm. Um, let's talk a little bit more about you and how you've experienced this journey thus far, Simon. If… What would you say the most challenging parts… Can you kind of enlighten us a little bit on the most challenging parts of the, of the role that you've experienced or whether they've been particular transitions, so maybe they're early stage, you know, you could identify the things that needed fixing or a novel approach, which you clearly did, and so on. You know, were the early days easier and the more recent days super challenging? Was it the other way around? Where have… You know, how have you experienced the challenge? So there's a saying in Paragraf, I'm sure it's the same in any other deep tech company, which is, "This year is our most important year." And that's because every year generally comes with new challenges as you get through each of the, the gates that you need to, to grow the company. The challenges exist. They are just different challenges or of a different scale. I think through, through the history of Paragraf, there have been some very, very difficult challenges, and whether that comes from capital, you know, the capital tech- Yep tech markets collapsed a few years ago. Yeah. Whether that comes from the infrastructure that's available to where we are, you know, building our own infrastructure in the UK has been very difficult. Semiconductor infrastructure- Yep … doesn't really exist in the UK. Um, whether it's been from trying to get that traction with the customers as we've just been talking about, trying to get over that hump of disbelief- Mm because of what's come before has been really challenging as well. But I think Looking back over the whole of the history of the company, the biggest challenge is people. If you can find the right people, then you can do immense things. And that means creating the right culture and the right vision and the right purpose for the company so that you bring the right minds in at the right time and the right capabilities with the right drive to want to do what essentially you're trying to do. And so the challenge for us, and I think this has happened in each of those phases, even though I could go through the series A, B, C phases, I could tell you what the biggest challenges were for those phases. The one underlying theme across all of it is people. Hmm. So we've got some great people here. I mean, they've, they've achieved amazing things, and sometimes I look at what they've done and go,"Wow, how, how did we get to there?" But it's really difficult to keep those people motivated when actually the final goal is big contracts with lots of devices sold to customers, and that's a long journey. It doesn't happen overnight. In fact, in the semicon world, it generally takes 10 to 15 years. So having that, I think for me the, the biggest challenge has been creating the, the right environment at the right time for the right people to get the most out of their skill sets that add up- adds up together- Hmm as that jigsaw to create the best and most powerful company we can- Mm-hmm … aiming towards that end goal. So we put a lot of time into that. We put a lot of effort and energy into, to trying to, to bring the right talent into the business And similarly, the flip side to that is you find your biggest challenges in, in the company are the people. People's motivations change over time. The business changes over time. What's good for somebody in the starting days is not necessarily so good when you're getting pressure from a customer now. Mm. People have different, different ways of working and different things they enjoy doing, and the business has to develop, and sometimes that business change is not suitable for some of the people that are in the business. And I think the biggest challenge I have is, and I th- a lot of people will say it, is knowing when you've reached the end of a road with an employee- Mm … and how to make the experience for them as comfortable and as good as you can for them to move on to the next challenge that they want to take, while also not damaging the business and not un- damaging the morale of the company. So everyone's, uh, kind of win-win. There's a… We don't like change. Humans don't like change, and so you reach a point where you'll have people in the business that are happy to stay because they're comfortable where they are, whereas the business needs something a bit more. And what they don't realize actually quite a lot of the time is that they'd actually be happier if they weren't just sitting and letting the business form around them, that they were off doing another challenge that was in their, if you like, in their wheelhouse. Yeah. Um, so, so yeah, it's all about successful business as far as I'm concerned is all about people and knowing… Having those hard conversations is my biggest challenge, always has been. And, and sometimes I've, I've let that go on for too long- Mm… uh, which when you take a harsh, you light of day, what I should be doing is putting the business first, um, for the investors, but also probably making it easier for people if I didn't take so long over those decisions. But we all have, you know, we all… As a, as a CEO of a company, one of the things you care about the most is the people- Yeah … and making sure that they're happy and making sure that they've got a career that they, they enjoy. Yeah. Particularly- Sorry, particularly for, for a company that's a startup, right? Because you're asking a lot of those first people, asking them to put their careers on the line generally. Um- Yeah … because they're, they're coming into somewhere where there isn't security, like if you're working at a large corporate. So, so yeah, you, you, you always kind of want to try and do the best for the people Have you changed the way you approach your role? Yeah, absolutely. Absolutely. I think there's a big part of this of, that goes a- a- alongside the transition from a technologist to CEO. The way I… When I see a technical challenge, an engineering challenge, I just wanna dive in, and you can't do that. You, you can't for many, many reasons, one of them being, as I just, as I just highlighted with people, you, you, you don't wanna piss off the people that are trying to do the technical stuff, and coming in as a CEO and going, "Oh, well, why are you not doing this?" Is… You really don't wanna do that. Um, but yeah, I think I've become harsher, I think. I think that's become a necessity. I think I've become more critical of myself, uh, and the decisions that I make I also think that I've managed to learn to rely on people more. And that's, for someone with a highly technical background that basically spent 12 years working on my own at customers where the person I relied on was me, that's, that's kind of a transition. And it comes back to that point of if you can get the right people, then you build a level of trust, and you build the ability to go, "Actually, you know, I can step back from this, and I can get on with the more immediate tasks that are required of the CEO of the business." Hmm. But that, that's … You're always, you're always, as an engineer, you always wanna fix the problem. And so if you … I, I generally spend a lot of my time over here in the, in the foundry now as opposed to over in the R&D side because I know if I walk round the lab I'm gonna start getting my, gonna s- start getting my mind set to think- Starting to assemble it all up. Yeah, exactly. Yeah. But e- e- on the, on the getting the right people, um, one of the things we've heard a bunch is the difficulty that some founder CEOs, depending on their background, you may not have this 'cause you had a lot of industry experience beforehand. But on some founder CEOs on actually defining what right looks like- Yeah … um, when they haven't had much industry experience, and even when they have had industry experience. Still, getting those matches quite right, building that core team is quite a tricky thing Absolutely. And I think the definition of what you need changes over time, um, a- as we were just discussing Going, I mean, Paragraf's eight years old now, and so it's been through several phases, and understanding that the team that you need to help you run the business will change over time. At the start, the business is not a corporate. It's, it is just a spinout. You need s- you need people w- with the mindset of, "If there's a problem, we're just gonna fix it. We're just gonna get on with it." You know? You know, if, if, if the toilet needs fixing, I'm just gonna go and fix the toilet. It's that level of, of thought path. Whereas you get towards the end that we're at now, and we are operating as a corporation, as a corporate, as a, as a company that is a real business, an SME, and the governance and the regulations and, and, and generally how you approach business operations is very different. It, it has to be… You know, you talked before about h- the ability to supply customers and resilience and having more than… Having the supply chain. Okay. All of that confidence needs to come out of the business into the customer. You didn't need that when we were doing experiments at CERN with our first, you know, with our first magnetic sensor. Everyone at CERN just wants to play with the new material. But, but now it's not. It's a- it's about, it's about showing that we're a real business, and that needs, that needs specific skill sets and a specific team, so. When did you need to make that transition? When do you… When did you identify that, "Okay, n- now we have to start moving quite quickly in order to, to demonstrate that we belong as a supplier to this community"? I don't think there was ever a trigger point. I, I think- Mm-hmm … it was, it was always in my mind over time that we would have to develop, and we have almost feathered in those changes, and that goes back to your point from before. I think that's because I've worked in industry. I, I was fully aware of the changes that were, that were going on in the business, but that's because I'd worked in a business before. Yeah. And I, I… So I was cognizant of what was going on. So I was on the lookout for good people. Always… I'm always looking around and saying, "Well, actually I've just met that per- They could be, they could be a really good asset in, you know, in two years' time." So keep in contact with them and make sure, make sure that the, the relationships still remains. So I don't really think there was ever one defining moment where I went, "Okay, we've gotta switch from this business type to this business type now." Yeah, yeah, yeah, yeah. And, and so think I've been quite fortunate in that way. We've got s- as I say, we've got a really good team. Yeah. Um, one more question for you. You mentioned earlier that one of the ways in which you've changed as a CEO or the way you've approached the role as CEO is you've become, and you used the term harsher. What do you mean by that? I think there's several aspects to it. From the simple aspects of I just don't have the time to spend with people anymore. So I'd, I'd normally I'd have an open door policy at Paragraf where anyone can come and speak to me. A- and I like that because it enables me to understand the mindset of the company, the mindset of people, where problems might be occurring. And I also want people to feel that I'm here to support them. So that, that simple end of the spectrum of not doing that anymore. Um, I, I do go and talk to people from time to time when, when I feel I need to take the temperature or something. But again, now I have a really good HR director and she handles all of that for me and she's, she's great at it. Mm-hmm. So, so that, that's simple then. But I think also on decision-making, um, from whether we need to keep people in the company, whether I need to start talking about moving people on, they were all things… I, I, I've always had this mindset of we need to give somebody another chance. What would've been three or four chances in the future is now two. Mm-hmm. And even though there may be a negative impact immediately on the business, whether that's to do with the morale of the business from the decision that's being made, it just has to be made. Mm-hmm. I also think that decision-making has to become more rapid as well, and you've got to stop caring about what other people think. And, and that's… It go-- it's-- you generally in a startup, unless, unless you've… Unless you're, like, highly egotistical or you're… Or, or you, you've got some kind of, uh, sociopath tendencies, um, it, it, it's run as a committee. You know, when you- in those early days, you want everybody's input, and you want everybody to talk, and you, and you want that cross-pollination, that cross-contamination to get to that end goal. As you get towards later stages, it can't be that way. It has to be a decision that's made for the good of the business, irrespective of the fact that you, you know that you're going to upset some people, and that just has to be a given. Of course, you can, you can work with people afterwards and sit down and say, "Look, this is why we made the decision. We'll carry…" So you have to carry the minds with you. But if you relate that back to that point about not having an open door anymore, how do you line those two things up together? Because- Yeah … you're not spending your time afterwards. But the goal now is to be successful. The goal is to get the value that this company can deliver, and the benefits. One of the biggest reasons we have some great people here is we are creating huge benefits for the world, for the people of the world, and you've got to think about that as well. I've got to make this hard decision because actually this molecular sensor that we want to get into testing potassium could save people's lives from kidney disease. So I don't want to get stuck in the decision now because somebody wants to make the PCB pink and not, not green. No, it's green. Done. Go."No, but I like pink." I don't care. Um, being a bit flippant about it, but it's, it's… You have to realize that particularly in the, in the strategic leadership positions, speed of decision is critical because- Hmm if it's not fast, the time you waste here gets magnified, and the time that gets wasted i- in, in the tasks that are being done to create the product or the technology, so. Yep. It's, it's a difficult… When you're very people focused, it's a very difficult transition. Yeah. Yeah, I'm sure, and especially when you've been the founder and you've built the team. Exactly. Exactly. Great. Simon, thank you so much. This has been wonderful. This has been- Yes, thank you for the opportunity. Really, really appreciate it. What a journey. What a journey. Thank you so much for, for, uh, for sharing all of this with us, your own personal journey and Paragraf's journey as well, and, uh, for really, you know, so, you know, for so clearly highlighting, uh, you know, the different points along the way, including the challenges. Uh, there's a lot of learning here, so, um, thank you very much. I get the feeling it's not finished yet. I'm sure it isn't. The, the, the learning, that is. I'm sure it isn't. Thank you again. All right. Thanks, Chris. Cheers. You've been listening to the Lab to Market Leadership Podcast, brought to you by Deep Tech Leaders. This podcast has been produced by Beauxhaus. You can find out more about us on LinkedIn, Spotify, Apple, or wherever you get your podcasts.