Episode 49 Transcript: Cleaning up the chemical industry

Molly Wood:

Welcome to Everybody in the Pool, the podcast where we dive deep into the innovative solutions and the brilliant minds who are tackling the climate crisis head-on. I'm Molly Wood.

Molly Wood:

Today, we’re plunging into the complex and often invisible world of chemicals … which are … obviously … everywhere … our products, our homes, our bathrooms, our bodies …

Molly Wood:

And the chemical industry … is responsible for at least 5 and as much as 10 percent of global emissions, largely because it relies heavily on fossil fuel feedstocks and energy. It’s a very tough … and very science-y nut to crack.

Molly Wood:

There are innovations in clean chemistry … from directly electrifying the production process … to creating NEW systems using synthetic biology …

Molly Wood:

Or … what we’ll talk about today … using electrochemistry to produce chemicals without the heavy carbon footprint.

Molly Wood:

Yes … we’ll explain … well … HE will …

Jeff Erhardt:

My name is Jeff Erhardt. I'm the CEO of Mattiq. And we're in the process of building what we think of as the world's most technologically advanced clean chemistry company. And specifically, we're really on a mission to tackle what's considered one of the big three hard to abate sectors within industry. 

People talk about cement, steel, and then chemicals writ large as these opportunities to leverage and go beyond what we've done, electrifying things like transportation, cleaning up power generation. But they're very challenging. And so we're going after that third one with some of what we believe is very novel approaches.

Molly Wood:

So let's start with the sector because the other ones I think are visible, whereas chemicals are a thing that surrounds us, is inside of us, is all around, and just that we're not necessarily thinking of. So talk about the sector itself and what climate impact it has and why it's so hard to abate.

Jeff Erhardt:

Yeah, so I think that's exactly right. And sort of the way I describe it sometimes is it's the things that build our world and the things we take for granted, whether it's fuels, whether it's the things that become plastics, or nylon or paint adhesives. These things are everywhere. And we don't think about them all the time. 

We take them for granted, but they're incredibly important. But they also are generated and produced today, largely using fossil fuel feedstocks and fossil fuel energy. And the result is, in addition to being very concentrated in their production, they are also responsible for probably approaching 10% of global emissions. And so the question is, how can we start to both reduce that climate impact by reducing the direct emissions, starting to drive towards creating a circular type of economy on those. 

And then how can we start to address some of the newer challenges around supply chain and some of the geopolitical issues that have come up recently sort of post pandemic time. So those are sort of all three challenges of why or what the opportunity is to start to decarbonize that chemicals and fuels industry.

Molly Wood:

Let's dig into the kind of individual pieces that you've talked about here. Let's maybe start with feedstock. Like which parts of those three challenges or three kind of opportunity buckets are you touching? Are you novel materials? Are you, run us through kind of the parts of the problem that Matique is tackling or all.

Jeff Erhardt:

Yeah, so what we're doing, so if we think about the opportunity to start to decarbonize the production of all of these things, there is, the way we think about the opportunity is there's a couple different ways that you can do that. One that has been quite hot for a while, is very interesting important is the use of synthetic biology, using microbes to start to convert again a feedstock or starting material into that chemical or fuel of interest. 

That's very interesting. That's companies like Lonsatec, quite well known for doing so. The other opportunity is you could start to say, okay, can I keep the same thermal processes that exist within these large sort of incumbent players? And can I convert that thermal heat into electrical heat? That's pretty interesting. But the third opportunity you could ask is, well, can I produce just like we've electrified everything else, we've electrified transportation, we've electrified many other parts of the economy, can we directly electrify the production of these materials? And the answer is yes. 

And so one of the sort of prominent examples of where that's done is the production of hydrogen. So hydrogen used all over the place, many, many different things, but one of the big potentials is as a clean fuel, for example, use in hydrogen fuel cells, as an industrial heat source or fuel source, et cetera. And over the past several years, there's been a tremendous effort in the production or development of capability of what's called green hydrogen. 

And green hydrogen is basically using electricity to start with water and to split that water into hydrogen and oxygen. And so the underlying technology that's used for that is called electrolysis or electrochemistry. And there's a tremendous opportunity in that market and it's starting to grow and there's tremendous funding for it, incentives in the US. They announced these hydrogen hubs recently.

But very importantly, there's an opportunity to not just use that same underlying technology platform for the production of so-called clean or green hydrogen, but to use that same type of architecture, that same type of technology, i.e. electrolysers and electrochemistry, and be able to apply it to solving this other problem we were just talking about, which is the clean production of an entire array of chemicals and fuels.

Molly Wood:

Right. So what, so, and I'm guessing based on your setup that you're in the business of number three. And so which parts of the kind of chemical production stack can be electrified in that way? We'll just continue to use that shortcut.

Jeff Erhardt:

Yeah, absolutely. So use that as the as the framework. So the way I like to think about this is the production of hydrogen using electrochemistry is you've got a magic black box and that magic black box has a couple parts. It has an anode, it has a cathode, and it's got a membrane in between. 

And on that membrane is a magic material called the catalyst. And the catalyst is the driver or really the enabler of a particular chemical reaction. And so the opportunity at what's going on there is to optimize that overall system. Each of those parts, what is that catalyst material? How does it interact with the membrane? 

What are the operating conditions? What is that feedstock? What is the, for example, electrolyte that's used within that system? That is a very highly dimensional space. And much like other industries where people have started to think about, how do I use the combination of experimentation. Sometimes people use the words combinatorial or massively parallel experimentation, coupled with the generation of large amounts of data plus math in the form of statistics, machine learning or AI, to bring those together to solve a problem. 

One of the most prominent examples of that is in the drug development and biotech industry. And you could think about it, you could use the analogy of following what's happened in that space. where traditionally drugs were developed using, one at a time experimentation in a very monolithic way. And then came in the genomics and the gene chip guys, where they could put on a chip and they could synthesize a pretty big number, thousands of proteins that could then be characterized and studied and put through a drug development pipeline to really accelerate the discovery and time to market of new solutions in that field. 

Effectively what we've done, and our core technology, the platform that we built and how we're solving this problem is a very similar way. But instead of doing it in the context of biotech proteins and drug discovery, we're doing it in the context of inorganic nanomaterials that can act as catalysts to drive these reactions in the context of these electrolyzer type systems that can then produce this very broad array of chemicals and fuels. 

And so what we do is you can think about us sort of like Genentech when they were being developed in the, what was that, the early 90s coming of sort of age relative to the traditional pharma companies like Roche, we have a similar approach that could be contrasted or compared to companies like BASF or the other chemical majors. So really transforming that process.

Molly Wood:

I love that analogy. And yes, to put a finer point on it, you're effectively a platform and a process, a new system that is relatively chemical or fuel agnostic. Is that fair to say?

Jeff Erhardt:

That is right, exactly. That is very fair. We started with solving one of the key bottlenecks within that hydrogen space we were just talking about. And if you look at that, the NREL, National Renewable Energy Labs, just released their updated cost analysis of driving down the cost of that clean hydrogen production, which today is not economically viable, right? 

It's actually very hard to split water using electricity and then produce enough hydrogen to meet the world's demands. And so right now they're relying on quite a bit of subsidies around the production of that.

Molly Wood:

Can I back you up and ask you a little bit more about why it's so hard?

Jeff Erhardt:

Please. Ah, exactly. Well, so there's two aspects of it. I think in the hydrogen case, so the way we think about this is that what we're doing is we're using electrons to drive a reaction. And so the very simple way that we think about this and that our scientists think about this is, how many electrons does it take to drive a particular reaction? 

And I'll use two different examples here. So water doesn't require very many electrons. That requires two electrons, but the end product is very light. And so you can think about it this way, the amount of energy or electricity that it takes to produce a ton of hydrogen, takes a ton of electrons. 

A similar example, you may have followed and heard, there's some people doing some very interesting work and a lot of interest in using carbon dioxide, CO2 as a feedstock, to be able to convert into other things like sustainable aviation fuel or ethylene, for example. That one is a slightly different challenge. That requires a lot of electrons to split the CO2, but then you get pretty good efficiency in terms of the weight of the end product. But the impact is the same, meaning that both of those things require a large, the so-called green premium. relative to the status quo and effectively relying on something. 

And that's okay, right? We'll get there, government is there to help incentivize those things. But separate that from an entire, in our analysis, an entire other universe of these chemicals and fuels that can be produced much more efficiently using fewer electrons per unit of output, coupled with the opportunity to really optimize a holistic system to be able to deliver those. And that's really where we're focused. is the opportunity to decarbonize all these building blocks in our world in a way that doesn't assume that people are going to be willing to pay more.

Molly Wood:

All right. Are you with me? We’re gonna take a quick break and when we come back we’ll talk about what chemicals Mattiq hopes to be replacing in the near term … on its way to becoming the Genentech … of the chemicals industry. Oh also nanotechnology. YESSS … let’s go.

Molly Wood:

Welcome back to Everybody in the Pool. We’re talking with Jeff Erhardt … CEO of Mattiq … a clean chemicals company that’s trying to decarbonize the chemicals the world uses every day … which leads to kind of an obvious question … where to even begin??

Molly Wood:

Are there top targets that you have? I mean, we've talked about hydrogen. Like, we've talked about fuel. I wonder, are there, there must be a list of fuels and or chemicals that are in such broad use and that this is quickly applicable to, right?

Jeff Erhardt:

There are absolutely. So yeah, so the hydrogen ones, it's an interesting, good story, right? So just to sort of put the punchline on that one and sort of lead into this next one. The reason, there's many reasons why hydrogen is expensive to produce, but one of them is one of the key material inputs is the catalyst that is used to drive it. That catalyst is a very rare earth element called iridium. 

It's concentrated in just a couple of countries around the world. And there's kind of a race going on to see who can capture the most of it. And so our work there was really to focus on how can we start to discover and design alternatives to that very rare element called iridium to be able to drive this reaction and help push that down the cost curve. 

Again, taking advantage of the fact that these systems need to be fully and completely integrated. Now, fast forward to some of those other materials that you asked about. So now let's take that same concept where now we need to care not just about the catalyst itself, but again, the design of this overall system to be able to drive towards being both clean, economically viable, and important, the last thing that I'll give you, and this will tee up some of the examples, is the ability to be modular and distributed. 

One of the great benefits of this underlying technology is that we don't rely on building massive monolithic factories with centralized production to be able to produce these chemicals. and we can start to drive towards smaller scale point of use type of production. So where does that apply? That can apply to a number of different things that are quite interesting. So let's take for example, hydrogen peroxide is produced in that way. 

It falls in the category that you said, okay, well, interesting, but how useful and how broad is it? It's used all over the place. It's the reason why you get your paper towels and they're white, the use of the bleaching agent used in all kinds of other things. How is that produced? It's produced in a massive centralized facility. But the problem is shipping is very dangerous. And so when they ship it, they dilute it back with something like 90% water to deliver to the end user, who then re-concentrates this again back to the state it needs to be used in. 

And so what have you done? Not only did you produce it in a not clean way, you're basically shipping water all over the place. You could say the same thing about ammonia. One of the world's most important chemicals uses a fertilizer starting to be considered as a fuel in some parts of the world, particularly Southeast Asia. Again, producing these massive, highly concentrated facilities, shipping is very difficult. Same story. 

You could ask the question, what if I was able to develop a system to produce ammonia using clean electricity, you know, maybe off the grid, solar wind, something like that, at the point of use where I need it, where it's gonna then be used and converted into fertilizer. And so those are two of the massive examples, two moonshot type of bets. But then you can start to get to all the ones that you mentioned before, these things that go into things like adhesives, paint. you know, acrylic acid, acetic acid, you know, ethylene, which is used in your plastic water bottles or a precursor to do those. 

All of those have the potential to be produced in a similar way with similar benefits, reducing, you know, their direct emissions, helping on, you know, transportation logistics, supply chain decoupling, and then starting to give other benefits like safety. within a plant operation, eliminating very dangerous intermediates like ethylene oxide, for example, which is effectively a bomb within some of these factors. So all of those together give the opportunity for both us as a company and the industry more broadly to start to tackle as it starts to move, as I like to say, beyond hydrogen as the sort of clean chemical and fuels that we're going to produce.

Molly Wood:

I've seen like bench scale just to even make it more concrete. I've seen, you know, bench scale examples of an electrolysis process that would create ammonia. And so you could imagine much more specifically a little solar powered electrolyzer in a farm on it in a field, right? Like we're literally talking about it's in a field, the fertilizer is produced on site. It doesn't have to get shipped anywhere or arguably even bought from anyone.

Jeff Erhardt:

That's exactly right. you hit the nail on the head. So now take that template, apply it not just to that, but to other similar things as well. And then the question becomes great. Can you do it in a way that is economically viable? And that's really the focus and the process that we're solving is that because the development process for these existing electrochemical systems is so complex, so antiquated, so siloed. people are ending up with suboptimal solutions. 

And so that's what we're doing. We're seizing the opportunity that you just mentioned and then taking out that green premium so that there's effectively no excuse to not be able to deploy these systems all over them.

Molly Wood:

And then let's go back to iridium for a second. You are synthesizing iridium. Tell me a little more about that. And it involves my favorite long lost topic, nanotechnology. What happened to nanotechnology? You're still doing it? Yeah.

Jeff Erhardt:

Oh, absolutely. Nanotechnology just went on hiatus for a little bit. So, no, it's one of those ones. Look, it's what's the current, let's call it, hype cycle that we're in, AI. And we'll come back and talk about that one. Exactly. Sort of joking about that aside, where does this go? And where did nanotechnology go? It started to just become everywhere. 

Just like eventually, you know, AI is not going to be a new separate thing by itself, it will be, you know, taken for granted and embedded within all of these things that we build, we use in our daily lives. So let's come back and talk about it. So are we synthesizing a iridium? The answer is not exactly, but what we are doing is we are synthesizing alternative nanomaterials that behave in similar ways to iridium.

Molly Wood:

Got it.

Jeff Erhardt:

So effectively what we've done and sort of the core technology of our company that was developed at Northwestern University was exactly that. It was the ability to combine and effectively do millions of experiments at a single time, combining different elements together and then studying their characteristics. When you start to combine things, when you shrink them down, their behavior is not always predictable. And so that's what we use. 

And that's what we did with this iridium problem was, okay, instead of allowing a single smart scientist to come in, read some papers, design an experiment and say, I'm gonna try to design this catalyst material that behaves as well as iridium, we're just gonna do millions of experiments at a single time. But importantly, millions of experiments that don't just exist in a simulated or in the digital world, millions of experiments that exist in the real world. 

And so it gives us the ability to start to solve that gap or that challenge between by definition, models and predictions, especially in the physical sciences world are by definition wrong. They have a large error bar around them. And so people spend a lot of time trying to simulate or do predictions at the sort of atomic scale and then they duplicate them in the real world and it doesn't work very well. 

We flip that paradigm on its head. We start with the massive number of experiments in the real world collect the data off of that, and then start to use that data to build intelligent systems to accelerate the overall process of identifying those interesting materials, how they should be integrated into these electrochemical systems and driving towards an optimal.

Molly Wood:

So there are multiple technology challenges inherent in what you've described. There's the initial technology challenge of devising tests that are going to produce good data. And then there's the testing, and then there's the analysis of the data. And then finally, you get to build great game-changing solutions on top of the results of this testing and data. And you at Matic are doing all of those, not the fourth part, right? You're doing the three.

Jeff Erhardt:

We are doing all three of those. That's right.

Molly Wood:

Right.

Jeff Erhardt:

We are doing all three of those. That's right. Think about us doing three things. We are doing real world development, let's call it synthesis, characterization, and systems development of materials and electrochemical systems, which in turn generates very large amounts of high quality verified in the real world data. that then allows us to apply math on top of it in the form of machine learning or AI to really accelerate that entire process. 

So you can think of step one being that, right? Is, okay, we've generated so much data, it's hard for people to sift through. How can we start to sift through it, interpret that better? And then as we get the flywheel going and get the loop going there, start to make better recommendations for what the next best experiment could be. 

That's step one. Then step two is, okay, as we start to think about, you said, you know, 15 minutes ago, boy, what we're doing is really interesting is we're not just tackling, say, hydrogen by itself, but what we've done is we've developed a platform, and again, the equivalent of sort of a drug development pipeline, to be able to tackle a large array of these different chemicals and fuels. 

Where it starts to get really interesting is then starting to say, how can we not just optimize within this one reaction we're thinking about, but now can how we start to make better predictions about if we have started to master the conversion, say, of CO2 into ethylene, what if we learned about that to talk about the conversion of CO2 into something else, right? Or into a third one, where we can start to do what I think of almost as synthetic discovery built on top of real world data, not just simulated data itself.

Molly Wood:

Now you’re probably already realizing this involves AI … and this is where I could not help myself and I had to ask about quantum computing because of having Pete Shadbolt and PsiQuantum on the show recently … so just ignore me because that’s really not the point … of course … the point is … how do we get people to ADOPT … the new processes … that companies like Mattiq are creating …

Jeff Erhardt:

We have right now the major tools that we need. We've got the physical sciences, experimentation, combinatorial, massive repair. We've got the AI and we've got the systems development that was really driven by the hydrogen guys. Our opportunity as a company now is, how do we bring those together? How do we start to develop, like you said, through those lab scale pilots, proof of concepts, but then really what's the scaling opportunity to be able to implement this, to make an impact at the scale and timeframe that the world needs.

Molly Wood:

Right. Which was in fact my non-quantum computing question, which is then what happens, right? After you've created this platform and the drug discovery, what are the barriers to that mass adoption and implementation? Is it like a workflow thing or is it just inertia? It's all of that.

Jeff Erhardt:

I wouldn't say it's workflow per se, really in the short term, it's inertia, right? And it's, you know, let's call it economics. That's exactly right. Which is why our whole mantra is we are doing the commercially viable decarbonization of these building blocks of our world. Because if we cannot do these things in a way that drives towards a limiting that green premium, it will be too easy for the incumbents to just embrace the status quo.

Molly Wood:

Right. So once it's commercially viable, I mean, sometimes even then, right? Like VHS still defeated Betamax, even though Betamax was better technology.

Jeff Erhardt:That's right.

Molly Wood:

It’s a long process … as you can imagine … turning this ship probably means incumbents adopting all kinds of solutions … INCLUDING this new type of electrochemistry and chemicals discovery platform they’re building …

Molly Wood:

But the opportunity to decarbonize this industry … is extremely large … some companies might have to collaborate more than they’re used to …

Molly Wood:

But Jeff also pointed out that there’s an opportunity to re-industrialize parts of the country that USED to be huge chemicals producers … from Houston to Chicago to the upper Midwest …

Molly Wood:

So I wondered …

Molly Wood:

Right. And then finally, this is fairly broad. But what is their need? What is the lever that's going to push them to change? Maybe it's not a need. Maybe it is, in fact, a stick. But there must be a carrot also.

Jeff Erhardt:

Yeah. So I think, yeah, the answer is, but you did it exactly right. Two sides of the same coin. There will be a carrot and a stick. So maybe to start with the stick is the obvious ones. It's, there is likely to be increasing regulation, incentives around, you saw, for example, 3M and the challenges that they've had in these so-called fluorinated chemicals, the so-called PFAS. That is one example. 

But there is similar things that are going to happen in these other processes produced in these legacy very dirty ways. And if they came for them, as the saying goes, they're going to come for you too. And that's from sort of the government, the regulation standpoint, putting aside any efforts to work on. a price of carbon or trading of carbon itself. There's simply the direct environmental damage that comes out of these things. 

Then there's some private aspects. There was a very, I think the Times on the front page a couple months ago had Michael Bloomberg as his next sort of one of his next things to go after is the chemical industry. And specifically, yeah, I'm gonna come after you to try to clean this up as well. So that's the stick side. The carrot side is some of the things that we were talking about earlier. which is, okay, let's take and put aside the fact that, evidence mostly says consumers are not willing to pay more for something just because it's sustainable, right? I think plenty of sort of examples now have shown the difference between stated preferences and revealed preferences in that. Put that aside.

Molly Wood:

Smugness still, you still cannot put a price on smug, but leaving that aside.

Jeff Erhardt:

That's exactly right. I agree, I agree. But leaving that aside, it really comes back to what we were chatting about a few minutes ago, which is developing these systems in this way beyond being good for the planet, okay? Beyond driving towards being competitively viable, beyond starting to provide risk reduction for you as a company that you're gonna get, you know, sort of, they're gonna come after you, so to speak. 

There's all those advantages in terms of being able to do modular distributed production on site at point of use. There is safety benefits as we talked about, which saves you money directly, which saves you money, right? They are safer, right? Which saves you money, right? Insurance, risk, et cetera. So the answer is that's what we're driving towards. And you just have to overcome this hurdle or inertia to change. Which saves you money. We should even put a finer point at which saves you money. So economics, again.

Jeff Erhardt:

That's right. And economics, again. Yep. of these existing systems, but it's going to happen. And we're starting to see it happen in these other spaces.

Molly Wood:

Say it with me, friends … we can … solve … hard problems …

Molly Wood:

One thing at a time. Or … you know … maybe a hundred or even a thousand or a million things at a time. The point is … we can.

Molly Wood:

And that's it for this episode of Everybody in the Pool. Thank you so much for listening.

Molly Wood:

Email me your thoughts and suggestions to in at everybody in the pool dot com and find all the latest episodes and more at everybody in the pool dot com, the website. And if you want to become a subscriber and get an ad free version of the show, hit the link in the description in your podcast app of choice.

Molly Wood:

Thank you to those of you who already have. See you next week.