speaker-1: But overall it could be four years easily in total.
speaker-2: given your experience, they most of these programs
speaker-1: It's anywhere from cell manufacturing back up the chain. think highly regulated industry where it can take two or three years to qualify and validate a product with the end customers. If you think about having two multi-story car parks and one of them is the cathode and one is the anode and the lithium ions are basically the cars which can be stored in either car park and have to move between them. And as things scale up, things change. So mixing patterns in vessels, tanks change. Heat transfer changes.
speaker-2: which is more of a big boys game rather than a start ups game. ⁓
speaker-1: I think it is, yes, largely. think it's very difficult for startups to break in. Getting to mass production is very tricky. Yeah, I don't have a crystal ball, but I guess looking at the general trends and know, AI is going to replace any of those anytime soon. It's not the case of an AI till predicting that. The proof of the pudding is in the eating. You you only know how good the pudding is when you eat it. So you've got to do the eating.
speaker-3: You only know
speaker-2: The failure you learn most
speaker-1: Always get by in and communicate. I learned that one hard way.
speaker-2: Welcome to the SP show. Today I'm joined by Stuart Johnson, founder and director at Solid Process Limited, a UK based firm. Stuart has over 30 years of experience in process engineering with a track record of turning complex industrial problems into repeatable scalable systems. In this episode, we'll get into the battery value chain with a practical lens, which battery chemistries are likely to matter the most over the next decade, what regulation will actually change on the ground, and where AI can genuinely speed up the material development and industrialization versus the where it is mostly the noise. Stuart, welcome to the show. Great to have you here. Can you please tell us briefly about your 30 years long career journey?
speaker-1: Thanks Sri, great to be here. Sure. Yeah, I started with really a chemical engineering degree at the University of Leeds followed by PhD. And then the following 30 or so years in the chemical industry in pigments, catalysts and battery materials for the last 10 years or so. And most of that's been in process development and business improvement and also some time in the commercial area in procurement for about five years. and also in process licensing. So quite a varied mix of things during my career. And a couple of years ago in 2023, I set up my own consultancy.
speaker-2: you very much. mean, 30 years is quite long. Let's try to dig something about what you have done and how it's helpful for batteries itself. So as you mentioned, you started your own consulting firm. So what is the consulting firm is about and why you started this?
speaker-1: Okay, well, the last company I was working for was coming to an end and I was looking kind of what to do next. Some colleagues had moved into the consultancy area, including yourself, Sri. So it seemed like a good opportunity to do something a little bit different from the corporate life. And really I wanted to use the experience that I had to help other companies in the battery industry. I'm very interested in the green sector. So that's really where I wanted to stay.
speaker-2: So what exactly you're considering from Solve process offers to its plying.
speaker-1: So really what we do is look at helping companies in the mid part of the battery materials value chain to help them to scale up, to develop their processes, to move towards commercialization, solve problems with technology or scale up and a host of other issues as well. But those are the main things.
speaker-2: getting back to the batteries, what exactly pulled you into the batteries originally and then what is keeping you still today here.
speaker-1: Really what pulled me in the first place, I was working for Johnson Matthew who were a large FTSE 100 company, actually in the catalyst division, the process catalysts, which are used in industrial processes. And I was there while they were setting up a battery materials division. And I happened to be in a meeting with the CTO at that time, Shirley Brown, and she basically head hunted me into the battery materials division, which was right at its infancy. ⁓ so that was my start and then worked, worked for several years after that in the new division. And what, what keeps me working in it is really, it's so interesting. It's a very complex and dynamic industry. mentioned earlier, I'm very interested in the green transition. That's the underlying purpose really for being in this part of the industry and also the people I work with and the places that I get to visit. It's really interesting pieces of work that I get to do.
speaker-2: So it's quite a quite obvious transition right from catalyst technologies to the batteries. Given the same context, can you please map out for us that what is the what is the meaning of battery value change?
speaker-1: Sure. And it's, it's quite a long value chain. So it starts right back at in mining. So I'll talk our way through the chain and then we can give a little bit of detail of each stage as well for a bit of context. the value chain starts with mining, mining of ores from the ground. And then those, the metals, the key metals are extracted and purified. And here we're talking about base metals like nickel, cobalt, manganese. Those are the basic building blocks. So we need pure forms of those metals from the ores. I mentioned those metals, but also the similar value chain for iron based materials to make LFP, lithium iron phosphate as well. But sticking to the base metal battery materials, those building blocks are made into what's called active materials. And that usually happens in two steps. So there's a precursor stage and then there's a cathode active material stage. And then those cathode active materials are made into batteries or cells. And then those individual cells themselves are built into a pack. It's a battery pack, which also has some electronics, which is known as the battery management system and cooling systems and a structural architecture to hold everything together. And then those packs are what are used in electric vehicles or EVs and also other energy storage applications for stationary storage, for example. And then at the end of the value chain, those packs, say from EVs can be used in, can be reused in other applications such as EV packs being used in stationary storage. And then when they're finally done, they can be dismantled and those metals can be recycled. So that's, there's a whole recycling loop back to the start of the, the chain again.
speaker-2: Okay. So going from the mining all the way across the recycling, getting back to again into the value chain. Given, given your experience, just I want to ask you this point that where most of these programs die. Because as you see it nowadays, have seen a lot of negative news around the value chain. So particularly around the, I would say, cell manufacturing side. But, based on your experience, where most of these programs die off.
speaker-1: I think it's anywhere from cell manufacturing back up the chain. think what most people don't realize is how tightly linked the elements of the chain are. So there's a very close relationship from the vehicle manufacturing, we call the OEMs, original equipment manufacturers, to the cell manufacturers, to the active material manufacturers, and even back into the battery chemicals. So all of those links have to be really close and we're talking about highly regulated industry where it can take two or three years to qualify and validate a product with the end customer. So it starts from the EV manufacturer backwards. it's very unlike a commodity chemicals industry where the customers can easily switch between suppliers if they're not happy because each product is specifically tailored for the customer's requirements and the... And then the production line which makes it also has to be approved by the OEM. So this is going right back to battery materials, the active materials, which are made from the battery chemicals. So a supplier therefore can't plan to build a factory to produce a generic precursor or cathode active material. say in two years time, without that material in process being well defined at the outset, they have to be very sure about what that's going to be. And that has to be backed up with an offtake agreement. And before that happens, it has to be two or three years worth of development and qualification work to provide samples and get the customer to buy into that. So there's lots of testing involved and lots of costs. So where things can fall down is where there's technical challenges occur along that value chain. ⁓ Or maybe there's cost or project overruns, which can cause issues. If things change, then everybody else is sort of dragged into that. And this means there's a huge load of barriers to entry for startups because the long time scales before getting to revenue, very high capex levels and very high learning curves, particularly around scale up. So we're talking about people scaling from lab up to full scale. number of steps to go through and a hugely challenging process to achieve that full scale production.
speaker-2: You mentioned there are different sorts of components out there, In a particular battery, for example, is it possible to slightly dig into the details of, for example, what is the role of a cathode or anode or a separator or electrolyte in a very simple lithium ion battery?
speaker-1: Sure, I think probably the most helpful analogy is that of multi-storey car parks. It might sound a bit odd to talk about car parks when discussing batteries, but if you think about having two multi-storey car parks and one of them is the cathode and one is the anode and the lithium ions are basically the cars which can be stored in either car park and have to move between them and the car parks really are essentially the layers within the battery materials within the particles. So that's where the lithium ions are sitting on each side of the battery. So during charging, lithium gets pushed from the cathode across to the anode. It comes out of the car park and goes across and into the other side. And then when the battery is being discharged or used, then it goes back the other way. It's reversible. And the lithium actually travels through the electrolyte, which is usually liquid, although people are developing solid electrolytes at the moment. And then between those two car parks is what's called a separator, which allows the lithium ions to pass through. but it also provides electrical insulation between the two sides. So the failure mechanisms that you asked about Sri, that's really where the car parks start to fall apart. So that's through things like particle breakage, where you can get layers of things building up around the particles or the car parks that's causing a bit like congestion around the car parks and the lithium can't get through. And also loss of lithium. So lithium is somehow getting trapped and that can be in various areas in the cell. And there was also quite a sudden failure which can occur when the separator can get punctured by things called dendrites, which think about things like stalactites or stalagmites in caves. Those are basically like a spike which can puncture the separator and create a short circuit with inside the cell. So that can be catastrophic to cell as well. And there are many issues during cell building itself. So as the cells being produced, you can get things like metal fragments coming in, which can create short circuits as well. You can get lack of alignment as the cells built, which can create issues with basically electricity flowing where it shouldn't be flowing. So that process has to be extremely high precision and high purity.
speaker-2: So just on the same point, among these three or four components, we mentioned cathode, anode, separator, and electrolyte. So which one is having the slightly higher probability of breakage, for example, which starts quite soon or very frequently happens?
speaker-1: Generally most cell degradation is due to cathode, cathodic material degradation. So that's usually through essentially cracking or pulverizing of the particles, which occurs during cycling because they slightly expand the contracturing cycling and over time that creates cracks which propagate. And so the material gradually degrades. That breakage can also occur when doing a process called calendaring, is when the, during the formation of the cell. The cathode layer is basically squeezed together onto a layer of aluminium underneath with a roller under high pressure. And that can also break some of the particles. there's a few different mechanisms, but those are generally the main ones.
speaker-2: So we often hear the words such as SCI, CEI, of key words. So can you please explain us, plain language, would say, what are these SCI and CEI?
speaker-1: Sure, SEI means solid electrolyte interface. It's really the interface layer between the particles and the electrolyte. At the moment, that's usually between solid and liquid. And generally what happens at that interface is you can get some side reactions which occur and you start to build up layers of material around the surface of the particle, which start to create resistance essentially to lithium ions moving. in and out of the solid. One of the main components within the SCI is lithium. So that means that lithium is trapped in that layer and can't participate in the electrochemical reactions occurring within the cell. Effectively, the SCI is going back to the car park analogy. It's a bit like blocking the car park entrances and exits around the particles. You can do various tests to check. like post-mortem analysis, you can pull the cell apart and have a look at particles. And you can also do some specialist testing without tearing the cell apart, such as cyclic voltammetry. Well, the cell is in cell testers. Through that, you can detect the SEI. Having said all of that, the SEI is somewhat inevitable. It's generally created during fast charging mostly because that tends to trigger more side reactions, but it's not completely undesirable. So it does provide some surface protection. The surface is generally, particularly in high nickel materials, is where a lot of side reactions do occur. So actually you want to build up a little bit of an SEI to protect that material from further side reactions.
speaker-2: what you're saying is that sometimes it's, we intentionally would like to build up these kinds of layers to protect the stability of the materials of the entire sand itself.
speaker-1: Correct, yeah. And quite often, in fact, almost always, the cell maker will often do some conditioning of the cell before the cell is actually sold. So they will cycle under certain conditions to create the SEI that they're looking for before the cell is sent out. that there's actually a conditioning or preconditioning step that goes on before the cell is released to the customer, which achieves the SEI that the cell manufacturer wants to achieve.
speaker-2: So earlier in the conversation, you mentioned that there is a calendaring process. So could you please explain main steps in order in the making lithium ion battery cell, for example?
speaker-1: Sure. So the materials that we've been talking about earlier, the cathoductive materials, those are made through something called precursor. So the precursor is a speciality powder with a very precise size and shape and chemistry. So the precursor is generally, if we're talking about nickel, generally nickel-based materials, that's a hydroxide material. That's blended with lithium and then it's fired at high temperature to create the cathoductive material. And also during that process, often coatings and dopants are added to enhance performance. So that cathoducting material is really the material that's sent to the cell manufacturer for making cells. And that's taken and mixed with other things like binders and conductive carbon and solvent. And it's made into a paste or a slurry. And then that's spread in a very thin layer onto a layer of aluminium. It's about the width of a human hair, so 60, 70 micron, something like that. So it spread us that paste on the aluminium foil and then that's dried. And then that's layer which is used as a cathode in the cell. And then a very similar thing happens on either side of the battery, the negative side with the anode material, which is generally at the moment, it's mostly graphite, but you're seeing some of the materials come in like silicon. base materials as well. And that's onto a copper foil instead of an aluminium foil. So those two parts, the anode and the cathode are the basic building blocks and then those are taken and they're built into something a bit like a sandwich inside the cell. they're put together with the separator, the roller or folded into certain shape, whether it's cylindrical cell or a prismatic or pouch cell. And then the cell is filled with electrolyte and then sealed up. So that process is extremely precise. As I mentioned earlier, you've got very tight tolerances for everything, extremely high purity. So if you think about something like silicon chip manufacturing, it's something along those lines, you know, to achieve the purities and tolerances that you need to make a good cell. And of course, not forgetting we... That requires extremely experienced and skilled people to do that as well as the machinery.
speaker-2: I had to manufacture the silicon layers manufacturing is quite interesting. may I ask you to mention that you usually coat the cathode material on aluminum foil, right? Whereas the anode material, if I understood right, it's on the copper foil. Why it is, why one is on the aluminum, other one is on the copper?
speaker-1: It's to do with the different potentials of those materials and the reactivity of those within the cells. it's, to be honest, I don't completely understand all the material science aspects of it, but that's, it's about the, the, the relative potentials of those metals and the effectively of those within, the electrolyte and at those voltages that each of them are seeing.
speaker-2: OK, yeah, I would like to slightly bring you back to your expertise, like process engineering and process aspects of it. So we often hear that the qualification of materials, qualification of equipment, and all these sorts of things, very common language in battery industry. So what does actually material qualification means? So is there any stage gate approach that will be there, or some sample sizes, or what are the timelines? We often hear that the qualification time lapse are at least few years, know, not few months, it's few years. What does it mean actually? What is the meaning of the middle qualification?
speaker-1: Yeah, it's quite a long and difficult process, but it all really starts off back at the cell level and what the OEM, the car manufacturer, for example, wants. So they start off with ⁓ a desired size, weight, capacity, power, performance, and lifetime of the battery cell that they're looking for. And again, that has to fit into the pack that is going to go into the vehicle. So it starts at the cell level and then the cell... The cell manufacturer will talk to the CAM manufacturers about their requirements and they in turn will talk to the precursor manufacturers to see whether they can provide precursor that's going to be able to deliver the cathodactive material or CAM for that particular purpose. So the cell manufacturer will go out with those requirements to usually several suppliers and then there'll be a sampling process takes place. where those suppliers will send in samples, they'll be tested and there's usually some iteration and improvement development as they get close to hopefully what the cell manufacturer is looking for. So each material is specifically tailored for particular requirements. So people talk about say an NMC 811 as being a generic material, but there isn't really such a thing as an 811. know, it's, they're all different. and they've all got different particle sizes and perform differently. So people often talk about four stages of sampling, ABCD. So the first step we've already touched on, which is the conceptual stage, which is where small samples, usually up to maximum of 10 kilograms is sent by those suppliers to the cell manufacturer. And they will use those to make for example, coin cells or single layer pouch cells, are larger cell than the coin cell. And then there's iterative process goes on to improve. And then at the end of the A stage, the cell manufacturer will normally narrow down to one or maybe two suppliers for that material. And then you're into B, C and D stages. So stage B is really a design freeze. So the material is frozen, properties are locked in and then the supplier or perhaps suppliers will be looking to produce samples from their pilot plant up to a thousand kilograms and that this is really proof that they can scale up and so they will be using the same type of equipment they will be using at full production scale. So that would involve thousands of tests by the cell manufacturer. thousands of cells to build up the statistical confidence in the production process to prove that it's reproducible and it's statistically a very smooth process. And this is a key failure point really in that testing. The scale-up is very difficult. And then the C testing is really about process validation at the commercial scale at full production volumes once the production line is built. So this involves very intense statistical process control. So the, line is run for probably several weeks. And this is to really validate the actual production line can produce exactly what was produced from the pilot line. And then the final stage, the D sampling is the final production and validation before the, the cell manufacturer gives a go ahead for start of production. So beyond that point, after D sampling you're into real production and any changes that are made, hopefully not, but there may need to be some changes made. Those have to be validated or checked with the customer. And it may mean actually going back to B or C stages. can't be basically from that point on, but you can't change anything without approval. All of this A to D testing has to be done by the CAM and the PCAM producer in synchronization between them. So they've got to ensure that they're in sync with any samples that are being produced. the relationships have to be extremely close because there's a mutual destiny for everybody in that supply chain.
speaker-2: So you mentioned that usually there are four different sampling stages, right? So is it okay to imagine that the sample is usually coming from the either from the lab or from the pilot line to start with? Because 10 kgs may not be lab sampling, right, by the time.
speaker-1: It usually is lamp sampling. Yes. It's usually large lab samples. Typically lab reactors are capable of making maybe a few hundred grams, something like that. So it might be several of those samples blended together into a single sample or they may keep them separate, but it's normally up to one kilogram. Sometimes 500 grams is enough. It doesn't have to be as much as a kilogram.
speaker-2: Okay, so and typically the B sample is coming from definitely from pilot line.
speaker-1: Yes, usually, yes.
speaker-2: usually, and the C is more of a demo slash commercial plant.
speaker-1: Yeah, that's normally the run on the actual production line, the commercial scale line.
speaker-2: And D-sampling is definitely a full-scale commercial line. obviously, like you always hear that, you know, there is a slippage of timings. So what is the usually the program slip and which stage and if so, how many months, for example?
speaker-1: It really depends on the circumstances. So slippage can occur during equipment commissioning once the process is built, but also back from there, you know, looking at the sampling stages along any of those stages. If the product for some reason isn't, isn't right, then the whole thing has to be started again. And that can set you back several months. I think quite often we focus on equipment as, as engineers and scientists, but it's often all about the people. So it's about the experience, the skills, the motivation of those people that are in each company that really makes the difference. So the slippage can occur really for a number of technical reasons, sometimes commercial reasons as well, but also purely project reasons in terms of the delivery of equipment or commissioning of equipment. So there's a whole host of things that can trip us up as we move towards commercialization, unfortunately.
speaker-2: So in terms of the qualification timeline itself, what is based on your experience, what is the typical timeline for, let us say, an MC 811 to qualify with a new cell manufacturer, for example.
speaker-1: So typically that I mentioned the several stages, normally the A and B stages can be, let's say up to two years. And then CD obviously takes place once the line is built, which can take up to two years. So in total, they're not necessarily all in completely aligned in series because they're separated by the time it takes to build the full commercial scale facility. Overall, it could be four years easily in total.
speaker-2: Just out of curiosity, I know the answer, but just I want to hear it from you. So why it will take two years, for example, for A and B samples? For example, I opened a shop. I can produce half a kg of samples. And once I have the lab, usually I build up the pilot line as well. That needs at least a few hundred kilograms, let us say. So why you need two years to do what?
speaker-1: Okay, so normally the first stage, there are several iterative cycles. So it could be, you could be close with the material that they're looking for initially, which will be extremely lucky. You could also be in a cycle of maybe five or 10 separate samples, which all have to be tested. And that testing can take at least a month for the customer to take that and make it into coin sales and at least two or three weeks. for initial testing in coin cells and probably longer for the larger format cells. So the cycle could be from at least say two months to produce them material and have it tested initially. then any more iterations from that each will be at least a couple of months. So if you've got 10 iterations, that's probably a year before you know it. And then the same thing can happen at the the pilot scale as well where you've got again those iterations you know you may already be very good at making that material in the pilot scale from the lab but there's usually some tweaking to go on to ensure that you produce the material that you want so you've got time to make the samples And then in this case, you've got to make more cells from those samples. So the testing takes even longer, maybe two to three months per set of samples. So you've got each cycle, maybe three or four months production and testing. So you've got maybe three or four cycles of testing. That's probably another year. It's very difficult to say exactly because every situation is different, but that's how you can get to two years.
speaker-2: That means typically not only the material manufacturing timelines, but also the, what I'm hearing is the cell testing timelines are the kind of real bottlenecks slash the challenges. One has to overcome to squeeze these qualification timelines.
speaker-1: Exactly. Yes. It's not just a question of putting the sample into the lab and then you get result back in a couple of days. What the cell manufacturer has to do is firstly build the cells, which takes some time. And then the cell is put into cell cycling equipment, piece of equipment, cell testing. And then the cells are cycled up and down repeatedly for thousands of cycles. that's... that takes time. can't be speeded up. Obviously people do try and accelerate that, you know, in different ways, but ultimately the final testing has to be without that acceleration. you've got a very long process to produce and test material.
speaker-2: lot of techings came up now, Like HPCE or some sort of monitor-corner simulations, modeling, so on, you know, just trying to accelerate the testing itself. So you mentioned several times that there is a kind of, know, due diligence process goes from the cell manufacturer to the cam manufacturer, down back to the precursor manufacturer. So based on your experience, what is the, like a one trap kind of thing you're seeing always, you know? or any particular documentation you will look for to avoid those kind of traps.
speaker-1: I don't think there's any particular documentation. I think the main thing, as you go around different facilities, you build a picture of what the company is like. As engineers and scientists, often focus on equipment. But quite often it's all about the people, the experience, the skills, the motivation of those people, which drive obviously the design of the installation of the equipment in the first place. And in the battery industry, those people are very uh, skilled and quite rare and quite mobile as well. So actually the organization chart of the company and the CVs or resumes, you know, say a lot about, about the company as well. Um, and likewise, you know, when you're walking around a facility and you're looking around, you can pick up a lot, you know, in terms of the cleanliness, the, how well ordered the facility is, the design of the facility, what the people are like. ⁓ and how they respond. I think there isn't one single thing when we talk about due diligence, ultimately the product is a product by process. So everything is important in delivering that product through the process from the raw materials, through the process equipment, through the way that equipment's operated to the labs. So there isn't really any, any one thing you can pick out, but you tend to build up a a picture as you walk around somewhere, you know, how well run this facility is. And I find, you know, a good question to ask is, you know, what would you do differently? Good open question. You know, if you could change something, what would it be? And that sometimes opens up people to maybe saying a bit about what doesn't quite go right, what they would change, what they would like to improve. And it's a good opening to. to explore that with them to discover more about the company.
speaker-2: Maybe I'm taking the liberty to ask you based on your experience, right? usually what I've seen based on my experience as well, on the lab scale, you'll get the beautiful product, the things in spec. Now everything is perfect. But when you kind of try to scale it up, know, where this process engineering concept, I think, you know, comes to the picture, the pilot demo or commercial, whatever it is, that scale up stage. So what are the usual kind of, know, bottlenecks or traps that kills a lot of startups, right? Do you have any secret sauce to fix those things?
speaker-1: I wish I did. think, yeah, obviously as things scale up, things change. So mixing patterns in vessels or tanks change. Heat transfer changes the way that gases move through material inside the kilns that we use to produce a cathodactive material. That changes as you scale up to larger scale. So a lot of things change as you scale up. A lot of it is about understanding what can change and being aware of some of those traps and doing the proper testing as far as possible at large scale equipment as you can. think a lot of startups don't focus on the end result and I love the quote or the saying from Stephen Covey's book, begin with the end in mind. So it's a really long journey to get to commercial production. It's a marathon rather than a sprint. So a good question to ask is what will this look like in five years time? How will this be commercialized? Who will the competition be? You know, what will the full scale process look like and cost? And what stages do we need to go through to get there? And startups particularly don't have a lot of experience of usually that scale up going from lab to pilot to full scale. And that's why they often need some help. trap, guess, just talking about scale up is perhaps people ignoring some of the glitches that occur. know, maybe something's clogging up or maybe once in a while, the batch doesn't quite turn out as expected and people perhaps will say, we'll work that one out later when we're at larger scale, but things will just be even worse. So it's better to fully understand what's going on at the time at the smaller scale before things get larger and more expensive. and can cause greater delay. guess another trap as well is planning, which can sometimes be too idealized. We can put together a great plan, but that needs to be stress tested. What happens if, for example, the scale-up takes twice as long and costs twice as much? How is that going to affect our business plan? Is that going to make the complete plan fail? If so, then how do we prevent that? Or how do we raise more money to mitigate that? And I guess the final thing is around people again, perhaps the team that is currently in place is not the right team to take things forward. And that requires people to look forward to what people will need to be in place. And that's quite a difficult thing to predict, but the right people need to be in place at the right time and they've got to be up to speed before then. there needs to be quite a lot of career planning or organisational planning for people to be in place at the right time.
speaker-2: That's very interesting. think I'm hearing a lot more times about the people and skills, those kinds of things.
speaker-1: Yeah, yeah, it's all about the people.
speaker-2: So just on the same context, do you say based on experience so far, particularly the cathode materials is more of a big boys game, actually rather than a startups game at the moment?
speaker-1: I think it is, yes, largely. think it's very difficult for startups to break in because of what we've already discussed. know, the barriers to entry, the high capital costs, the long time scale before reaching, you know, positive cashflow. Some of the technological barriers as well. Some of the people barriers, you know, maybe not being able to attract the top talent because either those people don't want to join something that's more risky or... more expensive. So I think, yeah, it does tend to be the big players dominating and particularly those with experience, you know, in Europe, there are already two big companies with that experience, Western companies plus a number of others coming in as well. it does not to say startups can't find a niche, but getting to mass production is very tricky. And unfortunately, we saw with no fault last year, know what, some of the outcome of that, it's a very hard business to be in.
speaker-2: I think when you mention about the particularly like a demo slash commercial plans, I think it needs a lot of, you know, patient capital as well, you know. And also another way, you have to wait almost like uncertainty for more than five years before you can actually see any light in the channel there. So it's quite...
speaker-1: Exactly. Absolutely. And add onto that very, very tight margins as well. You know, this isn't an industry which is, you know, being allied to the car industry who are well known for squeezing costs. The margins are not super high. So you, you know, you have to be prepared for pretty tight margins in this industry as well. So the whole thing combined together is quite a tough gig for most companies.
speaker-2: It's almost the same chain that you're coming from downstream and pushing towards upstream. it is kind of squeezing. Before digging into bit more details about the chemistries and materials. So could you please explain us what is the exact meaning of process engineering and how it is relevant to the battery industry or for that matter any chemical industry?
speaker-1: Exactly. Yeah, so process engineering is really about how to make the chemical and physical aspects of the process that you want to happen in the way that you want. So if you're talking about a reaction, for example, how to achieve the best yield and selectivity from that reaction in terms of getting the chemicals out that you want with minimum byproducts. And chemical engineering is really all about the physical design of the process, but also the conditions that you run the process at and then designing the equipment around that and all the controls to achieve that safely and cost effectively. it's basically about bringing something that you do in the lab at very small scale to a very large scale safely, cost effectively. and productively so that you can achieve the products that you want. That's a very condensed version of a four-year course usually to do a chemical engineering degree. yeah, it involves understanding the fundamentals of what's going on in the reactor, the kinetics of the reactions, the thermal side of things as well. And then all the physical changes that need to go on, heating, cooling. Boiling things off, condensing things, crystallizing things, separating things. So there's a whole host of things going on within the chemical process that all need to be very tightly controlled.
speaker-2: Being a metal chemist, kind of always underestimated the value of the process engineering actually, because it's a lot more complex than what I assume. So it's quite interesting concept itself.
speaker-1: Yeah, I think most people are surprised when they go into a chemical engineering course, there isn't actually a lot of chemistry in the course really. I mean, there is some obviously, but most of it is about physical processes and mass transfer. So heat and mass transfer are two of the core modules within every chemical engineering course. So what affects those and how you control those and achieve the best you can out of the process. with the equipment that you have or design equipment to achieve the best results.
speaker-2: But I still remember, think ⁓ long time ago when we were discussing, you always used to mention that we've done this mass heat balance. But when I saw like a scaling of the single crystals, particularly when we work together, it was quite important because the flow rate of the gases itself makes a huge impact on the performance as well as the homogeneity of the particles and so on. Right. So which process makes the most impacts, the human form of material as well as the homogeneity of the materials, particularly from the CAM perspective? And what is the, what is spec window that is non-negotiable for a process engineer.
speaker-1: Lots of questions in one question. Okay. Yeah. think, I think really to make a good cam cathode active material, really the precursor is the foundation of that. So the precursor is a very finely tuned product with a certain chemistry and particle size, which you use to make the cam. And we talked earlier about the precursors used with you take the precursor and mix it with lithium. And then that's baked to make the cam. So it's possible to make a bad cam from good precursor, but you can't make good cam from bad precursor. can, the precursor has to be good before you can make good cam. And really in terms of the precursor itself, there are basically two, two main categories of specifications. So one is chemical. What's the composition of that material? So not only the bulk composition, but the impurities. So that's one side of things. And the other side is more about the physical. attributes of that material, the particle size, particle size distribution, the crystal morphology, the porosity and the density of those particles. So there's a number of different aspects. So I mentioned earlier about product by process. It's not like most of the chemical products where you can buy a product and just look at specification on a piece of paper and say right that contains you know x percent of this y percent of that and that's okay that's specification so we can buy it. The precursor can only really be checked and tested through proper lab tests actually making that material into a battery cell and testing it so you've got to take the precursor through CAM and then into cells and that those cells has to be tested. And that's, that's quite a long drawn out process to do that. So it's not like you can tell from the specification how it's going to perform in the cell that there isn't a one-to-one relationship between those two things. You've got to do the cell testing to tell what that material is going to be like.
speaker-2: So is there any minimal viable process before shipping first pilot samples? So do you follow any kind of checklist before you ship your cam to a cell manufacturer, for example?
speaker-1: Yeah. So usually there's a number of things. So that viable process involves, you know, we talked earlier about chemical engineering, but process engineering, design of that plant has to be finalized. And that's usual chemical engineering aspects. So you've got flow sheets, equipment specifications, materials, you've got piping instrumentation diagrams, P and IDs. So there's the physical design of the plant, but then there's also the operational side, the operating procedures or SOPs, standard operating procedures, we call them. The control system has to be operational and tuned and working properly. And then we have to do a load of statistical testing by running the processor, probably several months to check that the process is in control and it's producing the product that we want. there's, there's a lot of work to do to, to get to that point before we have what we would call a viable process. And another thing which is forgotten actually is the, that pilot plant has to run using the same raw materials that we're planning to use at commercial scale. So before we get to that point, we've got to qualify those raw materials and get supply contracts in place to supply the pilot plant. So that can take several months to do as well. So there's a number of things that need to come together to have a viable pilot plant process. that can produce good quality samples for a customer. And then also, of course, we need our own lab alongside that for our own internal testing and validation of what's produced. That all has to be set up and operating properly as well, you know, to be able to samples coming off the pilot plant.
speaker-2: In that case, what you're saying is that we almost, we pre-qualifying the precursor as well for CAM. And also we need to qualify the equipment as well. It's a kind of, know, qualification process. It's part of the qualification process itself.
speaker-1: Yeah, I mean, that's more of an internal qualification to the pilot plant because at that stage when we're supplying, in this case, it will be B samples to the customer. The customer at that stage is probably going to have some degree of inspection and validation of that process just to check and confirm that they're happy with it, but it's not going to be as in-depth as the full scale line, but there will be, you know, for internal purposes, we need to make sure that whatever samples come off that production line are repeatable. and it can be a suitable quality. So we're not wasting our time and somebody else's time to test that material. So yeah, there needs to be several months of work once that pilot lines been put together to check it and check its operating as it should, you know, in control so that we confident that we can replicate that later, both from the pilot plant and also from the full scale plant. And a pilot plant is a great way of understanding what's likely to happen at larger scales as well as producing samples. The purpose of the pilot plant is to find out more about the process and check how robust that process is and what can go wrong.
speaker-2: That's very interesting discussion. And we'd like to slightly steer the gear and go more towards the actual chemistries of the chem itself. Okay. So as we know, there are multiple chemistries that are available in the market. Like NMCs, which is nickel manganese cobalt based compounds and LFPs, which is the new trend now. And LMFP, the derivatives of LFPs are also gaining a lot of traction. And also the big boy in game is the. sodium ion, where CIT are like betting a lot on it. So based on their experience, where this each category of cam will fit for next five to 10 years.
speaker-1: Yeah, I don't have a crystal ball, but I guess looking at the general trends, I'd see NMC still having a place, you know, at the high end, the longer range end of passenger EVs. Probably that is, the market share is probably declining, but the actual absolute volumes are increasing. LFP, lithium ion phosphate is continuing to improve and grow. and breaking more and more into the kind of low and medium range end of the segment. Particularly with the fourth and fifth generation LFP, the highly densified material that's coming in. So that's approaching NMC type performance overall. So LFPs made huge advances in the last few years and I think will continue to do so. That's probably on a... going to plateau out eventually. Sodium iron, yeah, you mentioned that that is the big boy in the game that's coming along. I've got some skepticism about how well that will break in. mean, that's really about the balance of pricing between nickel and lithium. So most of the sodium iron material is made using a nickel based cathode material. That may not continue, but that's where we are at the moment. And obviously nickel is fairly expensive compared to iron. So there's a disadvantage there, but obviously the advantage is that the sodium iron doesn't contain lithium. It's all about that balance and the performance. the performance of sodium iron material is fairly similar to LFP. That may increase in the same way that LFP is increasing. And I guess the other trend to mention, I guess, is Development from NMC is manganese rich. So this is a move away from the higher nickel content back to lower levels, maybe say 20 to 40 % nickel. She's starting to, that LMR, this in manganese rich material is starting to eat into some of the LFP and NMC space. One of the features of LMR is that it requires higher cell voltages. So There are some challenges that come along with that because of the changes that can occur both within the electrolyte and with manganese itself and trying to avoid manganese from leaching out of the cathode material into the electrolyte, which can occur. Manganese is quite a tricky material to manage in terms of its oxidation states. And so that material has quite a different structure. It's generally spinel type material. So again, it's all about understanding how to manage that material in the context of the cell in that whole electrochemical system that's happening within the cell itself. So that's the general overview. think the general trend is LFP increasing and sodium iron and lithium manganese rich coming in. to grab some market share over the next five years.
speaker-2: So talking about the sodium ion and also the LMR you mentioned, so what is the supply chain status for these two chemistries particularly?
speaker-1: LFP is generally China based. In fact, China restricted export of the most higher end of the fourth and fifth generation technologies to the West. it's primarily China driven LFP, although there are some other Western producers such as Nano One in Canada. That's developing. Again, it's mainly Chinese. dominated, but some in Korea as well. So LG, for example, have got an LMR offering, which looks good. I think it's a developing space. think a lot of NMC producers are looking towards LMR as a potential way to switch over production. The same facilities can basically be used with some modifications. LMR could be an option for an NMC producer already. already making NMC. So I think there are some options there, but the supply chains are like most things, primarily Asian dominated, mostly China at the moment. So it's a question of where other parts of the global economy can find ways to break into that.
speaker-2: So you mentioned that LMR is also kind of getting traction. So it is quite understood that LFP and sodium ion is kind of sharing between the EBS and ESS markets. So where do you see this LMR is going to play its role?
speaker-1: I think probably at the... towards medium to high range EVs, that's really where it's at the higher end. It's basically allowing a switch from a high nickel NMC to an LMR. So it's probably towards a higher end, ⁓ yet to be proven out in many vehicles yet, but it's starting to be released commercially. So we'll have to see where it really fits, I think because you can raise the voltage, although the specific performance of the material itself is lower, it's typically around the same level as LFP. The voltage allows you to achieve a higher level of storage per mass of material. LMR fits really at the higher end. But as I mentioned, there are some challenges with making that work in terms of the electrolyte and specific coatings that are required to stabilize the material.
speaker-2: So in terms of the ⁓ applications, right, of different chemistries, what I'm hearing is that probably three categories of applications we are looking into. One is the standard passenger EVs. It can be low-end, high-end segments. And middle one is kind of coming up, but with lot of interest from the OEMs is trucks, heavy-duty trucks particularly. Just kind of masked until you're developing the technology, of course. But third one is mostly... EIS market, but driven by the AI and its data centers requirements. So what is the kind of priorities, know, change across the energy or power, less on the material level, but more on the pack level? How do you distribute these chemistries across these three segments?
speaker-1: Yeah, it's a good question. think as with all battery technology, everything's a trade off. you improve one thing and you reduce another. So you're always looking at the trade off in every part. For the passenger EVs, really the main push, mean, safety is always a given. So safety is critical. The main push is towards longer range and faster charging. But primarily it's about range and therefore specific capacity. capacity per unit mass or per unit volume. Charging rates are also important. you know, the power, what we call the power performance of the battery has to be good as well to allow say charging within a few minutes rather than an hour. So that's always about the energy density is really the driver, but the other two need to be acceptable as well. Let's say LFP is coming to its own there really because its performance has improved and Because LFP is inherently much safer than NMC, some of the features of the pack can be modified for LFP and you can actually get more material in. So you end up with almost equivalent performance at a pack level and from LFP as from NMC because you don't need the extra safety controls that you would need for NMC. Trucks again, range is very important, but also charging. you think about a large battery like you would need in a truck, you know, The charging rate has to be sufficient within a few hours to charge such a large battery. also on the flip side, you've got the power required to accelerate a large vehicle like that. So you've got to have sufficient power performance. It's a very heavy power requirement. that's trucks. And then if we move to ESS or stationary storage, density isn't... particularly important now at all. It's all about the cost per amount of energy storing. And you you can effectively achieve a certain level of power by multiplying out the number of units by putting those in parallel. So that's not as important, the power aspect. It's all about energy storage per cost. And the safety aspect there is still important, but it's not as critical as in a vehicle, which somebody's traveling inside. And that can be controlled in other ways in a stationary storage location. So I think the other thing to remember from a high level, from a pack level is that the design of the pack is as important as design of and selection of the material itself within the cells. So a lot of aspects like thermal management are very important and the battery management system, the way that that controls. voltage in the current and can detect cells which are failing for example or problematic and that all requires a huge amount of thinking to get that right as well. So it's a very complex system because you've got thousands of these cells all linked thermally and electrically together which all have to operate as a single unit.
speaker-2: In general, we think are we over-engineering the material aspects and less focusing on the integrated system aspects? Why I'm asking actually is that particularly it's also maturity of the industry, Because that's the kind of, you're at an ancient stage where you're focusing a lot more on the, optimizing the materials and the cells, for example.
speaker-1: I think that optimization occurs at a number of levels. I think there's a level of optimization at starting at high level design. And one of our former colleagues is involved very much with that, you know, in terms of design of packs, thermal management, the battery management system, the BMS, how all those cells are placed, how we know what the cells structure are. So that's one level. And again, it's that flow back along the value chain back from packed level to cell. And then the cell, the cell designer looking at how that cell is put together and with a certain specification for that cell, therefore what the material needs to look like within that. how the, for example, the cathode material has to perform or what the volumetric and gravimetric energy densities need to be. What the voltage curves need to look like is the materials charged up and down. So there's a huge number of things that need to be thought about, that comes at a number of layers. there's pack level, cell level, and then at material level. If you think about it, the CAM and precursor manufacturer designing particles with a certain shape, size, density, porosity, crystallography. which has to deliver that final result for the battery materials. I guess it's a bit like a human body where you start from the overall structure, the size and shape of the body and how that performs. that you break it right down. It goes right down to the biology of individual cells and then within those cells, all the different structures within those. So there's a number of different layers within the battery, which are all critical to everything performing as they should just like in a body.
speaker-2: That's a very good analogy actually. So, I mean, we cannot finish the discussion without talking about the regulations, particularly if you're in Europe, where we are good at regulating the stuff. So what kind of requirements actually we need in terms of the design and the data systems around the batteries, particularly when you talk about batteries, a lot of noise around the carbon content, recycling content, battery passport, those kinds of things are coming up, particularly in Europe. So what's your take on that?
speaker-1: Sure. I think we have to distinguish between things which are really necessary for regulation for safety, for example, to provide consumer confidence in the end product, which is really important. And then things which are basically trade mechanisms to restrict or guide trade in a certain way, which will be in interests of the EU or the US or whatever it might be. And then there are things which probably naturally will happen commercially anyway. For example, supply chain due diligence will always need to happen. Any responsible supplier will always do that because they want to ensure that they're buying things from reputable sources and not from areas where there's things going on that shouldn't be going on. Recycling will always happen where it's economic because of the value of the materials in those cells. You know, it makes sense to recycle nickel, cobalt, lithium, et cetera. The carbon footprint, again, that will happen mostly due to economic reasons anyway, because the lowest cost areas are also green areas. For example, the Scandinavian countries, they've got huge amounts of hydroelectric power, which is very green. So that's where generally the industry is gravitating towards Scandinavia. Battery passports and rules of origin. That again will naturally happen because companies don't want publicity. They want good publicity about where they're sourcing material from. So I think, yes, some regulation is necessary. You you think about safety, human health, those types of things, but I think some of the other aspects will probably happen naturally anyway, just through companies behaving in the most economic way possible.
speaker-2: ⁓ So if you're given a power, what is the one decision you would change in you or in the states at the moment in order to catch up with Asia, particularly China?
speaker-0: .
speaker-1: That's a really difficult question. think, particularly in Europe, one of the barriers is the cost of energy. So that is really a barrier, particularly in the UK, but also in other parts of mainland Europe. There are energy costs. Obviously Scandinavia has got an advantage with its hydro power. So that's one thing I would certainly want to change to encourage the battery industry in Europe. Another aspect is people. And we talked about people a lot today and we don't have the people in Europe to really build the battery industry. You know, we can try and move people around. think upskilling people and actually developing people through apprenticeships, training programs, et cetera. That's all really important. And this is, I'm not talking about really academic research. It's more the industrial side and industrialization of the battery. industry that really is needed because let's face it most of the people who know how to do it are in China or other parts of Asia.
speaker-2: So I think that the energies keep coming up again and again. So why the battery manufacturing process energy intensive? Which part of the process is very highly energy intensive?
speaker-1: It's mainly the cell manufacturing and can manufacturing cathodic material. And they're beginning energy consumers, the cell production mainly because of the drying step after coating. So it uses a lot of energy to dry off the solvent. After coating, yeah. Obviously that will change with a dry cathode where there isn't the slurry. The dry coating doesn't involve the slurry so that
speaker-2: The after recording I mean.
speaker-1: that drying stage isn't required. In the CAM process, the CAM process uses very large furnaces, which are generally electrically fired. So those consume huge amounts of power as well. So that's really the main consumer within the CAM production process. Precursor, not so much. There's still quite a bit of energy use there in various pieces of equipment, but it's not as energy intensive as CAM.
speaker-2: So what you're seeing is that both those steps, we can correlate them back to the process engineering again, to really optimize those processes to reduce the energy intensity there.
speaker-1: Absolutely. The companies are already doing that. It's really when they're looking where to locate a factory, if they put it in, pick a country, put in France versus Sweden, what are the relative energy costs? And that makes a big difference because we're talking every few cents per kilo makes a difference. The operation is run on very tight margins, as we've already said. So all of that adds up, that's basically reducing the profit levels of the company. If the energy is more expensive, servicing counts.
speaker-2: So that again shows the importance of the processing engineering aspects of the whole value chain in this regard in a very general perspective. why the heat or cold, know, kill the batteries faster? Like what kind of impact it will have on the material level and how could you fix that one? Is there any kind of recipe or ⁓ secret sauce either on the material level or on the BMS side to fix these two extreme conditions of the the atmospheres.
speaker-1: Okay. Well, maybe if we take, I guess, the BMS first, the battery management system, that's the electronics, but also the cooling systems around the pack. So the cooling is really important and heating actually as well, because that system is usually used to heat. mean, some packs are air-cooled, majority now are liquid cooled. So they've got a essentially air conditioning system or not air conditioning, fluid system which circulates around through the various channels through the pack to remove or in cold weather to add heat into the battery so it can perform. I guess coming back to why the material can fail at high temperatures, it's mainly a result of side reactions. So the electrolytes, mostly liquid electrolytes, though that contains a number of chemicals which can react. The kinetic rate will increase at high temperatures. You can generate more gas at higher temperatures. You can generate more side reactions, which create more SEI type materials inside. And particularly combining that with high charging rates. If you're charging a battery quickly, you've got high current flowing through the battery, but you've also normally got a lot of heat there because the current is generating heat through resistance within the cell. The cell itself has some resistance inside. So you've got a combination of Fast lithium transfer with heat, which can create these side reactions and create gas and other deposits inside the battery. inside the cells, you've got various strategies of trying to address that, be it electrolyte additives to try and suppress some of those side reactions or coatings and dopants within the battery material itself. You're trying to again, suppress those side reactions, particularly at the surface from occurring. You've also got structural changes within the battery material itself. So you've got expansion and contraction and obviously that high temperatures that expansion is going to be greater. The material is probably slightly weaker as well. So you've got a number of changes occurring within the particles themselves and within the cell.
speaker-2: So what about the cold temperatures? Because we see a lot of chemistry slash EVs may not perform well at the very cold atmosphere, like in Sweden or Norwegian regions, particularly. So why the cold will also have impact on the batteries as well as materials from the past?
speaker-1: Yeah, again, it's kind of back to the kinetics of how quickly things happen. it's cold temperatures, everything slows down. Resistances tend to increase. So the electrolyte will be slightly more viscous. that the speed at which lithium can move, therefore through between the anode and the cathode and backwards the other way, will slow down. So essentially, it's just like everything going to slow motion really at cold temperatures and the BMS has to manage that. So it's basically going to limit the amount of ⁓ power that the pack as a whole can deliver. ⁓ to the vehicle or whatever it's attached to. So it's a bit like going to slow motion in a freezer. It's going to slow everything down.
speaker-2: Thank you. That's very helpful because just want to understand from the very simplistic perspective, course, everything, know. So other point I want to touch upon is that we are in AI era, artificial intelligence era. So how do you think this transformation is going to impact the battery value chain itself and where it will have its maximum impact, for example?
speaker-1: I guess on a number of fronts, mean, I think AI is a great thing. think this, you know, it's incredible what it can do. guess from one angle, you know, there's the power supplies required to power these data centers, which are running AI. So looking at it from that point of view, you know, the backup power supplies, the ESS systems, which will be involved in those, you know, will be, are going to be important. So that will itself consume a lot of battery material, battery cells in those ESS systems. But actually using AI itself, I found it's great for, you know, using it to refine something or even generate ideas, but we really need to watch the style because you can spot AI content a mile off. It's very obvious. It will obviously get better, but at the moment it's, it's very different from a human written article or, or a piece of, uh, piece of other writing. Um. I've been using specific profiles within, within, use Gemini generally. So I've been using something called Gemini Gems, which is within that. know you can do something similar within, within ChatGPT. So I've been able to create something which is more skeptical where it requires factual checking to, before it, before it works out and comes back with an answer. And I found that a lot better because I don't know about usury, but I found that, you know, I've got a lot of. hallucinations generated by AI and, you know, having it checking its work and going out and creating a very wide search and coming back and cross-checking facts before it presents them as facts is really important. I guess the other thing to mention is that in the battery industry, it's very secretive. So we don't see a lot of information coming out onto the internet from the battery industry. A lot of that is kept secret. trade secrets, so it's not out there. So some information is obviously contained within press releases and things like that, but there's not in terms of the technical side, particularly that's not really available on AI searches at the moment. That may change, but I tend to think not. think companies are protecting their IP in that way by keeping it off the internet.
speaker-2: I kind of agree with that in terms of the technical data, think the AI is not good. I'm not sure it's going to be improved a lot better in future because as you mentioned, more of a trade secret, it's no house than actual public floating information. The same information is coming again and again, you know, that's what these are about at the moment. But in terms of the AI utilization, do you think AI can help to accelerate the discovery process, get the new chemistries in? and also rapidified to go to markets, for example. Do you see any kind of value in the aspects of chemistry slash engineering?
speaker-1: Yeah, I think particularly with the advent of high throughput testing, you know, know companies like Navonics in the US are doing that and others, you know, in the US, for example, that I'm aware of. So I think with that quantity of data from all that test work using machine learning, which essentially is AI, being able to analyze those large data sets from multiple angles is really powerful. You can build insights, which you would never have seen before. So I think where that sort of machine Learning is brought together with human know-how as well. That's extremely powerful to be able to predict the performance of a certain formulation of material, a certain particle size, shape, cetera. That is really powerful, you know, to be able to guide research towards where you want to go.
speaker-2: So other point I want to touch upon is more about the guard rails before deploying. Because I think we might use the AI as certain aspects of it, but we have the very large data sets coming from the either from the chemistry side or from the cell testings or from the EV itself. Because as we know that the electric vehicles are more of a computer on the wheels at the moment. know, every data can be captured, refined and then utilized to improve the, you know. the safety or our performance or whatnot. So where do you think the guardrail should deploy before it goes out of our hands?
speaker-1: Yeah, I think, think the key thing is always safety. So safety at cell, pack and an EV level. you know, a number of tests are already done at each of those levels. For example, there are various tests on battery cells, know, temperature ramps and nail penetration tests, which are done on cells to test for safety. So those will... I'm sure always be in place, know, AI is going to replace any of those anytime soon. And the same at pack level, you know, once those packs are built, they will be put into a vehicle and tested for thousands of miles before that pack ever, you know, we're going to real production. And again, nobody, I'm sure no car executive, car company executive is going to release a car without that full testing being done because the car industry has got a long history of... ⁓ issues and recalls and nobody wants, ⁓ to have that recall on their hands from something that's faulty later on. costs millions of dollars or pounds to, ⁓ to correct that when it happens. That testing is always really important. So going back through the different layers from cell pack and at the EV level, you know, obviously that pack has to perform in sync with the EV, you know, in terms of when the driver presses accelerator, the pack responds and the vehicle accelerates. That's all got to be properly tested on the road. It's not the case of an AI till predicting that that's real testing.
speaker-2: Real test data should be there before we can analyze and optimize it.
speaker-1: can use AI to predict, course, but back to the English expression, the proof of the pudding is in the eating. You only know how good the pudding is when you eat it. So you've got to do the eating.
speaker-3: You only know how
speaker-2: That's a good expression, actually. Very practical one, I would say. So other thing before I would like to close this technical part is that, a hypothetical question to you, that if Fonda calls you in a pre-pilot level, so it's saying that, Stuart, could you please give me a 90 days plan to de-risk all the process as well as the sample risks? And what do you say yes to that?
speaker-1: I guess the first thing I would say is we need to look beyond just making samples. mean, making the samples is important, but we need to look beyond that. know, making samples isn't easy, but there are even harder things in future. So again, coming back to the, starting with the end in mind, thinking about what that will look like at full scale and then how we, how we build out the roadmap to get there. So the first thing would really be starting to think about that roadmap. How are we going to go from A to B in terms of lab to commercial scale. So the people, the equipment, the relationships and the contracts are needed. The different pieces of equipment that needs to be built and commissioned and run to generate samples and create learning. that's the first thing. The second thing would be at the current level, a technology risk assessment. looking at what are the risks of. the technology as it is and then as it's scaled up. So where can things go wrong? As it is and then how, they're scaled, as tank sizes are increased and dryers are increased in size, that type of thing. What can actually change as the process is scaled? So those are the two things, the roadmap and the technology and scale at risk assessment.
speaker-2: are very, important points in my opinion. ⁓ Other point I want to ask you, I mean, first of all, I would like to be very grateful for being my mentor during old country days. So it was one of the, I think discussion that I still remember when I would like to move from standard R &D role to strategic partnership role. And also I vividly remember what you mentioned at that time. given that aspect in my own experience, I just want to ask you, what is your one data should have it that improved over time by seeing actually what's happening on the floor.
speaker-1: Yeah, think the main thing for me, I'm in a number of things, but the main thing for me is staying curious about things, trying to ask good questions and trying to listen more than I speak. So I think staying curious is really important and always being ready to learn in every situation that you're in. That's really important and helps. Now that continual learning is really important, I think, keep moving forward rather than just staying stagnant. think that's essential, particularly in this industry.
speaker-2: Is there any one thing that I believe you changed based on the actual experience?
speaker-1: Not really. think, I think, you know, I've been through a number of experiences, good and bad over time. You know, I've had some, some really great experiences, some not so great. So I think, I think learning to appreciate all of that for the learning is important. I think maybe one thing that has changed in the last couple of years, when my time with my last employer was coming to an end, was thinking about retiring maybe, you know, 30 years, I was perhaps starting to think about that, but I've I've changed that now. You know, I'm not ready for that yet, so I'm going to keep going for a while yet.
speaker-2: I believe you have lot of you left Stuart. Other point I want to touch up more on the leadership level. Okay, so I'm sure based on your 30 years of experience, right? So what are the common traits you will notice to hire someone either on the early career stage or on the senior leadership level? One or two common traits you will see beyond their technical degrees or technical knowledge.
speaker-1: I think it comes back to what I said earlier about curiosity and flexibility and being able to respond while in different situations. I think, you know, a good question is, you know, what's the worst mistake you've ever made, for example, you know, is a good open question to see how somebody responds to that. You know, some people see that as a negative thing about making a mistake, but actually we all make mistakes and it's about learning. from those mistakes and moving on. So I think a key, part of any, you know, any, anyone working with other people is their ability to get on with other people and gain trust and communicate well. So that's again, a key attribute. think the sort of recap is, is really about curiosity and teamwork really in terms of. someone being able to fit well into a team and really work well through others as well as through their own work.
speaker-2: So because it has been to different stages of your care, right? Where you have been like an individual contributor to the manager and the leader as well. So what do you, how do you define a leadership, a good leadership? And what is the difference between a manager slash a leader?
speaker-1: question. think a lot of this is made in different publications and talks and everything, but essentially a leader is by definition somebody who has followers. So some of the best leaders that I've seen are people who can develop a team around them who will follow them through anything. You can build trust, you can build confidence in their team, a good team spirit and create a team who will go above and beyond for that leader. So that's different from a manager, which is more administration, I suppose, in terms of. developing plans and schedules for their team and administering the team around them. A leader kind of sets a direction and not only that gets people to follow that direction, which is sometimes difficult to achieve, but that's the goal really of leadership.
speaker-2: Okay, yeah, so I think can close the discussion by firing you some quick questions, maybe funny ones, they would say. So just to, we want to understand a bit more about you, beyond your batteries and this kind of world. So are you ready?
speaker-1: Go for it.
speaker-2: So are you a coffee guy or a tea guy?
speaker-1: Both coffee in the morning, tea in the afternoon.
speaker-2: Good. So, bird or night owl?
speaker-1: Alibert.
speaker-2: So a failure you learn most from.
speaker-1: Always get by in and communicate. I learned that one hard way.
speaker-2: by hardware.
speaker-1: Yeah, I'm not doing that.
speaker-2: I mean, what is the best way to learn that?
speaker-1: Yeah.
speaker-2: Any books? Favorite books?
speaker-1: Yeah, it's not a particular book. It's a series by J.K. Rowling called Strike. She writes under a pseudonym called Robert Galbraith. Recommend it to anybody. Some of it's quite dark, but it's really well written.
speaker-2: I only know Harry Potter from my daughter.
speaker-1: Yeah, you've got to look at Strike. There's a TV series on BBC iPlayer, but the books, particularly the audiobooks are very good.
speaker-2: I'll have a look at it. So best advice you ever received.
speaker-1: Best advice, check it again. And also tailor, remember your audience, tailor things to your audience.
speaker-2: One question you'll always ask in your interviews.
speaker-1: Yeah, coming back to what I said before, it's what's the worst mistake you've ever made. So it's interesting to see how people respond to that and what they learned from it.
speaker-2: So one mistake a junior engineer should avoid.
speaker-1: It's difficult to speak of one thing, probably the worst failing is assuming they know everything.
speaker-2: knowing everything. That's a good point. So one test you will always do before you run pilot line.
speaker-1: Yeah, I think, well, before we run it, checking it's clean. So checking cleanliness, but the one test I wished that there always was on every pilot line is in situ, particle size testing. So BSD testing. quite expensive, but they're valuable.
speaker-2: So dream travel destination.
speaker-1: Vietnam or Thailand. I think I'd love to go and see either of those two. Maybe both at the same time, but that's on the list.
speaker-2: favorite sports to watch or play.
speaker-1: The favorites at the moment are cycling and swimming. I'm mostly swimming at the moment. I've had a shoulder injury, so I'm trying to repair that. But so the swimming's the one.
speaker-2: So book you gift most to young people.
speaker-1: I don't necessarily gift it, but I recommend it. It's a book called The Seven Habits of Highly Successful People by Stephen Covey. It's an excellent book, well written and still after, I don't know how long, 50 years, it's still really good advice. More about self-management than anything else, but it's a really good book.
speaker-2: So one advice you give most often.
speaker-1: Probably look after yourself because without your health, without you can't do anything.
speaker-2: That's very important for founders actually. A career risk you glad you took.
speaker-1: There have been a few, probably the biggest is moving into consultancy. That was probably the riskiest. So yeah, I'm glad I did. It's challenging, but great.
speaker-2: Let us ship principle you changed ⁓ your mind on.
speaker-1: I'm not sure I've necessarily changed my mind on anything, but I think being patient, assuming that I've had to learn, I wasn't always patient and I'm still not always patient. But just focusing on delivery on the long-term and being appreciative of every experience that you have is quite powerful. So being in the moment in the now is really important rather than always. focusing on the future, just appreciating where you are as well. because that's life, your life isn't in the future, your life is now.
speaker-2: So if you got miraculously one more extra hour per day, how do you use it?
speaker-1: I'll probably use it to do more walking. I enjoy getting outside and walking. We live in a village, so it's really easy to get out into the countryside. So we have a dog, so he needs walking twice a day, but probably another hour, I'd probably go for another walk with him.
speaker-2: So last one, one line you will tell to yourself, anger yourself.
speaker-1: My younger self. Probably that it'll be okay. It'll all be okay in the end.
speaker-2: So any closing remarks before we close the discussion?
speaker-1: No, thanks for having me here today, Sri. Really enjoyed it and I hope it's been useful for your listeners.
speaker-2: Thank you Mr. Ward. think I really want to thank you not only for this particular session for agreeing immediately when I saw yesterday, but also being my mentor for a long time. I really thank you for that and really appreciate it.
speaker-1: Welcome. I don't feel like I did a lot,
speaker-2: Really appreciate that.
speaker-0: Okay.
speaker-2: given your experience, they most of these programs
speaker-1: It's anywhere from cell manufacturing back up the chain. think highly regulated industry where it can take two or three years to qualify and validate a product with the end customers. If you think about having two multi-story car parks and one of them is the cathode and one is the anode and the lithium ions are basically the cars which can be stored in either car park and have to move between them. And as things scale up, things change. So mixing patterns in vessels, tanks change. Heat transfer changes.
speaker-2: which is more of a big boys game rather than a start ups game. ⁓
speaker-1: I think it is, yes, largely. think it's very difficult for startups to break in. Getting to mass production is very tricky. Yeah, I don't have a crystal ball, but I guess looking at the general trends and know, AI is going to replace any of those anytime soon. It's not the case of an AI till predicting that. The proof of the pudding is in the eating. You you only know how good the pudding is when you eat it. So you've got to do the eating.
speaker-3: You only know
speaker-2: The failure you learn most
speaker-1: Always get by in and communicate. I learned that one hard way.
speaker-2: Welcome to the SP show. Today I'm joined by Stuart Johnson, founder and director at Solid Process Limited, a UK based firm. Stuart has over 30 years of experience in process engineering with a track record of turning complex industrial problems into repeatable scalable systems. In this episode, we'll get into the battery value chain with a practical lens, which battery chemistries are likely to matter the most over the next decade, what regulation will actually change on the ground, and where AI can genuinely speed up the material development and industrialization versus the where it is mostly the noise. Stuart, welcome to the show. Great to have you here. Can you please tell us briefly about your 30 years long career journey?
speaker-1: Thanks Sri, great to be here. Sure. Yeah, I started with really a chemical engineering degree at the University of Leeds followed by PhD. And then the following 30 or so years in the chemical industry in pigments, catalysts and battery materials for the last 10 years or so. And most of that's been in process development and business improvement and also some time in the commercial area in procurement for about five years. and also in process licensing. So quite a varied mix of things during my career. And a couple of years ago in 2023, I set up my own consultancy.
speaker-2: you very much. mean, 30 years is quite long. Let's try to dig something about what you have done and how it's helpful for batteries itself. So as you mentioned, you started your own consulting firm. So what is the consulting firm is about and why you started this?
speaker-1: Okay, well, the last company I was working for was coming to an end and I was looking kind of what to do next. Some colleagues had moved into the consultancy area, including yourself, Sri. So it seemed like a good opportunity to do something a little bit different from the corporate life. And really I wanted to use the experience that I had to help other companies in the battery industry. I'm very interested in the green sector. So that's really where I wanted to stay.
speaker-2: So what exactly you're considering from Solve process offers to its plying.
speaker-1: So really what we do is look at helping companies in the mid part of the battery materials value chain to help them to scale up, to develop their processes, to move towards commercialization, solve problems with technology or scale up and a host of other issues as well. But those are the main things.
speaker-2: getting back to the batteries, what exactly pulled you into the batteries originally and then what is keeping you still today here.
speaker-1: Really what pulled me in the first place, I was working for Johnson Matthew who were a large FTSE 100 company, actually in the catalyst division, the process catalysts, which are used in industrial processes. And I was there while they were setting up a battery materials division. And I happened to be in a meeting with the CTO at that time, Shirley Brown, and she basically head hunted me into the battery materials division, which was right at its infancy. ⁓ so that was my start and then worked, worked for several years after that in the new division. And what, what keeps me working in it is really, it's so interesting. It's a very complex and dynamic industry. mentioned earlier, I'm very interested in the green transition. That's the underlying purpose really for being in this part of the industry and also the people I work with and the places that I get to visit. It's really interesting pieces of work that I get to do.
speaker-2: So it's quite a quite obvious transition right from catalyst technologies to the batteries. Given the same context, can you please map out for us that what is the what is the meaning of battery value change?
speaker-1: Sure. And it's, it's quite a long value chain. So it starts right back at in mining. So I'll talk our way through the chain and then we can give a little bit of detail of each stage as well for a bit of context. the value chain starts with mining, mining of ores from the ground. And then those, the metals, the key metals are extracted and purified. And here we're talking about base metals like nickel, cobalt, manganese. Those are the basic building blocks. So we need pure forms of those metals from the ores. I mentioned those metals, but also the similar value chain for iron based materials to make LFP, lithium iron phosphate as well. But sticking to the base metal battery materials, those building blocks are made into what's called active materials. And that usually happens in two steps. So there's a precursor stage and then there's a cathode active material stage. And then those cathode active materials are made into batteries or cells. And then those individual cells themselves are built into a pack. It's a battery pack, which also has some electronics, which is known as the battery management system and cooling systems and a structural architecture to hold everything together. And then those packs are what are used in electric vehicles or EVs and also other energy storage applications for stationary storage, for example. And then at the end of the value chain, those packs, say from EVs can be used in, can be reused in other applications such as EV packs being used in stationary storage. And then when they're finally done, they can be dismantled and those metals can be recycled. So that's, there's a whole recycling loop back to the start of the, the chain again.
speaker-2: Okay. So going from the mining all the way across the recycling, getting back to again into the value chain. Given, given your experience, just I want to ask you this point that where most of these programs die. Because as you see it nowadays, have seen a lot of negative news around the value chain. So particularly around the, I would say, cell manufacturing side. But, based on your experience, where most of these programs die off.
speaker-1: I think it's anywhere from cell manufacturing back up the chain. think what most people don't realize is how tightly linked the elements of the chain are. So there's a very close relationship from the vehicle manufacturing, we call the OEMs, original equipment manufacturers, to the cell manufacturers, to the active material manufacturers, and even back into the battery chemicals. So all of those links have to be really close and we're talking about highly regulated industry where it can take two or three years to qualify and validate a product with the end customer. So it starts from the EV manufacturer backwards. it's very unlike a commodity chemicals industry where the customers can easily switch between suppliers if they're not happy because each product is specifically tailored for the customer's requirements and the... And then the production line which makes it also has to be approved by the OEM. So this is going right back to battery materials, the active materials, which are made from the battery chemicals. So a supplier therefore can't plan to build a factory to produce a generic precursor or cathode active material. say in two years time, without that material in process being well defined at the outset, they have to be very sure about what that's going to be. And that has to be backed up with an offtake agreement. And before that happens, it has to be two or three years worth of development and qualification work to provide samples and get the customer to buy into that. So there's lots of testing involved and lots of costs. So where things can fall down is where there's technical challenges occur along that value chain. ⁓ Or maybe there's cost or project overruns, which can cause issues. If things change, then everybody else is sort of dragged into that. And this means there's a huge load of barriers to entry for startups because the long time scales before getting to revenue, very high capex levels and very high learning curves, particularly around scale up. So we're talking about people scaling from lab up to full scale. number of steps to go through and a hugely challenging process to achieve that full scale production.
speaker-2: You mentioned there are different sorts of components out there, In a particular battery, for example, is it possible to slightly dig into the details of, for example, what is the role of a cathode or anode or a separator or electrolyte in a very simple lithium ion battery?
speaker-1: Sure, I think probably the most helpful analogy is that of multi-storey car parks. It might sound a bit odd to talk about car parks when discussing batteries, but if you think about having two multi-storey car parks and one of them is the cathode and one is the anode and the lithium ions are basically the cars which can be stored in either car park and have to move between them and the car parks really are essentially the layers within the battery materials within the particles. So that's where the lithium ions are sitting on each side of the battery. So during charging, lithium gets pushed from the cathode across to the anode. It comes out of the car park and goes across and into the other side. And then when the battery is being discharged or used, then it goes back the other way. It's reversible. And the lithium actually travels through the electrolyte, which is usually liquid, although people are developing solid electrolytes at the moment. And then between those two car parks is what's called a separator, which allows the lithium ions to pass through. but it also provides electrical insulation between the two sides. So the failure mechanisms that you asked about Sri, that's really where the car parks start to fall apart. So that's through things like particle breakage, where you can get layers of things building up around the particles or the car parks that's causing a bit like congestion around the car parks and the lithium can't get through. And also loss of lithium. So lithium is somehow getting trapped and that can be in various areas in the cell. And there was also quite a sudden failure which can occur when the separator can get punctured by things called dendrites, which think about things like stalactites or stalagmites in caves. Those are basically like a spike which can puncture the separator and create a short circuit with inside the cell. So that can be catastrophic to cell as well. And there are many issues during cell building itself. So as the cells being produced, you can get things like metal fragments coming in, which can create short circuits as well. You can get lack of alignment as the cells built, which can create issues with basically electricity flowing where it shouldn't be flowing. So that process has to be extremely high precision and high purity.
speaker-2: So just on the same point, among these three or four components, we mentioned cathode, anode, separator, and electrolyte. So which one is having the slightly higher probability of breakage, for example, which starts quite soon or very frequently happens?
speaker-1: Generally most cell degradation is due to cathode, cathodic material degradation. So that's usually through essentially cracking or pulverizing of the particles, which occurs during cycling because they slightly expand the contracturing cycling and over time that creates cracks which propagate. And so the material gradually degrades. That breakage can also occur when doing a process called calendaring, is when the, during the formation of the cell. The cathode layer is basically squeezed together onto a layer of aluminium underneath with a roller under high pressure. And that can also break some of the particles. there's a few different mechanisms, but those are generally the main ones.
speaker-2: So we often hear the words such as SCI, CEI, of key words. So can you please explain us, plain language, would say, what are these SCI and CEI?
speaker-1: Sure, SEI means solid electrolyte interface. It's really the interface layer between the particles and the electrolyte. At the moment, that's usually between solid and liquid. And generally what happens at that interface is you can get some side reactions which occur and you start to build up layers of material around the surface of the particle, which start to create resistance essentially to lithium ions moving. in and out of the solid. One of the main components within the SCI is lithium. So that means that lithium is trapped in that layer and can't participate in the electrochemical reactions occurring within the cell. Effectively, the SCI is going back to the car park analogy. It's a bit like blocking the car park entrances and exits around the particles. You can do various tests to check. like post-mortem analysis, you can pull the cell apart and have a look at particles. And you can also do some specialist testing without tearing the cell apart, such as cyclic voltammetry. Well, the cell is in cell testers. Through that, you can detect the SEI. Having said all of that, the SEI is somewhat inevitable. It's generally created during fast charging mostly because that tends to trigger more side reactions, but it's not completely undesirable. So it does provide some surface protection. The surface is generally, particularly in high nickel materials, is where a lot of side reactions do occur. So actually you want to build up a little bit of an SEI to protect that material from further side reactions.
speaker-2: what you're saying is that sometimes it's, we intentionally would like to build up these kinds of layers to protect the stability of the materials of the entire sand itself.
speaker-1: Correct, yeah. And quite often, in fact, almost always, the cell maker will often do some conditioning of the cell before the cell is actually sold. So they will cycle under certain conditions to create the SEI that they're looking for before the cell is sent out. that there's actually a conditioning or preconditioning step that goes on before the cell is released to the customer, which achieves the SEI that the cell manufacturer wants to achieve.
speaker-2: So earlier in the conversation, you mentioned that there is a calendaring process. So could you please explain main steps in order in the making lithium ion battery cell, for example?
speaker-1: Sure. So the materials that we've been talking about earlier, the cathoductive materials, those are made through something called precursor. So the precursor is a speciality powder with a very precise size and shape and chemistry. So the precursor is generally, if we're talking about nickel, generally nickel-based materials, that's a hydroxide material. That's blended with lithium and then it's fired at high temperature to create the cathoductive material. And also during that process, often coatings and dopants are added to enhance performance. So that cathoducting material is really the material that's sent to the cell manufacturer for making cells. And that's taken and mixed with other things like binders and conductive carbon and solvent. And it's made into a paste or a slurry. And then that's spread in a very thin layer onto a layer of aluminium. It's about the width of a human hair, so 60, 70 micron, something like that. So it spread us that paste on the aluminium foil and then that's dried. And then that's layer which is used as a cathode in the cell. And then a very similar thing happens on either side of the battery, the negative side with the anode material, which is generally at the moment, it's mostly graphite, but you're seeing some of the materials come in like silicon. base materials as well. And that's onto a copper foil instead of an aluminium foil. So those two parts, the anode and the cathode are the basic building blocks and then those are taken and they're built into something a bit like a sandwich inside the cell. they're put together with the separator, the roller or folded into certain shape, whether it's cylindrical cell or a prismatic or pouch cell. And then the cell is filled with electrolyte and then sealed up. So that process is extremely precise. As I mentioned earlier, you've got very tight tolerances for everything, extremely high purity. So if you think about something like silicon chip manufacturing, it's something along those lines, you know, to achieve the purities and tolerances that you need to make a good cell. And of course, not forgetting we... That requires extremely experienced and skilled people to do that as well as the machinery.
speaker-2: I had to manufacture the silicon layers manufacturing is quite interesting. may I ask you to mention that you usually coat the cathode material on aluminum foil, right? Whereas the anode material, if I understood right, it's on the copper foil. Why it is, why one is on the aluminum, other one is on the copper?
speaker-1: It's to do with the different potentials of those materials and the reactivity of those within the cells. it's, to be honest, I don't completely understand all the material science aspects of it, but that's, it's about the, the, the relative potentials of those metals and the effectively of those within, the electrolyte and at those voltages that each of them are seeing.
speaker-2: OK, yeah, I would like to slightly bring you back to your expertise, like process engineering and process aspects of it. So we often hear that the qualification of materials, qualification of equipment, and all these sorts of things, very common language in battery industry. So what does actually material qualification means? So is there any stage gate approach that will be there, or some sample sizes, or what are the timelines? We often hear that the qualification time lapse are at least few years, know, not few months, it's few years. What does it mean actually? What is the meaning of the middle qualification?
speaker-1: Yeah, it's quite a long and difficult process, but it all really starts off back at the cell level and what the OEM, the car manufacturer, for example, wants. So they start off with ⁓ a desired size, weight, capacity, power, performance, and lifetime of the battery cell that they're looking for. And again, that has to fit into the pack that is going to go into the vehicle. So it starts at the cell level and then the cell... The cell manufacturer will talk to the CAM manufacturers about their requirements and they in turn will talk to the precursor manufacturers to see whether they can provide precursor that's going to be able to deliver the cathodactive material or CAM for that particular purpose. So the cell manufacturer will go out with those requirements to usually several suppliers and then there'll be a sampling process takes place. where those suppliers will send in samples, they'll be tested and there's usually some iteration and improvement development as they get close to hopefully what the cell manufacturer is looking for. So each material is specifically tailored for particular requirements. So people talk about say an NMC 811 as being a generic material, but there isn't really such a thing as an 811. know, it's, they're all different. and they've all got different particle sizes and perform differently. So people often talk about four stages of sampling, ABCD. So the first step we've already touched on, which is the conceptual stage, which is where small samples, usually up to maximum of 10 kilograms is sent by those suppliers to the cell manufacturer. And they will use those to make for example, coin cells or single layer pouch cells, are larger cell than the coin cell. And then there's iterative process goes on to improve. And then at the end of the A stage, the cell manufacturer will normally narrow down to one or maybe two suppliers for that material. And then you're into B, C and D stages. So stage B is really a design freeze. So the material is frozen, properties are locked in and then the supplier or perhaps suppliers will be looking to produce samples from their pilot plant up to a thousand kilograms and that this is really proof that they can scale up and so they will be using the same type of equipment they will be using at full production scale. So that would involve thousands of tests by the cell manufacturer. thousands of cells to build up the statistical confidence in the production process to prove that it's reproducible and it's statistically a very smooth process. And this is a key failure point really in that testing. The scale-up is very difficult. And then the C testing is really about process validation at the commercial scale at full production volumes once the production line is built. So this involves very intense statistical process control. So the, line is run for probably several weeks. And this is to really validate the actual production line can produce exactly what was produced from the pilot line. And then the final stage, the D sampling is the final production and validation before the, the cell manufacturer gives a go ahead for start of production. So beyond that point, after D sampling you're into real production and any changes that are made, hopefully not, but there may need to be some changes made. Those have to be validated or checked with the customer. And it may mean actually going back to B or C stages. can't be basically from that point on, but you can't change anything without approval. All of this A to D testing has to be done by the CAM and the PCAM producer in synchronization between them. So they've got to ensure that they're in sync with any samples that are being produced. the relationships have to be extremely close because there's a mutual destiny for everybody in that supply chain.
speaker-2: So you mentioned that usually there are four different sampling stages, right? So is it okay to imagine that the sample is usually coming from the either from the lab or from the pilot line to start with? Because 10 kgs may not be lab sampling, right, by the time.
speaker-1: It usually is lamp sampling. Yes. It's usually large lab samples. Typically lab reactors are capable of making maybe a few hundred grams, something like that. So it might be several of those samples blended together into a single sample or they may keep them separate, but it's normally up to one kilogram. Sometimes 500 grams is enough. It doesn't have to be as much as a kilogram.
speaker-2: Okay, so and typically the B sample is coming from definitely from pilot line.
speaker-1: Yes, usually, yes.
speaker-2: usually, and the C is more of a demo slash commercial plant.
speaker-1: Yeah, that's normally the run on the actual production line, the commercial scale line.
speaker-2: And D-sampling is definitely a full-scale commercial line. obviously, like you always hear that, you know, there is a slippage of timings. So what is the usually the program slip and which stage and if so, how many months, for example?
speaker-1: It really depends on the circumstances. So slippage can occur during equipment commissioning once the process is built, but also back from there, you know, looking at the sampling stages along any of those stages. If the product for some reason isn't, isn't right, then the whole thing has to be started again. And that can set you back several months. I think quite often we focus on equipment as, as engineers and scientists, but it's often all about the people. So it's about the experience, the skills, the motivation of those people that are in each company that really makes the difference. So the slippage can occur really for a number of technical reasons, sometimes commercial reasons as well, but also purely project reasons in terms of the delivery of equipment or commissioning of equipment. So there's a whole host of things that can trip us up as we move towards commercialization, unfortunately.
speaker-2: So in terms of the qualification timeline itself, what is based on your experience, what is the typical timeline for, let us say, an MC 811 to qualify with a new cell manufacturer, for example.
speaker-1: So typically that I mentioned the several stages, normally the A and B stages can be, let's say up to two years. And then CD obviously takes place once the line is built, which can take up to two years. So in total, they're not necessarily all in completely aligned in series because they're separated by the time it takes to build the full commercial scale facility. Overall, it could be four years easily in total.
speaker-2: Just out of curiosity, I know the answer, but just I want to hear it from you. So why it will take two years, for example, for A and B samples? For example, I opened a shop. I can produce half a kg of samples. And once I have the lab, usually I build up the pilot line as well. That needs at least a few hundred kilograms, let us say. So why you need two years to do what?
speaker-1: Okay, so normally the first stage, there are several iterative cycles. So it could be, you could be close with the material that they're looking for initially, which will be extremely lucky. You could also be in a cycle of maybe five or 10 separate samples, which all have to be tested. And that testing can take at least a month for the customer to take that and make it into coin sales and at least two or three weeks. for initial testing in coin cells and probably longer for the larger format cells. So the cycle could be from at least say two months to produce them material and have it tested initially. then any more iterations from that each will be at least a couple of months. So if you've got 10 iterations, that's probably a year before you know it. And then the same thing can happen at the the pilot scale as well where you've got again those iterations you know you may already be very good at making that material in the pilot scale from the lab but there's usually some tweaking to go on to ensure that you produce the material that you want so you've got time to make the samples And then in this case, you've got to make more cells from those samples. So the testing takes even longer, maybe two to three months per set of samples. So you've got each cycle, maybe three or four months production and testing. So you've got maybe three or four cycles of testing. That's probably another year. It's very difficult to say exactly because every situation is different, but that's how you can get to two years.
speaker-2: That means typically not only the material manufacturing timelines, but also the, what I'm hearing is the cell testing timelines are the kind of real bottlenecks slash the challenges. One has to overcome to squeeze these qualification timelines.
speaker-1: Exactly. Yes. It's not just a question of putting the sample into the lab and then you get result back in a couple of days. What the cell manufacturer has to do is firstly build the cells, which takes some time. And then the cell is put into cell cycling equipment, piece of equipment, cell testing. And then the cells are cycled up and down repeatedly for thousands of cycles. that's... that takes time. can't be speeded up. Obviously people do try and accelerate that, you know, in different ways, but ultimately the final testing has to be without that acceleration. you've got a very long process to produce and test material.
speaker-2: lot of techings came up now, Like HPCE or some sort of monitor-corner simulations, modeling, so on, you know, just trying to accelerate the testing itself. So you mentioned several times that there is a kind of, know, due diligence process goes from the cell manufacturer to the cam manufacturer, down back to the precursor manufacturer. So based on your experience, what is the, like a one trap kind of thing you're seeing always, you know? or any particular documentation you will look for to avoid those kind of traps.
speaker-1: I don't think there's any particular documentation. I think the main thing, as you go around different facilities, you build a picture of what the company is like. As engineers and scientists, often focus on equipment. But quite often it's all about the people, the experience, the skills, the motivation of those people, which drive obviously the design of the installation of the equipment in the first place. And in the battery industry, those people are very uh, skilled and quite rare and quite mobile as well. So actually the organization chart of the company and the CVs or resumes, you know, say a lot about, about the company as well. Um, and likewise, you know, when you're walking around a facility and you're looking around, you can pick up a lot, you know, in terms of the cleanliness, the, how well ordered the facility is, the design of the facility, what the people are like. ⁓ and how they respond. I think there isn't one single thing when we talk about due diligence, ultimately the product is a product by process. So everything is important in delivering that product through the process from the raw materials, through the process equipment, through the way that equipment's operated to the labs. So there isn't really any, any one thing you can pick out, but you tend to build up a a picture as you walk around somewhere, you know, how well run this facility is. And I find, you know, a good question to ask is, you know, what would you do differently? Good open question. You know, if you could change something, what would it be? And that sometimes opens up people to maybe saying a bit about what doesn't quite go right, what they would change, what they would like to improve. And it's a good opening to. to explore that with them to discover more about the company.
speaker-2: Maybe I'm taking the liberty to ask you based on your experience, right? usually what I've seen based on my experience as well, on the lab scale, you'll get the beautiful product, the things in spec. Now everything is perfect. But when you kind of try to scale it up, know, where this process engineering concept, I think, you know, comes to the picture, the pilot demo or commercial, whatever it is, that scale up stage. So what are the usual kind of, know, bottlenecks or traps that kills a lot of startups, right? Do you have any secret sauce to fix those things?
speaker-1: I wish I did. think, yeah, obviously as things scale up, things change. So mixing patterns in vessels or tanks change. Heat transfer changes the way that gases move through material inside the kilns that we use to produce a cathodactive material. That changes as you scale up to larger scale. So a lot of things change as you scale up. A lot of it is about understanding what can change and being aware of some of those traps and doing the proper testing as far as possible at large scale equipment as you can. think a lot of startups don't focus on the end result and I love the quote or the saying from Stephen Covey's book, begin with the end in mind. So it's a really long journey to get to commercial production. It's a marathon rather than a sprint. So a good question to ask is what will this look like in five years time? How will this be commercialized? Who will the competition be? You know, what will the full scale process look like and cost? And what stages do we need to go through to get there? And startups particularly don't have a lot of experience of usually that scale up going from lab to pilot to full scale. And that's why they often need some help. trap, guess, just talking about scale up is perhaps people ignoring some of the glitches that occur. know, maybe something's clogging up or maybe once in a while, the batch doesn't quite turn out as expected and people perhaps will say, we'll work that one out later when we're at larger scale, but things will just be even worse. So it's better to fully understand what's going on at the time at the smaller scale before things get larger and more expensive. and can cause greater delay. guess another trap as well is planning, which can sometimes be too idealized. We can put together a great plan, but that needs to be stress tested. What happens if, for example, the scale-up takes twice as long and costs twice as much? How is that going to affect our business plan? Is that going to make the complete plan fail? If so, then how do we prevent that? Or how do we raise more money to mitigate that? And I guess the final thing is around people again, perhaps the team that is currently in place is not the right team to take things forward. And that requires people to look forward to what people will need to be in place. And that's quite a difficult thing to predict, but the right people need to be in place at the right time and they've got to be up to speed before then. there needs to be quite a lot of career planning or organisational planning for people to be in place at the right time.
speaker-2: That's very interesting. think I'm hearing a lot more times about the people and skills, those kinds of things.
speaker-1: Yeah, yeah, it's all about the people.
speaker-2: So just on the same context, do you say based on experience so far, particularly the cathode materials is more of a big boys game, actually rather than a startups game at the moment?
speaker-1: I think it is, yes, largely. think it's very difficult for startups to break in because of what we've already discussed. know, the barriers to entry, the high capital costs, the long time scale before reaching, you know, positive cashflow. Some of the technological barriers as well. Some of the people barriers, you know, maybe not being able to attract the top talent because either those people don't want to join something that's more risky or... more expensive. So I think, yeah, it does tend to be the big players dominating and particularly those with experience, you know, in Europe, there are already two big companies with that experience, Western companies plus a number of others coming in as well. it does not to say startups can't find a niche, but getting to mass production is very tricky. And unfortunately, we saw with no fault last year, know what, some of the outcome of that, it's a very hard business to be in.
speaker-2: I think when you mention about the particularly like a demo slash commercial plans, I think it needs a lot of, you know, patient capital as well, you know. And also another way, you have to wait almost like uncertainty for more than five years before you can actually see any light in the channel there. So it's quite...
speaker-1: Exactly. Absolutely. And add onto that very, very tight margins as well. You know, this isn't an industry which is, you know, being allied to the car industry who are well known for squeezing costs. The margins are not super high. So you, you know, you have to be prepared for pretty tight margins in this industry as well. So the whole thing combined together is quite a tough gig for most companies.
speaker-2: It's almost the same chain that you're coming from downstream and pushing towards upstream. it is kind of squeezing. Before digging into bit more details about the chemistries and materials. So could you please explain us what is the exact meaning of process engineering and how it is relevant to the battery industry or for that matter any chemical industry?
speaker-1: Exactly. Yeah, so process engineering is really about how to make the chemical and physical aspects of the process that you want to happen in the way that you want. So if you're talking about a reaction, for example, how to achieve the best yield and selectivity from that reaction in terms of getting the chemicals out that you want with minimum byproducts. And chemical engineering is really all about the physical design of the process, but also the conditions that you run the process at and then designing the equipment around that and all the controls to achieve that safely and cost effectively. it's basically about bringing something that you do in the lab at very small scale to a very large scale safely, cost effectively. and productively so that you can achieve the products that you want. That's a very condensed version of a four-year course usually to do a chemical engineering degree. yeah, it involves understanding the fundamentals of what's going on in the reactor, the kinetics of the reactions, the thermal side of things as well. And then all the physical changes that need to go on, heating, cooling. Boiling things off, condensing things, crystallizing things, separating things. So there's a whole host of things going on within the chemical process that all need to be very tightly controlled.
speaker-2: Being a metal chemist, kind of always underestimated the value of the process engineering actually, because it's a lot more complex than what I assume. So it's quite interesting concept itself.
speaker-1: Yeah, I think most people are surprised when they go into a chemical engineering course, there isn't actually a lot of chemistry in the course really. I mean, there is some obviously, but most of it is about physical processes and mass transfer. So heat and mass transfer are two of the core modules within every chemical engineering course. So what affects those and how you control those and achieve the best you can out of the process. with the equipment that you have or design equipment to achieve the best results.
speaker-2: But I still remember, think ⁓ long time ago when we were discussing, you always used to mention that we've done this mass heat balance. But when I saw like a scaling of the single crystals, particularly when we work together, it was quite important because the flow rate of the gases itself makes a huge impact on the performance as well as the homogeneity of the particles and so on. Right. So which process makes the most impacts, the human form of material as well as the homogeneity of the materials, particularly from the CAM perspective? And what is the, what is spec window that is non-negotiable for a process engineer.
speaker-1: Lots of questions in one question. Okay. Yeah. think, I think really to make a good cam cathode active material, really the precursor is the foundation of that. So the precursor is a very finely tuned product with a certain chemistry and particle size, which you use to make the cam. And we talked earlier about the precursors used with you take the precursor and mix it with lithium. And then that's baked to make the cam. So it's possible to make a bad cam from good precursor, but you can't make good cam from bad precursor. can, the precursor has to be good before you can make good cam. And really in terms of the precursor itself, there are basically two, two main categories of specifications. So one is chemical. What's the composition of that material? So not only the bulk composition, but the impurities. So that's one side of things. And the other side is more about the physical. attributes of that material, the particle size, particle size distribution, the crystal morphology, the porosity and the density of those particles. So there's a number of different aspects. So I mentioned earlier about product by process. It's not like most of the chemical products where you can buy a product and just look at specification on a piece of paper and say right that contains you know x percent of this y percent of that and that's okay that's specification so we can buy it. The precursor can only really be checked and tested through proper lab tests actually making that material into a battery cell and testing it so you've got to take the precursor through CAM and then into cells and that those cells has to be tested. And that's, that's quite a long drawn out process to do that. So it's not like you can tell from the specification how it's going to perform in the cell that there isn't a one-to-one relationship between those two things. You've got to do the cell testing to tell what that material is going to be like.
speaker-2: So is there any minimal viable process before shipping first pilot samples? So do you follow any kind of checklist before you ship your cam to a cell manufacturer, for example?
speaker-1: Yeah. So usually there's a number of things. So that viable process involves, you know, we talked earlier about chemical engineering, but process engineering, design of that plant has to be finalized. And that's usual chemical engineering aspects. So you've got flow sheets, equipment specifications, materials, you've got piping instrumentation diagrams, P and IDs. So there's the physical design of the plant, but then there's also the operational side, the operating procedures or SOPs, standard operating procedures, we call them. The control system has to be operational and tuned and working properly. And then we have to do a load of statistical testing by running the processor, probably several months to check that the process is in control and it's producing the product that we want. there's, there's a lot of work to do to, to get to that point before we have what we would call a viable process. And another thing which is forgotten actually is the, that pilot plant has to run using the same raw materials that we're planning to use at commercial scale. So before we get to that point, we've got to qualify those raw materials and get supply contracts in place to supply the pilot plant. So that can take several months to do as well. So there's a number of things that need to come together to have a viable pilot plant process. that can produce good quality samples for a customer. And then also, of course, we need our own lab alongside that for our own internal testing and validation of what's produced. That all has to be set up and operating properly as well, you know, to be able to samples coming off the pilot plant.
speaker-2: In that case, what you're saying is that we almost, we pre-qualifying the precursor as well for CAM. And also we need to qualify the equipment as well. It's a kind of, know, qualification process. It's part of the qualification process itself.
speaker-1: Yeah, I mean, that's more of an internal qualification to the pilot plant because at that stage when we're supplying, in this case, it will be B samples to the customer. The customer at that stage is probably going to have some degree of inspection and validation of that process just to check and confirm that they're happy with it, but it's not going to be as in-depth as the full scale line, but there will be, you know, for internal purposes, we need to make sure that whatever samples come off that production line are repeatable. and it can be a suitable quality. So we're not wasting our time and somebody else's time to test that material. So yeah, there needs to be several months of work once that pilot lines been put together to check it and check its operating as it should, you know, in control so that we confident that we can replicate that later, both from the pilot plant and also from the full scale plant. And a pilot plant is a great way of understanding what's likely to happen at larger scales as well as producing samples. The purpose of the pilot plant is to find out more about the process and check how robust that process is and what can go wrong.
speaker-2: That's very interesting discussion. And we'd like to slightly steer the gear and go more towards the actual chemistries of the chem itself. Okay. So as we know, there are multiple chemistries that are available in the market. Like NMCs, which is nickel manganese cobalt based compounds and LFPs, which is the new trend now. And LMFP, the derivatives of LFPs are also gaining a lot of traction. And also the big boy in game is the. sodium ion, where CIT are like betting a lot on it. So based on their experience, where this each category of cam will fit for next five to 10 years.
speaker-1: Yeah, I don't have a crystal ball, but I guess looking at the general trends, I'd see NMC still having a place, you know, at the high end, the longer range end of passenger EVs. Probably that is, the market share is probably declining, but the actual absolute volumes are increasing. LFP, lithium ion phosphate is continuing to improve and grow. and breaking more and more into the kind of low and medium range end of the segment. Particularly with the fourth and fifth generation LFP, the highly densified material that's coming in. So that's approaching NMC type performance overall. So LFPs made huge advances in the last few years and I think will continue to do so. That's probably on a... going to plateau out eventually. Sodium iron, yeah, you mentioned that that is the big boy in the game that's coming along. I've got some skepticism about how well that will break in. mean, that's really about the balance of pricing between nickel and lithium. So most of the sodium iron material is made using a nickel based cathode material. That may not continue, but that's where we are at the moment. And obviously nickel is fairly expensive compared to iron. So there's a disadvantage there, but obviously the advantage is that the sodium iron doesn't contain lithium. It's all about that balance and the performance. the performance of sodium iron material is fairly similar to LFP. That may increase in the same way that LFP is increasing. And I guess the other trend to mention, I guess, is Development from NMC is manganese rich. So this is a move away from the higher nickel content back to lower levels, maybe say 20 to 40 % nickel. She's starting to, that LMR, this in manganese rich material is starting to eat into some of the LFP and NMC space. One of the features of LMR is that it requires higher cell voltages. So There are some challenges that come along with that because of the changes that can occur both within the electrolyte and with manganese itself and trying to avoid manganese from leaching out of the cathode material into the electrolyte, which can occur. Manganese is quite a tricky material to manage in terms of its oxidation states. And so that material has quite a different structure. It's generally spinel type material. So again, it's all about understanding how to manage that material in the context of the cell in that whole electrochemical system that's happening within the cell itself. So that's the general overview. think the general trend is LFP increasing and sodium iron and lithium manganese rich coming in. to grab some market share over the next five years.
speaker-2: So talking about the sodium ion and also the LMR you mentioned, so what is the supply chain status for these two chemistries particularly?
speaker-1: LFP is generally China based. In fact, China restricted export of the most higher end of the fourth and fifth generation technologies to the West. it's primarily China driven LFP, although there are some other Western producers such as Nano One in Canada. That's developing. Again, it's mainly Chinese. dominated, but some in Korea as well. So LG, for example, have got an LMR offering, which looks good. I think it's a developing space. think a lot of NMC producers are looking towards LMR as a potential way to switch over production. The same facilities can basically be used with some modifications. LMR could be an option for an NMC producer already. already making NMC. So I think there are some options there, but the supply chains are like most things, primarily Asian dominated, mostly China at the moment. So it's a question of where other parts of the global economy can find ways to break into that.
speaker-2: So you mentioned that LMR is also kind of getting traction. So it is quite understood that LFP and sodium ion is kind of sharing between the EBS and ESS markets. So where do you see this LMR is going to play its role?
speaker-1: I think probably at the... towards medium to high range EVs, that's really where it's at the higher end. It's basically allowing a switch from a high nickel NMC to an LMR. So it's probably towards a higher end, ⁓ yet to be proven out in many vehicles yet, but it's starting to be released commercially. So we'll have to see where it really fits, I think because you can raise the voltage, although the specific performance of the material itself is lower, it's typically around the same level as LFP. The voltage allows you to achieve a higher level of storage per mass of material. LMR fits really at the higher end. But as I mentioned, there are some challenges with making that work in terms of the electrolyte and specific coatings that are required to stabilize the material.
speaker-2: So in terms of the ⁓ applications, right, of different chemistries, what I'm hearing is that probably three categories of applications we are looking into. One is the standard passenger EVs. It can be low-end, high-end segments. And middle one is kind of coming up, but with lot of interest from the OEMs is trucks, heavy-duty trucks particularly. Just kind of masked until you're developing the technology, of course. But third one is mostly... EIS market, but driven by the AI and its data centers requirements. So what is the kind of priorities, know, change across the energy or power, less on the material level, but more on the pack level? How do you distribute these chemistries across these three segments?
speaker-1: Yeah, it's a good question. think as with all battery technology, everything's a trade off. you improve one thing and you reduce another. So you're always looking at the trade off in every part. For the passenger EVs, really the main push, mean, safety is always a given. So safety is critical. The main push is towards longer range and faster charging. But primarily it's about range and therefore specific capacity. capacity per unit mass or per unit volume. Charging rates are also important. you know, the power, what we call the power performance of the battery has to be good as well to allow say charging within a few minutes rather than an hour. So that's always about the energy density is really the driver, but the other two need to be acceptable as well. Let's say LFP is coming to its own there really because its performance has improved and Because LFP is inherently much safer than NMC, some of the features of the pack can be modified for LFP and you can actually get more material in. So you end up with almost equivalent performance at a pack level and from LFP as from NMC because you don't need the extra safety controls that you would need for NMC. Trucks again, range is very important, but also charging. you think about a large battery like you would need in a truck, you know, The charging rate has to be sufficient within a few hours to charge such a large battery. also on the flip side, you've got the power required to accelerate a large vehicle like that. So you've got to have sufficient power performance. It's a very heavy power requirement. that's trucks. And then if we move to ESS or stationary storage, density isn't... particularly important now at all. It's all about the cost per amount of energy storing. And you you can effectively achieve a certain level of power by multiplying out the number of units by putting those in parallel. So that's not as important, the power aspect. It's all about energy storage per cost. And the safety aspect there is still important, but it's not as critical as in a vehicle, which somebody's traveling inside. And that can be controlled in other ways in a stationary storage location. So I think the other thing to remember from a high level, from a pack level is that the design of the pack is as important as design of and selection of the material itself within the cells. So a lot of aspects like thermal management are very important and the battery management system, the way that that controls. voltage in the current and can detect cells which are failing for example or problematic and that all requires a huge amount of thinking to get that right as well. So it's a very complex system because you've got thousands of these cells all linked thermally and electrically together which all have to operate as a single unit.
speaker-2: In general, we think are we over-engineering the material aspects and less focusing on the integrated system aspects? Why I'm asking actually is that particularly it's also maturity of the industry, Because that's the kind of, you're at an ancient stage where you're focusing a lot more on the, optimizing the materials and the cells, for example.
speaker-1: I think that optimization occurs at a number of levels. I think there's a level of optimization at starting at high level design. And one of our former colleagues is involved very much with that, you know, in terms of design of packs, thermal management, the battery management system, the BMS, how all those cells are placed, how we know what the cells structure are. So that's one level. And again, it's that flow back along the value chain back from packed level to cell. And then the cell, the cell designer looking at how that cell is put together and with a certain specification for that cell, therefore what the material needs to look like within that. how the, for example, the cathode material has to perform or what the volumetric and gravimetric energy densities need to be. What the voltage curves need to look like is the materials charged up and down. So there's a huge number of things that need to be thought about, that comes at a number of layers. there's pack level, cell level, and then at material level. If you think about it, the CAM and precursor manufacturer designing particles with a certain shape, size, density, porosity, crystallography. which has to deliver that final result for the battery materials. I guess it's a bit like a human body where you start from the overall structure, the size and shape of the body and how that performs. that you break it right down. It goes right down to the biology of individual cells and then within those cells, all the different structures within those. So there's a number of different layers within the battery, which are all critical to everything performing as they should just like in a body.
speaker-2: That's a very good analogy actually. So, I mean, we cannot finish the discussion without talking about the regulations, particularly if you're in Europe, where we are good at regulating the stuff. So what kind of requirements actually we need in terms of the design and the data systems around the batteries, particularly when you talk about batteries, a lot of noise around the carbon content, recycling content, battery passport, those kinds of things are coming up, particularly in Europe. So what's your take on that?
speaker-1: Sure. I think we have to distinguish between things which are really necessary for regulation for safety, for example, to provide consumer confidence in the end product, which is really important. And then things which are basically trade mechanisms to restrict or guide trade in a certain way, which will be in interests of the EU or the US or whatever it might be. And then there are things which probably naturally will happen commercially anyway. For example, supply chain due diligence will always need to happen. Any responsible supplier will always do that because they want to ensure that they're buying things from reputable sources and not from areas where there's things going on that shouldn't be going on. Recycling will always happen where it's economic because of the value of the materials in those cells. You know, it makes sense to recycle nickel, cobalt, lithium, et cetera. The carbon footprint, again, that will happen mostly due to economic reasons anyway, because the lowest cost areas are also green areas. For example, the Scandinavian countries, they've got huge amounts of hydroelectric power, which is very green. So that's where generally the industry is gravitating towards Scandinavia. Battery passports and rules of origin. That again will naturally happen because companies don't want publicity. They want good publicity about where they're sourcing material from. So I think, yes, some regulation is necessary. You you think about safety, human health, those types of things, but I think some of the other aspects will probably happen naturally anyway, just through companies behaving in the most economic way possible.
speaker-2: ⁓ So if you're given a power, what is the one decision you would change in you or in the states at the moment in order to catch up with Asia, particularly China?
speaker-0: .
speaker-1: That's a really difficult question. think, particularly in Europe, one of the barriers is the cost of energy. So that is really a barrier, particularly in the UK, but also in other parts of mainland Europe. There are energy costs. Obviously Scandinavia has got an advantage with its hydro power. So that's one thing I would certainly want to change to encourage the battery industry in Europe. Another aspect is people. And we talked about people a lot today and we don't have the people in Europe to really build the battery industry. You know, we can try and move people around. think upskilling people and actually developing people through apprenticeships, training programs, et cetera. That's all really important. And this is, I'm not talking about really academic research. It's more the industrial side and industrialization of the battery. industry that really is needed because let's face it most of the people who know how to do it are in China or other parts of Asia.
speaker-2: So I think that the energies keep coming up again and again. So why the battery manufacturing process energy intensive? Which part of the process is very highly energy intensive?
speaker-1: It's mainly the cell manufacturing and can manufacturing cathodic material. And they're beginning energy consumers, the cell production mainly because of the drying step after coating. So it uses a lot of energy to dry off the solvent. After coating, yeah. Obviously that will change with a dry cathode where there isn't the slurry. The dry coating doesn't involve the slurry so that
speaker-2: The after recording I mean.
speaker-1: that drying stage isn't required. In the CAM process, the CAM process uses very large furnaces, which are generally electrically fired. So those consume huge amounts of power as well. So that's really the main consumer within the CAM production process. Precursor, not so much. There's still quite a bit of energy use there in various pieces of equipment, but it's not as energy intensive as CAM.
speaker-2: So what you're seeing is that both those steps, we can correlate them back to the process engineering again, to really optimize those processes to reduce the energy intensity there.
speaker-1: Absolutely. The companies are already doing that. It's really when they're looking where to locate a factory, if they put it in, pick a country, put in France versus Sweden, what are the relative energy costs? And that makes a big difference because we're talking every few cents per kilo makes a difference. The operation is run on very tight margins, as we've already said. So all of that adds up, that's basically reducing the profit levels of the company. If the energy is more expensive, servicing counts.
speaker-2: So that again shows the importance of the processing engineering aspects of the whole value chain in this regard in a very general perspective. why the heat or cold, know, kill the batteries faster? Like what kind of impact it will have on the material level and how could you fix that one? Is there any kind of recipe or ⁓ secret sauce either on the material level or on the BMS side to fix these two extreme conditions of the the atmospheres.
speaker-1: Okay. Well, maybe if we take, I guess, the BMS first, the battery management system, that's the electronics, but also the cooling systems around the pack. So the cooling is really important and heating actually as well, because that system is usually used to heat. mean, some packs are air-cooled, majority now are liquid cooled. So they've got a essentially air conditioning system or not air conditioning, fluid system which circulates around through the various channels through the pack to remove or in cold weather to add heat into the battery so it can perform. I guess coming back to why the material can fail at high temperatures, it's mainly a result of side reactions. So the electrolytes, mostly liquid electrolytes, though that contains a number of chemicals which can react. The kinetic rate will increase at high temperatures. You can generate more gas at higher temperatures. You can generate more side reactions, which create more SEI type materials inside. And particularly combining that with high charging rates. If you're charging a battery quickly, you've got high current flowing through the battery, but you've also normally got a lot of heat there because the current is generating heat through resistance within the cell. The cell itself has some resistance inside. So you've got a combination of Fast lithium transfer with heat, which can create these side reactions and create gas and other deposits inside the battery. inside the cells, you've got various strategies of trying to address that, be it electrolyte additives to try and suppress some of those side reactions or coatings and dopants within the battery material itself. You're trying to again, suppress those side reactions, particularly at the surface from occurring. You've also got structural changes within the battery material itself. So you've got expansion and contraction and obviously that high temperatures that expansion is going to be greater. The material is probably slightly weaker as well. So you've got a number of changes occurring within the particles themselves and within the cell.
speaker-2: So what about the cold temperatures? Because we see a lot of chemistry slash EVs may not perform well at the very cold atmosphere, like in Sweden or Norwegian regions, particularly. So why the cold will also have impact on the batteries as well as materials from the past?
speaker-1: Yeah, again, it's kind of back to the kinetics of how quickly things happen. it's cold temperatures, everything slows down. Resistances tend to increase. So the electrolyte will be slightly more viscous. that the speed at which lithium can move, therefore through between the anode and the cathode and backwards the other way, will slow down. So essentially, it's just like everything going to slow motion really at cold temperatures and the BMS has to manage that. So it's basically going to limit the amount of ⁓ power that the pack as a whole can deliver. ⁓ to the vehicle or whatever it's attached to. So it's a bit like going to slow motion in a freezer. It's going to slow everything down.
speaker-2: Thank you. That's very helpful because just want to understand from the very simplistic perspective, course, everything, know. So other point I want to touch upon is that we are in AI era, artificial intelligence era. So how do you think this transformation is going to impact the battery value chain itself and where it will have its maximum impact, for example?
speaker-1: I guess on a number of fronts, mean, I think AI is a great thing. think this, you know, it's incredible what it can do. guess from one angle, you know, there's the power supplies required to power these data centers, which are running AI. So looking at it from that point of view, you know, the backup power supplies, the ESS systems, which will be involved in those, you know, will be, are going to be important. So that will itself consume a lot of battery material, battery cells in those ESS systems. But actually using AI itself, I found it's great for, you know, using it to refine something or even generate ideas, but we really need to watch the style because you can spot AI content a mile off. It's very obvious. It will obviously get better, but at the moment it's, it's very different from a human written article or, or a piece of, uh, piece of other writing. Um. I've been using specific profiles within, within, use Gemini generally. So I've been using something called Gemini Gems, which is within that. know you can do something similar within, within ChatGPT. So I've been able to create something which is more skeptical where it requires factual checking to, before it, before it works out and comes back with an answer. And I found that a lot better because I don't know about usury, but I found that, you know, I've got a lot of. hallucinations generated by AI and, you know, having it checking its work and going out and creating a very wide search and coming back and cross-checking facts before it presents them as facts is really important. I guess the other thing to mention is that in the battery industry, it's very secretive. So we don't see a lot of information coming out onto the internet from the battery industry. A lot of that is kept secret. trade secrets, so it's not out there. So some information is obviously contained within press releases and things like that, but there's not in terms of the technical side, particularly that's not really available on AI searches at the moment. That may change, but I tend to think not. think companies are protecting their IP in that way by keeping it off the internet.
speaker-2: I kind of agree with that in terms of the technical data, think the AI is not good. I'm not sure it's going to be improved a lot better in future because as you mentioned, more of a trade secret, it's no house than actual public floating information. The same information is coming again and again, you know, that's what these are about at the moment. But in terms of the AI utilization, do you think AI can help to accelerate the discovery process, get the new chemistries in? and also rapidified to go to markets, for example. Do you see any kind of value in the aspects of chemistry slash engineering?
speaker-1: Yeah, I think particularly with the advent of high throughput testing, you know, know companies like Navonics in the US are doing that and others, you know, in the US, for example, that I'm aware of. So I think with that quantity of data from all that test work using machine learning, which essentially is AI, being able to analyze those large data sets from multiple angles is really powerful. You can build insights, which you would never have seen before. So I think where that sort of machine Learning is brought together with human know-how as well. That's extremely powerful to be able to predict the performance of a certain formulation of material, a certain particle size, shape, cetera. That is really powerful, you know, to be able to guide research towards where you want to go.
speaker-2: So other point I want to touch upon is more about the guard rails before deploying. Because I think we might use the AI as certain aspects of it, but we have the very large data sets coming from the either from the chemistry side or from the cell testings or from the EV itself. Because as we know that the electric vehicles are more of a computer on the wheels at the moment. know, every data can be captured, refined and then utilized to improve the, you know. the safety or our performance or whatnot. So where do you think the guardrail should deploy before it goes out of our hands?
speaker-1: Yeah, I think, think the key thing is always safety. So safety at cell, pack and an EV level. you know, a number of tests are already done at each of those levels. For example, there are various tests on battery cells, know, temperature ramps and nail penetration tests, which are done on cells to test for safety. So those will... I'm sure always be in place, know, AI is going to replace any of those anytime soon. And the same at pack level, you know, once those packs are built, they will be put into a vehicle and tested for thousands of miles before that pack ever, you know, we're going to real production. And again, nobody, I'm sure no car executive, car company executive is going to release a car without that full testing being done because the car industry has got a long history of... ⁓ issues and recalls and nobody wants, ⁓ to have that recall on their hands from something that's faulty later on. costs millions of dollars or pounds to, ⁓ to correct that when it happens. That testing is always really important. So going back through the different layers from cell pack and at the EV level, you know, obviously that pack has to perform in sync with the EV, you know, in terms of when the driver presses accelerator, the pack responds and the vehicle accelerates. That's all got to be properly tested on the road. It's not the case of an AI till predicting that that's real testing.
speaker-2: Real test data should be there before we can analyze and optimize it.
speaker-1: can use AI to predict, course, but back to the English expression, the proof of the pudding is in the eating. You only know how good the pudding is when you eat it. So you've got to do the eating.
speaker-3: You only know how
speaker-2: That's a good expression, actually. Very practical one, I would say. So other thing before I would like to close this technical part is that, a hypothetical question to you, that if Fonda calls you in a pre-pilot level, so it's saying that, Stuart, could you please give me a 90 days plan to de-risk all the process as well as the sample risks? And what do you say yes to that?
speaker-1: I guess the first thing I would say is we need to look beyond just making samples. mean, making the samples is important, but we need to look beyond that. know, making samples isn't easy, but there are even harder things in future. So again, coming back to the, starting with the end in mind, thinking about what that will look like at full scale and then how we, how we build out the roadmap to get there. So the first thing would really be starting to think about that roadmap. How are we going to go from A to B in terms of lab to commercial scale. So the people, the equipment, the relationships and the contracts are needed. The different pieces of equipment that needs to be built and commissioned and run to generate samples and create learning. that's the first thing. The second thing would be at the current level, a technology risk assessment. looking at what are the risks of. the technology as it is and then as it's scaled up. So where can things go wrong? As it is and then how, they're scaled, as tank sizes are increased and dryers are increased in size, that type of thing. What can actually change as the process is scaled? So those are the two things, the roadmap and the technology and scale at risk assessment.
speaker-2: are very, important points in my opinion. ⁓ Other point I want to ask you, I mean, first of all, I would like to be very grateful for being my mentor during old country days. So it was one of the, I think discussion that I still remember when I would like to move from standard R &D role to strategic partnership role. And also I vividly remember what you mentioned at that time. given that aspect in my own experience, I just want to ask you, what is your one data should have it that improved over time by seeing actually what's happening on the floor.
speaker-1: Yeah, think the main thing for me, I'm in a number of things, but the main thing for me is staying curious about things, trying to ask good questions and trying to listen more than I speak. So I think staying curious is really important and always being ready to learn in every situation that you're in. That's really important and helps. Now that continual learning is really important, I think, keep moving forward rather than just staying stagnant. think that's essential, particularly in this industry.
speaker-2: Is there any one thing that I believe you changed based on the actual experience?
speaker-1: Not really. think, I think, you know, I've been through a number of experiences, good and bad over time. You know, I've had some, some really great experiences, some not so great. So I think, I think learning to appreciate all of that for the learning is important. I think maybe one thing that has changed in the last couple of years, when my time with my last employer was coming to an end, was thinking about retiring maybe, you know, 30 years, I was perhaps starting to think about that, but I've I've changed that now. You know, I'm not ready for that yet, so I'm going to keep going for a while yet.
speaker-2: I believe you have lot of you left Stuart. Other point I want to touch up more on the leadership level. Okay, so I'm sure based on your 30 years of experience, right? So what are the common traits you will notice to hire someone either on the early career stage or on the senior leadership level? One or two common traits you will see beyond their technical degrees or technical knowledge.
speaker-1: I think it comes back to what I said earlier about curiosity and flexibility and being able to respond while in different situations. I think, you know, a good question is, you know, what's the worst mistake you've ever made, for example, you know, is a good open question to see how somebody responds to that. You know, some people see that as a negative thing about making a mistake, but actually we all make mistakes and it's about learning. from those mistakes and moving on. So I think a key, part of any, you know, any, anyone working with other people is their ability to get on with other people and gain trust and communicate well. So that's again, a key attribute. think the sort of recap is, is really about curiosity and teamwork really in terms of. someone being able to fit well into a team and really work well through others as well as through their own work.
speaker-2: So because it has been to different stages of your care, right? Where you have been like an individual contributor to the manager and the leader as well. So what do you, how do you define a leadership, a good leadership? And what is the difference between a manager slash a leader?
speaker-1: question. think a lot of this is made in different publications and talks and everything, but essentially a leader is by definition somebody who has followers. So some of the best leaders that I've seen are people who can develop a team around them who will follow them through anything. You can build trust, you can build confidence in their team, a good team spirit and create a team who will go above and beyond for that leader. So that's different from a manager, which is more administration, I suppose, in terms of. developing plans and schedules for their team and administering the team around them. A leader kind of sets a direction and not only that gets people to follow that direction, which is sometimes difficult to achieve, but that's the goal really of leadership.
speaker-2: Okay, yeah, so I think can close the discussion by firing you some quick questions, maybe funny ones, they would say. So just to, we want to understand a bit more about you, beyond your batteries and this kind of world. So are you ready?
speaker-1: Go for it.
speaker-2: So are you a coffee guy or a tea guy?
speaker-1: Both coffee in the morning, tea in the afternoon.
speaker-2: Good. So, bird or night owl?
speaker-1: Alibert.
speaker-2: So a failure you learn most from.
speaker-1: Always get by in and communicate. I learned that one hard way.
speaker-2: by hardware.
speaker-1: Yeah, I'm not doing that.
speaker-2: I mean, what is the best way to learn that?
speaker-1: Yeah.
speaker-2: Any books? Favorite books?
speaker-1: Yeah, it's not a particular book. It's a series by J.K. Rowling called Strike. She writes under a pseudonym called Robert Galbraith. Recommend it to anybody. Some of it's quite dark, but it's really well written.
speaker-2: I only know Harry Potter from my daughter.
speaker-1: Yeah, you've got to look at Strike. There's a TV series on BBC iPlayer, but the books, particularly the audiobooks are very good.
speaker-2: I'll have a look at it. So best advice you ever received.
speaker-1: Best advice, check it again. And also tailor, remember your audience, tailor things to your audience.
speaker-2: One question you'll always ask in your interviews.
speaker-1: Yeah, coming back to what I said before, it's what's the worst mistake you've ever made. So it's interesting to see how people respond to that and what they learned from it.
speaker-2: So one mistake a junior engineer should avoid.
speaker-1: It's difficult to speak of one thing, probably the worst failing is assuming they know everything.
speaker-2: knowing everything. That's a good point. So one test you will always do before you run pilot line.
speaker-1: Yeah, I think, well, before we run it, checking it's clean. So checking cleanliness, but the one test I wished that there always was on every pilot line is in situ, particle size testing. So BSD testing. quite expensive, but they're valuable.
speaker-2: So dream travel destination.
speaker-1: Vietnam or Thailand. I think I'd love to go and see either of those two. Maybe both at the same time, but that's on the list.
speaker-2: favorite sports to watch or play.
speaker-1: The favorites at the moment are cycling and swimming. I'm mostly swimming at the moment. I've had a shoulder injury, so I'm trying to repair that. But so the swimming's the one.
speaker-2: So book you gift most to young people.
speaker-1: I don't necessarily gift it, but I recommend it. It's a book called The Seven Habits of Highly Successful People by Stephen Covey. It's an excellent book, well written and still after, I don't know how long, 50 years, it's still really good advice. More about self-management than anything else, but it's a really good book.
speaker-2: So one advice you give most often.
speaker-1: Probably look after yourself because without your health, without you can't do anything.
speaker-2: That's very important for founders actually. A career risk you glad you took.
speaker-1: There have been a few, probably the biggest is moving into consultancy. That was probably the riskiest. So yeah, I'm glad I did. It's challenging, but great.
speaker-2: Let us ship principle you changed ⁓ your mind on.
speaker-1: I'm not sure I've necessarily changed my mind on anything, but I think being patient, assuming that I've had to learn, I wasn't always patient and I'm still not always patient. But just focusing on delivery on the long-term and being appreciative of every experience that you have is quite powerful. So being in the moment in the now is really important rather than always. focusing on the future, just appreciating where you are as well. because that's life, your life isn't in the future, your life is now.
speaker-2: So if you got miraculously one more extra hour per day, how do you use it?
speaker-1: I'll probably use it to do more walking. I enjoy getting outside and walking. We live in a village, so it's really easy to get out into the countryside. So we have a dog, so he needs walking twice a day, but probably another hour, I'd probably go for another walk with him.
speaker-2: So last one, one line you will tell to yourself, anger yourself.
speaker-1: My younger self. Probably that it'll be okay. It'll all be okay in the end.
speaker-2: So any closing remarks before we close the discussion?
speaker-1: No, thanks for having me here today, Sri. Really enjoyed it and I hope it's been useful for your listeners.
speaker-2: Thank you Mr. Ward. think I really want to thank you not only for this particular session for agreeing immediately when I saw yesterday, but also being my mentor for a long time. I really thank you for that and really appreciate it.
speaker-1: Welcome. I don't feel like I did a lot,
speaker-2: Really appreciate that.
speaker-0: Okay.
