Fire Science Show

194 - Playing with batteries with Xinyan Huang

• Wojciech Węgrzyński

Share episode

Professor Xinyan Huang from Hong Kong Polytechnic University shares his expertise on battery fires and the various experimental methods researchers use to trigger thermal runaway events under controlled conditions.

• Terminology matters - "thermal runaway" more accurately describes battery failure than "ignition" as the critical reactions occur inside the cell
• Nail penetration testing is widely used but contains surprising complexities, including nail material, penetration depth, velocity and battery orientation
• Mechanical abuse tests (crushing, dropping, squeezing) simulate real-world accidents but often lack repeatability
• Thermal abuse via heating typically targets 200°C surface temperature using methods including flame exposure, electrical coils, and laser heating
• Electrical abuse through overcharging (150-200% SOC) significantly increases risk, while poor-quality charging equipment creates additional hazards
• State of charge plays a crucial role in how batteries respond to abuse tests
• New research aims to bridge the gap between micro-scale material testing and cell-level testing

Professor Huang is organising the 4th International Symposium on Lithium Battery Fire Safety (ISLBFS 2025) in Hong Kong from October 30th to November 2nd - the largest battery fire safety conference in the world.

I intended to link Xinyan's papers on batteries, but there is 19 of them!?! Let me link the most recent ones:

Cover image source: https://doi.org/10.1016/j.est.2024.111337

Thank you to the SFPE for recognizing me with the 2025 SFPE Fire Safety Engineering Award! Huge thanks to YOU for being a part of this, and big thanks to the OFR for supporting me over the years.

----
The Fire Science Show is produced by the Fire Science Media in collaboration with OFR Consultants. Thank you to the podcast sponsor for their continuous support towards our mission.

Wojciech:

Hello everybody, welcome to the Fire Science Show. We're talking batteries today, so an exciting episode coming your way. I had my good friend, professor Xinyan Huang from Hong Kong Polytechnic University visit us as a visiting professor at the ITB. We've spent a great two weeks doing research together and while his time in Poland, I could not miss an opportunity to conduct an interview with Xinyan, and he's been a guest on the podcast multiple times, mostly talking about AI and smart firefighting. His group is also huge on batteries and therefore we chose that the batteries would be the theme of this episode. And given that we are both experimentalists and we've talked a lot about the different battery hazards and battery challenges with batteries so far in the Fireside Show, I thought you know what An interesting episode could be on how do you set off a battery or ignite it where ignite is not the perfect word and you will hear the explanation in the episode how you set off a battery and experiment it's been said in a podcast by multiple guests it's quite difficult to ignite the batteries sometimes.

Wojciech:

Podcast by multiple guests. It's quite difficult to ignite the batteries sometimes, and in this episode we will talk about all different kinds of nasty stuff you can do to batteries nailing them, breaking them, dropping them, burning them, putting a coil around them, short-circuiting or overcharging. All of this is in this podcast episode. So if you're keen to learn how fire scientists, fire researchers, abuse the batteries, then you'll learn it in this episode episode. So if you're keen to learn how fire scientists, fire researchers, abuse the batteries, then you'll learn it in this episode, and I think this piece of information is very important. All of us, fire site engineers, are seeking for information about battery fires. All of us are dealing with some sort of battery hazards. And what you resort to? You resort to fire literature, but in the literature it's very difficult to distill the information presented and understand the information in the context of your own project. So in this case, knowing how exactly do scientists abuse the batteries can allow you to create better, more appropriate design scenarios for your own case.

Wojciech:

So I think in this case, it's critical for engineers to understand how the science is done, and that's what is in this podcast episode. So let's not prolong this anymore. Let's spin the intro and jump into the episode. Welcome to the Firesize Show. My name is Wojciech Wigrzyński and I will be your host.

Wojciech:

This podcast is brought to you in collaboration with Ofar Consultants, a multi-award-winning independent consultancy dedicated to addressing fire safety challenges. Established in the UK in 2016 as a startup business of two highly experienced fire engineering consultants, the business has grown phenomenally to eight offices across the country, from Edinburgh to Bath. Colleagues are on a mission to continually explore the challenges that fire creates for clients and society, applying the best research experience and diligence for effective, tailored solution. In 2025, there will be new opportunities to work with OFR. Ofr will grow its team once more and is keen to hear from industry professionals who would like to collaborate on FHIR safety features this year. Get in touch at ofrconsultantscom. Hello everybody, I am joined here today by Professor Sinyan Huang from Hong Kong Polytechnic University, nice to have you here, Hi Wozniak.

Xinyan Huang:

It's great to see you in person and have this interview.

Wojciech:

It's unusual for us to have an in-person interview, and it always makes me happy to see my guests not in the window of the computer. So I hope you're enjoying your stay in Poland so far. But let's do some serious work and let's talk about batteries, and first things first. You told me you like to play with batteries, so what do you have in mind saying that?

Xinyan Huang:

So in general, I think a battery fire. From combustion point of view it's definitely a fascinating phenomenon. You always see new things like you have a very unconventional ignition, you have the flame, you have very different fuels sometimes generating sparks, explosions. So from scientific point of view, this is definitely a mystery to be solved. And the fun aspect the fun aspect is you always see some unexpected phenomenon. So from researcher point of view, if you see something unusual, you can write a new paper on that okay, yeah, that sounds like our sort of fun also, like I believe the world of batteries is.

Wojciech:

People would put a lot of different things into one umbrella of a battery, but there is so many chemistries, so many technologies out there, so many ways they they are used. Is there any specific focus you put on those?

Xinyan Huang:

like the kinds of batteries that you test or their intended way of use. So, in general, the lithium-ion battery is definitely dominating the entire market. From the laptop to electric vehicle to large energy storage unit, they all use very advanced lithium-ion battery. Of course, there are some new battery chemistry involved, but most of them are still in the laboratory stage and not really in mass production. So I think everything we're talking about today is applied to the lithium-ion battery.

Wojciech:

I'm mostly concerned about stuff related to buildings. So from my perspective, I think all of it is interesting because today, an electric scooter is my design scenario for a building, electric vehicle is a design scenario for a building, a power wall in a garage is a scenario for a building. So yeah, all of this is relatable. Anyway, let's start because you're an experimentalist and you play a lot with those batteries. As you said, I wanted to go deep into igniting the batteries or starting the battery fires for the experiments, and I had some colleagues in the podcast. I had Francesco Restuccia, elena Fong, I had Professor Peter Sturm. Many of them shared the story that it's for them kind of hard to ignite the battery. I assume they were talking about, you know, devices, not just the single cell units. So maybe let's try and talk about how one can ignite the batteries for the purpose of research and maybe let's go from that point. So if you could just give me a brief overview of methods of igniting cells for experiments, thank you.

Xinyan Huang:

I think you touched something very complex. So even when you propose, the question makes me thinking are we really using the right words to describe this process? For example, can we really say ignition, Because in many cases you don't see the flame? We usually consider you trigger some combustion reaction, then you can call it ignition. But in fact many cases something happens within the battery cell. Of course they release some flammable gas, but not every time they ignite it, and combustion or flame is not a necessary condition for the ignition of the battery and the major triggering reaction happens inside of the cell, the battery cell.

Xinyan Huang:

I think we still know very little about what's really happening within the cell. There are different kind of theories. People say maybe it's some hydrogen attacked the cathode materials. Some people may think it's metal reactions. So it's definitely different from our conventional combustion fire. That is basically chemical reacting flow. We don't really have that flow inside of the battery cell. So that's the reason some people don't like the words of ignition. They use the thermal runaway. I think that's a reasonable description because ignition is one potential thermal runaway reaction.

Xinyan Huang:

So how to trigger that ignition? I would say it's so different from our conventional fuel, from the gas fuel, liquid fuel, solid fuel there's no clear concept of ignition temperature here. Solid fuel there's no clear concept of ignition temperature here. So that makes us have to think a new approach to define that process, how to trigger the battery thermal runway or battery ignition, Because we don't know too much about the chemistry. It's different from the solid fuel, like pyrolysis dominated, so there are different stages or different steps of reaction inside the battery cell and they generate a lot of gases and these gases take some time to be released to the environment. Then you may have a second, the real combustion ignition in the gas phase. So I think what we really care, or most research conducted so far, is how to trigger the battery reaction inside. There are limited research talking about the ignition of the gas released from the battery.

Wojciech:

So perhaps, indeed, how you described it, the ignition is not the most perfect word to start, it's just a word that any fire scientist is familiar and keen to use Thermal runaway I also like that, but I think it describes two separate things. So one would be a thermal runaway of a cell, where it just heats up and ends up as a burned out cell, and another thing would be a thermal runaway of an entire battery pack, where we would be talking more about the propagation of the event through multiple cells. So I also think this would be completely two different mechanisms. But on one hand, you can just even though I don't have a good word for it I imagine that I have a battery in a state in which it is stable, in which it just, you know, gives me back energy, takes energy to charge itself.

Wojciech:

The chemical composition for it is as designed, there are no things happening with it, and suddenly I put it into a state which A it becomes unstable. So stuff happens to it Chemical reactions, mechanical damage, energy release. There are multiple phenomena which we'll definitely talk about. And the second thing I feel is irreversible from that point it's not that oh, I can like quickly cool it down and it's going to be back to its normal state. So I think it's like taking something from a stable phase into like irreversible chaotic phase but we really could use a good word for that transition.

Xinyan Huang:

Yeah, I think you are touching a really good point. In fact, most of the combustion reaction is irreversible, especially in the solid phase, I would say. But there are more things that happen inside the battery. In fact, every time you charge it, you discharge it, you generate something new inside these battery cells, some like lithium, metal particles. This could be the major triggering for future more intense reactions, but I would say, every process is in general irreversible in that point of view. Of course, you cannot return those batteries.

Wojciech:

Who has went to the, or perhaps, if you find a way, you would be a very rich person.

Xinyan Huang:

Yes, that's a very challenging part.

Wojciech:

Okay, so let's talk about how do we put those batteries into a state at which those events would come into play and they would cascade. I will start with one nail penetration and you go further, so let's explain. How do you actually?

Xinyan Huang:

do it. So neo-channel penetration is one of the most widely used methods to trigger the battery thermal runway. It's reaching many different standards. It has its own good I would say quality or perspective, because it's relatively repeatable in that sense and it also mimics the real battery damage. For example, if you have two electric vehicles collide so you may have something inserted into the battery cell that triggers the thermal runway. So the nail test somehow mimics that kind of mechanical damage to the battery. But on the other hand if you put a nail into the battery, of course it will change the heat transfer process inside the battery. In our experiment we found if you put a very big needle you actually will cool the battery during the penetration process, especially if you use a copper one which has a very large thermal conductivity. So you actually cool down the battery during the nail penetration. So on the other side you also have the short circuit you may generate with the metal needle. So we also tried some non-metal needle to infer to the battery. You get very different phenomena.

Wojciech:

Really so. Okay, it's such a simple thing, a nail penetration, but there's so much aspects to that. In one of your papers I've read you also care about one, the depth at which it goes. Two, the velocity at which it goes. Three, also the battery orientation. So it matters if you nail it from up down or from the sides.

Xinyan Huang:

I guess Now you bring more elements yes, so our original idea is for many battery jelly rolls there are multiple layers, so we want to quantify, okay, how many layers the nail breaks can trigger that some wrong way. Maybe there is a certain limit. But during the we found out it's so difficult to control that nail penetration, especially when the battery has a metal shell. You actually have to break that shell first and during that process it's not perfect. The battery may be deformed, so the jelly roll inside also deforms, so eventually it just makes us so difficult to control how much we actually break the layers of the roll. So finally we can only get a rough estimation. But definitely the penetration depths will also affect the thermal runway triggering.

Wojciech:

So it's not enough to just break a single sandwich of cathode anode and the separation between them in the battery. You have to get multiple.

Xinyan Huang:

Yes, so definitely one single layer is not strong enough to trigger very intense. So in a wrong way, if you break more definitely, what's happening is more dramatic and you have more gas reaction.

Wojciech:

You start the chain reaction in a more intense way and how about nailing from a top or from the side? So if you nail from top, you not go that much through the layers, right, so you just squeeze them, or what? What happens then?

Xinyan Huang:

so this is another very important point. So when we think about a battery, this is a three-dimensional object, so the thermal runway can also propagate in different directions inside the battery. Sometimes it can propagate along the Jerry Roll, sometimes it can propagate across different layers. The speed is actually different. So if you penetrate the battery from different directions you can get a very different result.

Wojciech:

And to bring this more into relationship with the real world. Does different nail penetration velocities or depths, does this reassemble some real world scenarios or is it just a way of parametrically studying it?

Xinyan Huang:

So I guess every kind of standard test mimics some real situations, but during that simplification of course you ignore something, you deviate from the reality. But nonetheless these kind of standard tests still provide some good data for you at least to compare different batteries under different conditions.

Wojciech:

And we were just talking about a single cell right now, and that's also how we do it in our laboratory. We have a nail penetrator, a very sophisticated piece of equipment for something that really is a nail, a very large cabinet for a single nail to house. What if the cells are assembled into a battery and that battery is, you know, packed now within a case enclosure, and those, of course, can vary Like. You can have those packs that literally look like wrapped in a tin foil. They're very loose, something that you would put as a replacement package in whatever power bank you have. You could have a power bank device that you charge your phones in a plastic casing and you can have a structural floor of a car vehicle shielding those batteries. In the nail penetration test you only test the cell cell. Have you done any tests in which you would penetrate through a casing? And again, would that change the outcomes or so overall, this is a very challenging question.

Xinyan Huang:

In fact, battery is not a material. It's different from like a certain plastic or like a wood, it's an assembly itself.

Wojciech:

It's a device, right?

Xinyan Huang:

Yes, it's already a combination, even without complicated cases outside, because all the batteries they have films or case. It's a shell itself. So you have also multiple layers, different materials mixed inside, also some mechanical devices maybe to preventing or facilitating the venting. All these are actually really complex and of course you can go and try to measure the material property of these batteries, but that kind of experiment is very tiny, small scale, maybe some gravity analysis. If you only test the material, you're also not closing to the reality because there are more than 10 different materials inside a battery. So we don't really have a good test between the cell scale and the material scale. I think that's something research can be looking to it. We need some better test to quantify to avoid the complexity, but not just focusing on one or two materials.

Wojciech:

I was also asking from the perspective of the thermal runaway between the cells, because also the device is designed in a certain way to be able to take some heat away. Perhaps over-ventilate, over-pressure created by batteries. If you mechanically abuse the casing you also create new pathways in that system that perhaps change. But I guess that's a level of complexity we're studying. It will be very hard because of the numerous ways it can go. What do you do?

Xinyan Huang:

Machine learning- so indeed, if we look at the battery like a regular fire in a room. So if you put a single battery cell into a battery pack or a battery energy storage container, you are like putting some fuel into the room. So you have to consider the built environment, ventilation, the radiation, smoke, movement, so that also creates another level of complexity. You have the safety of the battery itself, you have the overall environment, whether that's promoting the battery fire or kind of like stopping the battery fire.

Wojciech:

Yeah, very good. How about different mechanical abuse tests, Because the nail is not the only one. Are there other ways, like dropping the batteries, breaking them, half-squeezing them, hitting them? How do you abuse them mechanically besides nail?

Xinyan Huang:

There are some standard tests where you can squeeze the battery to make it deform. I think some companies they also do the falling test, trying to throw the battery from a certain height and see how the battery will react. But these kind of tests are kind of like random. The repeatability is no better than flipping a coin. So that's the reason it's not written in many standards of these throwing tests. And I think the battery shape, the battery type is also not so standard. You have a big one, you have a small one, different chemistry inside as well. So I'm not sure if there will be some mechanical way to testing it.

Wojciech:

But to create some effects inside, those irreversible effects, whether they are like immediately, because I also know from my colleagues from other laboratories that you sometimes would abuse those batteries in different ways and they would, let's say, not go off, and then you put them in a safe space and they can go after three days, five days.

Wojciech:

There's famous cases of vehicle fires where the vehicle would go to a scrapyard and start going off multiple times after a very long period of time. So definitely some irreversible, like cascading effects are happening inside the battery, but it's not immediately. Perhaps the velocity of those changes is is very slow. I wonder, like, what mechanically has to happen inside the battery, like, do you have to break the separator? Do you have to physically change the condition of the cathode, anode, or or just, I don't know, squeezing layers, making them closer to each other, is is enough? I just wonder if, like high enough g-force acceleration on the battery, like you imagine, the vehicle is going into a crush and you just abuse this battery, but by a very rapid acceleration, but perhaps not mechanically breaking it At what point it becomes unstable and dangerous.

Xinyan Huang:

Oh, that's another hard question I only have a hard question.

Xinyan Huang:

Yes, I have to think more about that. So in general, for example, if you have a very heavy battery, if you throw it, then its impact definitely is larger. And internally, what's happening internally? That's something difficult to observe. Of course you can use like X-ray to scan it, but to capture a very dynamic process with the battery is moving. I feel that's quite challenging and I think a lot of things can be only guessed, because once you trigger the thermal runway, the battery basically burns out. You don't know what's going on at the beginning. So some better measurement mechanism is also important, but I would say, very challenging, because the battery file always destroys the device if you put anything too close.

Wojciech:

It's a challenging question, but I would call it a million-dollar question because we need to understand those mechanisms In this particular point of time, between the battery is abused and the battery is damaged up to a state where no diagnostics can be done. We need to understand that part to be able to figure out how to rule out dangerous batteries from the useful batteries. You know, because it's also from from the perspective. I mean batteries, energy, energy storage systems. This is a tool for sustainability, right that that's the reason we have them. And it's very unsustainable.

Wojciech:

If you throw out every suspicious battery and I also have a feeling that we're in Europe, in a fairly rich country, we can afford throwing those batteries, but someone will make a business out of taking those batteries that we throw away and just send them to some less fortunate place in the world where they desperately need devices like that, and we'll sell them there as a new device or as refurbished device without touching it, just taking the risk which we are not comfortable taking. So I think it's it's very. Do you know any diagnostics after mechanical abuse that you can use? Are you, if you abuse a battery and it doesn't go off, what to do with it?

Xinyan Huang:

That's something we have trying to look into it. So if some battery, you're trying to penetrate it but it didn't go off as we expected, then maybe you can peel off and check what's inside the battery and sometimes you can see some area was triggering some reaction, some did not. You can also do some element analysis and see what kind of element increased, what chemical increased. But I feel what's really challenging is not the abusing by purpose, Because if you hit a battery, if you hit a battery, you expect it will go off. But in many cases you are driving a car, driving an electric vehicle, and suddenly self-ignite.

Xinyan Huang:

That's something we're not really doing, the abusive condition. But still, the battery goes somewhere wrong. It goes to the fire. That's the challenge part. And something may happen inside a battery and that thing may be too small to be observed. That's the challenge part. And something may happen inside the battery and that thing may be too small to be observed by the current X-ray or like the other diagnostic method. If there are some good way can detect these small changes inside the battery, for these non-abuse conditions I would say this is a billion-dollar question.

Wojciech:

I'm interested, though. Well, I'm interested, but not for this reason. Okay, let's move into different abuses, because we just covered the mechanical abuse very widely, but there are different other abuses Electrical abuse, thermal abuse. Which would you like to follow?

Xinyan Huang:

So I would say maybe just heat. I think heating is still the most widely used method to trigger the thermal wrong way. And when I say heating I say the external heating. And of course you can use a flame to heat the battery. You can also put some electrical coin to heat it. We also try some laser heating, trying to heat it remotely. These are all good ways to do the test, because when we do the test we want to do a repeatability check. If you have a good method of repeating the test result, that's a good method to trigger the thermal runway.

Wojciech:

Okay, but you said there's no ignition temperature, but is there a temperature at which you are fairly sure that the thermal runaway will occur in the battery when you're hitting it? Are you aiming at some specific temperature or are you noting down the temperature at which the thermal runaway did they start?

Xinyan Huang:

I think if you really want a number, I would say 200 degree is a good number to start with and, of course, what different batteries.

Wojciech:

It can vary and what we need to know 200 degrees at the electrolyte level at the casing in surrounding.

Xinyan Huang:

That's a very good question, because you only measure the surface temperature.

Wojciech:

Surface temperature okay yeah.

Xinyan Huang:

So you don't know what exactly temperature inside. Of course you can do some reverse modeling trying to figure out the internal temperature. But that also makes that number varies a lot because if you have a large battery, so heat transfer process inside is more important. It's different from the conventional ignition of the solid fuel because the surface temperature is usually higher, highest point, and that's the point you release the flammable gas out. But for the battery they have to escape from its own cell first and that's for the second ignition process of the gas. But for the internal process if you heat it you are heating it from externally and you don't really know where the initial reaction happens inside.

Wojciech:

and you don't really know where the initial reaction happens inside. Okay, heating up. It sounds very simple, but we are fire scientists, we know the intricacies of heat transfer problem. So if you put a battery inside the flame, you will have extremely strong convective heating, extremely strong convective heating. If you wrap it in a coil, you will have very uniform boundary condition around the battery. Right. If you point a laser in it, you're going to have an extreme gradient between the heated point and the surrounding. So at this point do you think uniformly the battery must heat up to some point, or is it like local damage that triggers this reaction? If you could compare those three heating methods, like something in the direct flame contact, something heated by a laser and something heated by a coil, do you have the same observations from those three methods or these are three completely different scenarios?

Xinyan Huang:

I guess the heating method matters Also. The battery cell itself also matters. If you have a metal shell, like a cylindrical battery, that metal shell actually helps you to uniform the heating process.

Wojciech:

Was the metal shell made of steel?

Xinyan Huang:

Something like that, so definitely not aluminum but relate to the steel.

Xinyan Huang:

Okay, Sometime. Maybe nickel, yeah, so have to check the detailed components. But in general, if you have the metal shell, the overall heating will be uniform. In fact the flame heating is also quite uniform because it's convective heating by the flame and it's also widely used in many reactors. Flame provides a very uniform heating, but the problem with flame is just too intense. So if you want to have a relatively gentle heating, you can put that battery into the oven and using the oven to heat it convectively plus some radiation. And if you want to use a coin, okay, coin has its own problem because the coin itself is very hot, hotter than the. So of course the place near the coin is much hotter than the other places. And we also try the laser. Of course, if you near the coin is much hotter than the other places and we also tried a laser Of course if you heat it with a laser, then that laser point is also much hotter. But I think a laser is one of the ways you can really remote heating it, not touching it.

Wojciech:

Why are you heating them with lasers? Is it going again into your theme of playing with batteries, lasers and batteries.

Xinyan Huang:

There's better than batteries yes, that's, that's started with the research, started with the farm. Because we have a laser in the lab to measure the smoke movement, okay. So we say, okay, we have this laser, why don't we use it to hit the battery? See, something interesting happens. And in fact the laser power also matters a lot because there are some research. With super strong laser they actually just break some battery cell, just heat inside directly. So we use a relatively weaker laser so we can heat it externally without breaking the battery structure. Okay, cool, cool.

Wojciech:

And if we could relate those heating conditions into real-world scenarios, I guess the coil or slow oven would reassemble just omic heating of the battery during their normal use and perhaps heating up to high. How would you relate those modes of ignition by thermal abuse into real world scenarios?

Xinyan Huang:

So I guess, for example, many people use a hot plate to heat the battery. Yes, so the hot plate could act as a battery that has already gone some wrong way, so you can use that to trigger the.

Wojciech:

So it's like a cascading effect within the module of the battery, from one battery to another.

Xinyan Huang:

Yes, but eventually you need to get a good heat flux, the overall heat flow. If you can mimic the heat flux of batteries that already went through some wrong way, you are basically reproducing that process very well.

Wojciech:

And the flame contact is like when the battery goes off completely and there's this flame bloom emerging. Is this a scenario that you're looking?

Xinyan Huang:

for this is very challenging, of course we know. For example, when you have an electric vehicle burning, the fire may not start from the battery. It may be burning, maybe the tire, maybe start from the battery. It may burning the, maybe the tire, maybe some other plastic inside, so they generate some fire to heat the battery. Yes, if you use a flame to ignite the battery, you can mimic this situation for the battery cell. Of course it generates a jet flame, but that jet flame does not last forever. It it's like sometimes pulsating. So you have some jet flame for 10 seconds, then they stop, then after another minute you have another jet flame. So it's quite difficult to really reproduce the process. But getting the heat flux is important so you can simulate any kind of heating.

Wojciech:

And here again, if we talk about realistic scenarios with those, if it's just a cell that you're studying, that's of course simple. But in a real battery, like a vehicle battery, you would have some systems to mitigate those heat transfer phenomena between the cells and also hopefully to limit the chance of telmo runaway of the entire module. So you're kind of like playing against those systems in your scenarios. When you design the experiments, do you take that into account? Or you use those experiments to scout the safety systems of the batteries or it's just a nuisance for you?

Xinyan Huang:

This is definitely some research needed in this area because we have multiple very mature thermal management process. That is, for example, the cold plate. Sometimes you use the phase change material around the battery or simply air cooling. That is also very efficient. But these cooling methods are not really used for stopping thermal runway. They are thermal management. I guess we cannot ask too much. We cannot expect a thermal management system to control the thermal runway. But in case of the thermal runway happens, can this thermal management system also slow down or suppress the thermal runway process? That something can be studied.

Wojciech:

So we've talked about igniting by heating up, but is it possible to ignite by cooling? I, you know, I'm living in Poland. I keep my electric bicycle on my outdoor terrace because it's a better place for it than on my evacuation pathways. That's what you all told me to not put it inside my living room. So did anyone study like freezing as a condition to ignition, or it's just unlikely?

Xinyan Huang:

Okay, that's in general. I would not say you can cool it to have some wrong way. But I guess if you put the battery into a very cold environment and you use that for like some time discharging, charging it, I think things inside the battery may go bad. So that definitely increase the chance of having some wrong way. And we all experience that if you use the battery in a cold environment the it does not last very long. So definitely something happens inside. Whether that will have permanent damage to its structure, I think that's something to be studied. And also people are proposing using liquid nitrogen to cool the battery for when you have some wrong way. But what if these liquid nitrogen when they cool the battery, they you have some wrong way? But what if these liquid nitrogen when they cool the battery, they also change something inside? That may also increase the safety risk of the future use of that battery.

Wojciech:

How realistic is cooling and reacting to that thermal runaway process, because I also know a lot of people would propose some suppression solutions. There are discussions about how do we extinguish lithium-ion batteries. I guess, again, if we're talking about five cells spread on a desktop of a scientist and you just spray water on them, it's pretty okay to cool the surrounding batteries. But if you're talking about the product, the module, a power bank or whatever, with all all the complexities, the casing, the internal management, is it realistically possible to have a meaningful cooling effect from?

Xinyan Huang:

outside. Another wonderful question Also. Let me think a little bit more about this problem. So how do we really think about this process? So we are all fire scientists, so most important thing is never about suppression, it's about preventing. Or if you want to suppress it, extinguish it. You do it when the fire is very small. So if I want to make an analogy here, I think you can treat the thermal runway the same as the flashover of that room. If you have the cell go through the thermal runway already, that cell is saveless, so you don't have to spend time to save it. So what you really care is whether that thermal runway will go to another cell. That's what you care most. So if you really want to do something, you have to do it before the thermal runway happens. For example, if, say, 200 degrees is the temperature of the thermal runway, say, 200 degree is temperature of somewhere around way, then you need to do something before the battery reach that critical 200 degree Celsius. And that's more important.

Wojciech:

And I know you also had those research on optimizing the cooling of batteries with those. It's a really good paper. I recommend I'll put it in the show notes. I'll put most of your battery papers in the show notes. They can only take 4,000 characters, so I hope it will be sufficient. You had this paper on genetic algorithms optimizing the cooling pathways in the battery, like how to shape the channels and the heat sinks inside the batteries. Is this only for thermal management? The heat sinks inside the batteries Is this only for thermal management or is there a realistic chance that they could also act as this additional layer of reaction to the processes inside the battery or perhaps in between the state of normal operation and those 200 degrees? You know where you would be in a thermal runaway. You also consider them for that?

Xinyan Huang:

so in general, there are different ways for cooling. I personally think the liquid immersion cooling is definitely most efficient. That can basically stop what was liquid. The liquid could be some I don't know the exact name. They used to cool the CPU chips, the GPU chips, so these are inorganic liquid. I'm not talking about the waters, of course water is a good way to go to be filled with water, right yes that's so these kind of liquid is.

Xinyan Huang:

they're not conducting electricity. So there are a bunch of group of liquid like that used, but they are conducting electricity. So there are a bunch of group of liquid like that used, but they are expensive, I would say. And there are battery storage units in China has already using the immersion cooling of the entire container with that liquid immersion cooling system, but of course it's super expensive. I'm not sure if it will be widely used in future, but regarding the cooling performance that's definitely the best.

Wojciech:

One more thing for thermal management, because there's also like a good temperature of operation for batteries. I know that if you want to use this ludicrous mode in Teslas, you have to like set it up before, because it has to warm battery to, you know, be able to deliver this much power. Is there a specific temperature at which those batteries are supposed to operate? I mean elevated temperature, and, uh, what's the role of the cooling system back then is just to to prevent it going beyond that point.

Xinyan Huang:

So in general the thermal management system is different from the thermal safety system. I would say so for the thermal management you have like a peak temperature you want to prevent to reach. You may also have a maximum temperature difference you want to control it. So these values are relatively low. I would say 45, maybe 50 is already very high. So the thermal management system wants to control everything below these values. But accidents always happen. You may have a collision, your car may be burned by the car next to you. So these extremely heating or mechanical damage condition, then you are not really using thermal management system you need, as a file safety system to handle these problems.

Wojciech:

Okay, and let's now move to the electrical abuse. So all the concepts of overcharging, discharging, short circuits and so on, Can you tell me about those? Are you using those in your laboratories and what's the angle on those Again, how they relate to reality and how useful they are, how repeatable they are?

Xinyan Huang:

So in general some batteries have the protection mechanism for external short circuit. I know most of the cylindrical batteries they have that mechanism. They were tested whether they will go off under the short circuiting situation.

Wojciech:

So in the system, not in the battery.

Xinyan Huang:

Short circuit In the battery. So I think the manufacturer tested that if you put a wire outside of the battery, whether it In the battery. So I think the manufacturer tested that if you put a wire outside of the battery, whether it will go off, okay, so it's just okay, but not for all the types. I think some prismatic battery they may not have that design mechanism. So still, external electric short circuit still can happen. Some other potential situation maybe when you are, for example, have some transformers nearby, you may have a very strong arc and that may also hit the battery. Or sometimes you may consider the lightning strike of the batteries. These are the unknown areas. I would say they definitely will trigger something in the battery.

Wojciech:

If you have a battery storage facility hit by lightning, there's a huge uncertainty of what to do with them.

Xinyan Huang:

That's some experiment I want to explore, but so far too dangerous.

Wojciech:

What about overcharging? Is this something done for the purpose of changing the chemical composition in the battery or doing some reactions inside? Or is it just a way of introducing ohmic heating and it could be considered another thermal abuse mechanism in here, or a combination?

Xinyan Huang:

I would say I don't know too much about this protection mechanism for overcharging, but I would say if you overcharge the battery, definitely you are making it more dangerous. So I know some battery can be charged 150%, 200% SOC, so you're definitely pushing the limit of the safety and some other things. To be cared more is the charging device, because we see so many battery fire not really happens in the battery itself, it's happening in the these cheap, poorly made as the charging device and these can cause a cable fire and also potentially trigger the battery fire and if you're charging in a relatively unstable current, unstable voltage, definitely reduce the life also the safety of the battery cell.

Wojciech:

You've wrote about state of charge. Actually I was supposed to ask that even at the stage of nail penetration state of charge is a variable in those mechanical and thermal abuses. A big factor, right?

Xinyan Huang:

Yes, so we suspected that if you put a metal into the battery you cause some internal short circuit that may accelerate or decelerate the thermo. A runaway process, but overall it's too complicated because you cannot really extinguish what happens inside.

Wojciech:

Okay, and finally a real world scenario. So I know FS1 is doing those experiments where they do flashovers in the rooms and they have scooters next to it. We intend to do some back-trafficked experiments with batteries present in the fire scene. So batteries, actually as a secondary ignited item, Is this again something that you study in your laboratory or is it within your interest? And if so, how do you deal with that?

Xinyan Huang:

So battery fire is very dangerous. Even for us, very experienced fire experiment list, we still feel it's quite dangerous to do large-scale battery fire tests in our lab. So usually you need a relatively large testing facility to make sure everything is safe. Then you do the test and of course if you want to burn an electric vehicle, that's another level and definitely you need to consider more. But of course the burning behavior, the combustion behavior, is way more complex.

Wojciech:

I guess we'll be wrapping it up because we're running out of time, but I think it was an interesting discussion, way more complex. I guess we'll be wrapping it up because we're running out of time, but I think it was an interesting discussion about abusing batteries in different ways in order to play with battery fires. So maybe for the final thought of the interview, could you give me examples of the new research directions for your group in terms of firing More into experiments, more into modeling. Share a little bit. What can we directions for your group in terms of FIRI More into experiments, more into modeling Share a little bit. What can we expect from your group.

Xinyan Huang:

So one of my current research is we are proposing some potential standard test that can bridging the micro scale material test to the cell-based test. So I feel something missing between these scales and because if you go to the material scale it does not reflect everything. It just shows that single material, but if you go to the cell scale it's already too complex. So we need something in between and that could be a standard test used for future battery development. So that's something I'm currently working on.

Wojciech:

And the conference. Let's tease in the conference.

Xinyan Huang:

Yes, so we are organizing the fourth International Symposium of Elysian Battery Fire Safety Conference in Hong Kong from this October 30th to November 2nd. So this will be the largest battery fire conference in the world and it has once every two years and we have so far received 160 paper submissions to the conference and many from the industry. So I believe we'll have very good discussion in the conference and many from the industry. So I believe we'll have very good discussion in the conference and discuss about the future of battery fire safety.

Wojciech:

So everyone's invited to Hong Kong. I think we sent the paper. Kuba, did we send the paper? Yeah, kuba, thumbs up. We've sent the paper. So hopefully, if it's good enough I don't know, maybe we're bad, but we'll see if it's good enough. I don't know, maybe we're bad, but we'll see Hopefully see you in Hong Kong and it was a huge pleasure to host you in the Far Science Show and even a big pleasure to have you as a guest researcher, visiting professor at our Building Research Institute, itb in Poland. What a two weeks it was. Thank you, Xinyan.

Xinyan Huang:

Thank you for the invitation. It's about the new research.

Wojciech:

Yes, Thank you, and that's it. Thank you for listening. I really enjoyed the in-person interviews. I need to do more of those.

Wojciech:

In two weeks I'm going to SFP Europe conference in Edinburgh, so perhaps there will be a chance to conduct some in-person interviews. We'll see. I'll try to line up some guests. Maybe that's an easy way for me to build up the library of interviews for the podcast, who knows?

Wojciech:

Anyway, coming back to this episode with Xinyan, it's difficult to talk in great depth about all of those things, so we chose to rather give you an overview of the methods that are used by fire researchers, fire scientists, and an overview of the factors that influence them scientists, and an overview of the factors that influence them, and perhaps try to create in our minds some representative scenarios of what those abuses represent to.

Wojciech:

And from this perspective, I think we've done quite a good job giving you a very broad overview of all the test methods that are used in the fire community. So when you read the research paper and you see that there was a fire underneath the battery, there was a nail penetration, there was overcharging or there was a coil attached to the battery into the cell inside the battery, now you know what it means and what challenges are related to that. And one more thing cnm te is the conference in Hong Kong at the end of October, I believe 30th October, 2nd November I could get that wrong, but more or less that time. Halloween time in Hong Kong Sounds not bad. If you feel like you're interested in more knowledge about battery fires, you will find a ton more in there, so a lot more opportunities to learn about battery fires and uh yeah, what.

Wojciech:

What an interesting problem we've been given by the industry to solve. Fire engineering is not going anywhere. We're more needed than ever. Thanks for listening to the fire science show next wednesday another piece of fire science going your way. Thanks for supporting me, thanks for being a fan of this show. Thanks for talking to me and sending me emails. I really appreciate that. I appreciate you listening and I hope you enjoy and benefit from listening to the show. See you next Wednesday. Bye, thank you.