118 - Different batteries different challanges with Francesco Restuccia

I can see that in the statistics for the podcast and I'm very keen to learn also a lot about that , because it so closely relates to my other works in tunnels , in car parks and the general problems that I'm meeting in my own engineering every day . So I'm very happy to present you another episode on batteries .

This time I have invited again Dr Francesco Restutia from Kingsgloves , london , and we will talk about how different types of application require different chemistries , different types of batteries , how different batteries perform in a different way and what that means for fire safety .

How do you handle fire safety of these various types of batteries , how different industries end up with different systematic and legislative solutions that perhaps already create different regimes of safety in batteries .

I'm not even sure if it is okay to simply talk about battery fires anymore , because the battery you would have in your car , the one that you would have in your energy storage , the one you have in your phone or the one you have in your scooter or hoverbolt which does not hover , by the way these are completely different devices . So a very interesting world .

Some new information in this episode , a good summary of what the problem is at the level of a single cell in a battery module . I hope you will enjoy that . So , yeah , let's spin the intro and jump into the episode . Welcome to the Firesize Show . My name is Vojci Winkżinski and I will be your host .

Firesize Show is produced in partnership with Offer Consultants , a multi-award winning independent consultancy dedicated to addressing fire safety challenges . Offer is the UK's leading fires consultancy with the globally established team . I've just learned that OFAR won the Small and Medium Enterprise of the Year award in the 2023 Engineering Talent Award .

I've met a lot of OFAR's talent and I understand why this award went to them . So huge congratulations to the OFAR Very well deserved award . And if you are a talent that would like to join such an esteemed team , ofar is always looking to hear from industry professionals who would like to collaborate . Get in touch at OFARConsultantscom .

And now let's see what's new in the world of batteries . Hello everybody , welcome to the Firesize Show . I'm here today , joined again by Dr Francesco Rastuccia of Kings College London . Hey Francesco . Good morning Ortrec . Thank you for having me .

Hey man , good to have you back on the podcast Last time we've discussed a lot of chemistry on battery fires and today I guess we do a follow up on the battery issue .

But before we jump into more interesting things on how the batteries evolve and how different devices use different batteries that are characterized with different safety , let's say , or different hazards , let's quickly recap how a battery burns , because someone listening may not be up to date . So tell me what's inside the battery and why the hell does it burn ?

Most of those cylindrical cells have a cathode , a nanode , some kind of electrolyte in between . You charge the battery so it has an energy capacity . The bigger the battery usually , the bigger the volume , the bigger the amount of energy it can store .

That's not always the case and maybe we can go into that when we talk about the limitations , but it has a certain amount of what are stored in chemical energy . Now that chemical energy is within the battery , the materials within the batteries react and as you use the energy there is some internal resistance in the battery .

So if you're charging or discharging a battery you also produce some heat .

Always there is some ohmic resistances Now within the battery , when you're doing discharge and discharge , the battery can start having internal reactions that change the structure and so you can start having a CI decomposition , so you start having a layer build up on , you can start having you have a separator in between , which is usually a polymer , so that polymer

could start degrading over time and that electrolyte that's in it is also a flammable mixture . Now those reactions that happen are usually exothermic , and so if they're happening at low enough rates , then the environment cools them , or if there's active cooling , they're cool , the battery keeps its temperature .

But if your battery starts having temperature differences one , the power that you can deliver changes , and two , you start developing bigger resistances through , and so you're going to start having different currents at different parts of the battery . That starts really exorbitant , those exothermic reactions , and then you can have failure .

Now failure can also happen from so that's just internal failure , so you start creating small electric short circuits . Could also happen from mechanical failure , puncture or damage . And it can also happen from external behavior . So for example , heating . If it's heated outside of a usual temperature behavior , then it can fail .

Any of those scenarios can bring you to failure , which usually comes with thermal runaway of the battery and then ignition and then some kind of either vapor cloud .

But all of the components generate heat , and so you can have heat generation from the anode and the cathode . It's just at what temperatures do they happen ? So the first ones is , yeah , the electrolyte , and then you start having a cathode and anode , and all of them generate new gases as well , and so the gas that is generated from them will vary .

Now , obviously , the cathode material can change between different batteries and that's why you read about LFP batteries and MC batteries , and those are all different cathodes , and so those cathodes will have different characteristics .

So maybe you can tap that into some specific chemistries that are common .

That's a good point , so the number's in front to tell you the dimensions . So 18650 just means 18 is the diameter . So the first numbers are the diameter of the battery and the second one is the length . So it's 18 millimeters in diameter and 65 in length .

Another example , another classical one we use for electric vehicle is the 21700s , for example Tesla , and the 21700s means 21 diameter , 70 in length , and then the biggest one .

So if you look at , maybe when Tesla made that announcement two years ago about their scaling up of their new battery for Tesla Y , which was the 4680s , and that's 46 millimeters in diameter , 80 millimeters in length . So these are getting fatter and fatter as you increase the energy density . Now , increasing that dimension has its downside .

So remember I said that you can have different generations within it . So if you have an 18650 , the diameter is small . Normally you have tabs at the end of the battery and those tabs create some resistance and actually that resistance brings you in homogeneous current distribution .

So actually , as you get bigger and bigger with the cell , that tab on the top would make a big , big difference because it's a huge , huge inhomogeneous electric current distributor , and so as you get fatter , let's say with the batteries , as they get wider , then you would have a bigger inhomogeneity so the batteries actually might not have enough energy .

And so there are some moves that have been done into that . So , for example , the big announcement from the Tesla one was it's tabless right . So the big big fat one they've made now doesn't have a tab end , but that actually does have a tab . They call it tabless because the tab is actually the entire top and the entire bottom .

So it looks flat to you because it's the entire surface , but that means you don't have this little bit on the top . So if you look at a battery , you always have that tiny bit on the top right .

That's it Total energy storage capacity . So it goes up quite large scale . But again , if you have that inhomogeneity , then you start having issues when it comes to charging because you cannot charge it fast enough If you start having those current gradients here .

So then you have to worry about cooling , you have to worry about tabs , for example , but as you get bigger and wider , usually you get more energy in , and so , obviously , depending on what you're trying to use it for , you then first choose a size that gives you the energy capacity you need , and then everything that goes around it .

So , for example , if you're looking at consumer electronics and you have a phone , you don't need a battery that lasts 20 days .

So if you have something , that where you need a really flat application , cylindrical cells are not very good for that because they waste space , and that's why in your phone you don't have cylindrical cells , you have flat cells , you have prismatic cells .

So you could come up with a fantastic battery , for example , that charges within I don't know two minutes and has no fire risk , but only can cycle twice and then it degrades very , very quickly .

So it can only charge and discharge twice and then it dies off , and so you have to have a compromise there , and so different chemistries give you slightly different compromises on charge and cost , because some of them have a much , much higher material cost . And so it depends on chemistry type there which one is most useful .

The other one is the electrolyte you use inside , but most companies don't tell you what that electrolyte is , and so you can slightly change the electrolyte to change a little bit the behavior . And the third one is how homogeneous can you keep the temperature while it's charging .

So if you're charging very fast and you can keep the batteries to a very homogeneous temperature so that you don't have too much internal resistance generating heat , then you can charge faster without degrading the battery .

But if you start charging faster and you start having gradients through the battery , then it starts decomposing the battery , and so that usually is the limiting factor on how fast you can charge .

And so there are some studies that show actually that I don't have the name on the top of my head , but it was this year and they were showing that some of the cars which have gone through the fast charging cycles for their entire lifetime , versus cars that have gone through the normal charging cycle effectively , we have had the same degradation over two years ,

and that's probably because of the management system around it . So they've kept the temperatures constant . They've kept the cells balanced .

If you were not to balance the cells between them and you don't have a good management system of the battery , so battery management system then the cells would have slightly different voltages as they're charging , you start generating different heats and they don't last very long .

So the two things I would say that let you have fast charging are , one , a very good battery management system and two , a chemistry that allows you to have many cycles , because you can have fast charging for a few cycles , but you want a chemistry that allows you to have many cycles and like consumer electronics , my phone or my laptop .

For your laptop that would not be feasible . If your laptop's battery degrades 20% a year , you have to get a new laptop in three years , and it used to be , if you think back 15 years ago , our laptop batteries would probably last a year before they get to 80% in life .

But nowadays we require them to have much longer battery lives and so the batteries tend to have a longer battery life . I think for cars it's much , much longer battery life . We think they're designed to have 12 years . So depends on your application what the lifetime you want it to be before it degrades to a certain percentage is .

And that doesn't mean they stop working right . So when they degrade to 80% they can still be used , but maybe for different applications 80% is some sort of metric , yeah . So normally the degradation is you go to the 80% . So a lot of the data is always how long will it take , how many charges will it take to get to 80% of its life ?

And so how long will it take to get to 80% of its capacity ? And that's because in engineering we tend to over design right , and so things will still work fine even with 80% .

So you can just have a very large battery system and you don't need to charge it very fast because it is very large . You can have many batteries and so you can spread the charge over many of them .

The problem there can come in maybe everything that goes around the management , because maybe you don't have a thermally controlled environment and the temperature will change rapidly . And let's say that you are in Poland versus Britain versus maybe Southern Italy , where I'm from Nevada desert , exactly .

Yeah , if you think about the energy storage fires that they've had in Nevada , for example , the conditions of the environment are very different . In a car , in your phone , you tend to be .

In a house , you tend to be in a system where you can control the temperature much better , and so I think the problem is in those energy storage systems are that they're exposed to very different temperatures even throughout the same day , and so you have slightly different issues there with control , and the biggest issue there is that if they are outside , not a

problem , you know you don't have to worry about the people inside . But if they're inside a house , you need to worry about what happens with anything that happens , with it failing or off-casting and so on , where there are humans present and maybe where you don't have all the sensors that you would have in a car , for example .

I must say . I'm talking about the batteries with people for a long , long time , but I think the importance of homogenities in heating up in the batteries and in homogenities in electrical currents inside the battery are perhaps the first time I'm really hearing about this as one of the influential factors .

I guess that's either a sign I'm incompetent or the science is advancing . I hope it's the latter .

One of our last presentations , one of my colleagues presented to me a great , great new electrolyte which is non-flammable . He says , and I said but this was made two years ago in Japan . I was like , if it is truly so amazing , inflammable , low cost , why is it not in every single battery ? There must be something else that we don't know about .

So I think everything comes at a compromise . So electrolytes , for example . These days , toyota has announced a solid-state battery . They say solid-state , you get rid of that liquid electrolyte . Solid-state batteries have been around since , I think , the 19th century . It's not a new thing . It is where you take a battery , you remove that electrolyte you use .

One of the ends is solid lithium and so you're getting rid of that flammable electrolyte . But it's one . It can be very expensive . Two , it can be very , very heavy and because you're going to have all of these solids which takes weight . The other one is sensitivity and stability .

So one of the biggest issues there is , it's very , very sensitive in applications .

So your anode , in this case , is the lithium solid and you might start getting lithium dendrites developing , then they penetrate through your separator and then you have a failure as well mechanical failure in this case and so , compared to some of these batteries , they have good volumetric energy densities , good high energy density overall , but often the geometry is a

big constraint , right , because if you need something small , something with a special geometry , you can't do it , and there is usually a big change in volumetric behavior as it's charging and discharging , and so again , it depends what you choose in the material . But there is a downside there .

But the biggest issue there was always cost , and so now that they've brought the cost down , I was very surprised to read that in three years they're planning to use them commercially for electric vehicles Toyota , I believe , in the announcement they made in June , and so it is looking like technology is very advancing there , but everything always comes at a cost .

So different materials , different space requirements , the other issue , so that the electrolyte one definitely one . The other one , I think , is we don't have a very good understanding of how to suppress . Suppression of batteries varies massively in our community .

We read a lot of papers on suppression and well , I mean , you're the organizer , you're one of the organizers for the workshops in IFSS soon , and we're having a workshop on battery fires and one of our three main topics is suppression .

Not because there is infinite work on suppression , but the opposite , because we do not know enough about suppression because it varies so much based on the application , the chemistry and so on . And so , yeah , those two topics I think we don't know much about In terms of manufacturing .

So look at the positives we have gotten so much better at manufacturing batteries because we're doing it at such a scale now that the number of failures that come from internal short circuits in manufacturing , so on , are detected well before they go to commercialization , before they go out in the vehicle , and so we are much , much better , having standardized the

product .

Because now , from what you say , it's not just removing the heat but making sure that while you remove the heat , you heat up everything around to the same temperature . So , we should put them all in the block of aluminium .

roughly , and so it's much , much easier to have that heat homogeneity . And so , yes , although designs are similar , we still do packs the way we used to do them .

We have more sensors , first of all , and second , we're much better now at the thermal management , and so I think overall , the in operation electric vehicles are maintained very , very well now in terms of homogeneity and so on .

When they're passive and they're being transported , maybe different story because we're maybe less careful with temperature , but when they're being used , you know .

And in terms of suppression , to finish the structural discussion on the batteries , I guess one thing is what to use to efficiently take out the heat and stop the chemical reactions and I assume water would still work but the other thing is how to get it in there , because I don't think the biggest challenge is with how to cool it .

The biggest challenge is how to get the cooling medium , whatever it is , or the extinguishing medium , whatever it is , into the particular place in the battery where the fire emerges .

Because , also , when you have 1,000 cells on a fire , your need to suppress is not that huge anymore , because the car is lost and you're probably managing damage outside of the vehicle , but perhaps you would like to stop them earlier .

So for liquid water injector , water sprinkler , water mist additives , foams , classic ones have been looked at , depending on the battery . One of the concerns there with liquid was heat absorption is what you get . Lots of right Liquid good , you absorb the heat . It is very good cooling capacity .

However , sometimes you're still getting emissions from the batteries that are not ideal . So another one that they usually look at is chemical suppression . So if you have a chemical suppressant , that's great because you stop the reactions early on .

But the problem with some of the chemical suppressants are that actually some of the chemical pathways that they can generate instead produce I don't know hydrofluoric acids , cfos , and those are extremely , extremely toxic , and so there is no silver bullet there . You have one . Oh , that's what I forgot as well . To mention is they can easily reignite .

So actually suppressing it is not super hard once you know what type , but they can reignite very , very easily and the toxicity of the gases around it after they've stopped igniting can be a factor .

If you're producing a lot of hydrofluoric acid , if you're increasing so for example , matter-based agents have been shown to increase the HF production then that's a concern for the people around it .

Because if you're a fire fighter and you've put it off and you're in a closed environment let's say energy storage system , you don't want there to be lots of hydrofluoric acids left over after you've put the fire out .

There's just another thing that can happen with batteries . Let's just say we get a nice diffusion flame . If I have a turbulent diffusion flame , I'm going to start having the same problems we have when we have pool fires .

Right , I'm going to start having radiation effects , I'm going to start having convection from the air around it , and that we're moderately comfortable in in fire science , because we've dealt with pool fires for a long time .

So we know that if we have a pool fire and we know the size , we know roughly how it scales , and so if we can actually from fire scientists , if we can get it to be , this is the type of flame I get . It looks pretty much like this pool fire .

It is much easier for us to understand how propagation would happen , right , because then I can think about how will things ignite near it . I remember last time we spoke , I had just finished some work with Newcastle looking at flame flares and there , you know , you start having the same concepts .

So it doesn't matter what the chemistry was , they have a three meter long flame flare . There's a turbulent flame and it behaves like a classical , you know , methane flame . I can tell you what's going to happen to the thing that's going to be impacted by it without knowing about the chemistry within the battery .

And so I think those we can separate the two problems once we know what that input is . So if I get the input from the electrochemist and the battery people and I get this as my composition , then I can look at the fire modeling .

In fact that's some of the work we do as part of a grant that we're working on here with the fire day institution is there are people who look at all the different chemistries and how to detect and how to you know instrument , the cell .

But ourselves and another group really look at using this data to then you know this cell level modeling to then look at okay , how does the thermal runaway propagate , how does the fire propagate , how do the reaction networks propagate , if we want to look at some of the detailed kinetics that say we're doing the combustion side and that's , you know , more classical

research of venue and also industrial venues for us in fire science .

You need to have this chain of knowledge passed from one group to another and I think , again , communication is the key with knowing . What do you want to know from them ? No , because the chemists may not necessarily be perfectly aware of what fire engineer needs to understand about a battery fire and vice versa .

The fire engineer may not be aware of what impacts the decisions of the engineers and the engineers of the electrolyte . That is also the point of doing podcast episodes like this one , which are perhaps not easy to understand for a lay person , but critical to see what's on the other side of the pond and what people are looking into .

So I feel connecting these fields is immensely valuable . Now the reason I've invited you to this podcast episode with different types of batteries , different electrolytes , chemistries and so on .

I had an episode with Professor Sturm about their tests in tunnels with electric vehicles and he mentioned that , to his surprise , it was very difficult to ignite the vehicle batteries . They were expecting it's going to be very easy , just over heat them a bit and then it's going to pop . And they said , oh well , it actually was not that easy .

And the second thing that triggered me was I saw some statistics out of my head . I cannot really recall the numbers . It was something like 30-ish percent of the car fires with electric car fires had a battery involved in the fire . I think it was a statistic coming either from Poland or from Europe and the number was roughly 30 percent .

So let's say , one in the three fires of electric vehicle included the battery , which means it's not that any like an edge generation in the vehicle leads to a catastrophic failure of the battery and a huge fire that we're all worried about . So I would say the things are looking quite bright in this space .

Well , on the other hand , you have like the small electric mobility devices and you hear about the tragic stories of them going to turn around away occupants of flats and in general you can see that this problem is on the grow , not on the decay , and I start wondering like did we just like split the timeline ?

You know that one industry does one thing , is it even okay to speak about battery fires anymore ? Like is this generalization harmful to some of the industries ? Because , intuitively , I see there is a difference , you know , in quality of those devices and in what fire hazard they bring . What was your opinion working on that ?

Actually Because , yeah , because if electric vehicles had so much regulations adopted to them , anything I think would above , I want to say , two kilowatt hours has a lot , a lot of regulations around it . And so you see , said versus , don't really struggle with ignition . That's usually due to the battery management system .

So you know , we've ignited a lot of batteries electric vehicle batteries without the battery management system . So you take off the battery management system , then they're just cells right , then you can treat the cells , you can ignite them .

And the reason that you struggle much more igniting a full pack that's instrumented is there is a lot of cutoff procedures well , well before the failure . And so actually igniting , you know , today's electric vehicles is much more challenging than , say , electric scooter , because electric scooters maybe have very little regulations or worse , some of them are .

You know , a lot of those fires you read about in people's homes are people have made their own packs and so you make your own pack . Maybe you've made it , even if you add a battery thermal management system or a battery electric management system , you've made it with things that are not compatible or things that are not controlling .

The circuits between them and the cells can make can have a lot of difference . So you know , we build our own battery packs here and , out of curiosity , in a very safe environment in a chamber , I charge and discharge normally , you know , low , low charge and discharge some batteries without having control between the batteries , and they started identical .

I already had a 10% difference between them . So imagine the second and third cycle . It gets bigger and bigger , and so having these cells that you've put together yourself , even in controlled environment , is not necessarily following a lot of those regulations that are there for bigger systems , and that's why I think we see a lot more failure .

So unfortunately , there it's regulation , because these are , you know , when we have new technology and new application for things , regulation is always lagging a bit and until regulation comes into place , there is more risk there .

The other one is which now I think the European Union has added is you want a battery passport , right ?

You want to know where the batteries have gone , so when you use it for second or third life application , you know exactly its life cycle , you know exactly what behavior it's gone through , and so you want the data to be there so that you can actually use it for the right application . And this is you know . We do this already in engineering .

Normally , when you have something that you're going to use for something else , you usually know its life story , and so you know if it's applicable or not to the new problem . And for the smaller scale battery systems , we don't have any of that at the moment .

So we do for electric vehicles and we do for energy storage systems , although for energy storage systems , as I said earlier , we have other issues , which is the surroundings , but not so much for smaller scale application . And those are the fires we tend to see these days .

And are there any regulations in how should you build the interior of the battery , Like how should you separate ?

I think in Europe , so in the European Union , by the end of 2031 , you want 80% of the batteries to be recyclable the materials . So that means that you need to ensure whatever you've used to make it is reusable , and so there's a lot of regulations around that .

But then in the user so let's say , you want to do an electric bicycle there is much less there , and so I think , yeah , the regulations should be there , applying to all batteries , like we've done already for waste , right . So now every battery doesn't matter what the consumer size is .

It could be your phone battery , it could be the battery you use for your flashlight . You now need to recycle it , right ? You can't just put it in the normal bin , and so we want maybe something like that for operation as well , and not just end of life , where there are strict regulations for operation , irrelevant of the size .

And why was it never an issue before ? Well , you know , for phones it was an issue 15 years ago , you know , there were very famous fires of companies we will not mention that almost bankrupted the said companies , right , and so usually it's the industry itself that self regulates because they don't want the publicity of .

I am the person , I am the company that has caused the first fire and the issue with a lot of . And so why is it not an issue now ?

Because , for example , for those companies where they've had fires and planes and they've had fires and it was , you know , almost bankrupting those companies , and the reason that we don't have this with some of these consumer systems is that it's second life and so actually it's not that company anymore and so it's people who've made their own or it's third suppliers ,

and so then the publicity is not the issue there anymore , and so it's missing that step of how do you deal with this second or third life application and when it wasn't designed for it , maybe , but we're still using it because it's cheaper to do it that way .

The battery passport is a good start , the European battery passport because now you need to know what's happened to the battery before , and so you then need to know that before the new use . But for smaller scale use , where people are just putting batteries together , it's still a very , very big issue .

And then the recyclability one is the other big , big issue , because , yeah , if you don't have a good second life and you don't have a good circular economy for that whole battery operation , then one , you're going to run out of materials relatively quickly and two , you're going to have all of this waste to deal with , which was waste that actually could have been

used either for making new batteries or for using it for lower power applications . Wojciech Mrodsik in Newcastle works on second life you know they're part of the Fire Day Institution as well and the composition of the battery .

There is a big issue as well , because if you're using different type of batteries in different applications , maybe it's not what it was designed for and so can you still use it for something that it wasn't designed for .

For most things , yes , for some things no , and so , especially when it comes to storage systems , so for me especially interesting is cross application .

Now we're kind of talking about cross application use , so you would use already decayed battery from electric vehicle , for example , and perhaps use it as your home storage when , if you just purchase a new one , you would expect a completely different characteristics of that device . Is placing these batteries outside of their intended use also a difference ?

I guess you would have to battle with the BMS because it probably would not allow you to do some things you would like to do . So we're into that , into hacking the systems and exponentially increasing the risk ? I guess yes exactly that .

But then after a few iterations or a few cycles you get into territory where it's not tolerable anymore and then maybe it degrades much faster . It becomes unsafe .

Actually , I know your group is researching a lot of that and I know you do a lot of modeling , so perhaps you can also tell me how does one model the interior of the battery , and I guess the knowledge of this in homogenous insight came from somewhere .

So the quick answer is you model the different aspects , like we mentioned before , so you can model the electrochemistry to thermal interaction . For example , you can look at how different temperatures or different cycles or different materials generate those inefficiencies .

And then you would solve mass continuity , species conservation , some electrochemistry actions to get an equation or an output of heat , and then you know what that heat is and then you use it as a feedback loop to look at how that heat would then generate something internally , and that's purely from a thermal side .

And that's effectively how we work with the battery management system . So the battery management system is actually a system that is designed to keep you within that safe operating conditions .

And so if you've done plenty of models before to look at the link between heat generation and electrochemical reactions at specific temperatures or at specific voltages or specific charges and discharge , you can do lots of cycling in a lab . Then you have those parameters to set your management system . The trickier bit happens when you go outside those parameters .

So once you go through something that's not studied , how do we model it Right ? You need parameters and you're a modeler as well . Wojcik , you also know that if you have a system where you have more than six equations , so you have an over constrained system .

You know , in these electrochemical system and thermal system , when I add fire , I think I can come up with about 25 equations . Now , if you give me six equations , I can probably fit an elephant through your equations .

If you give me 25 , I can fit your entire department through the equations that you wouldn't even notice , right , and so it's the same concept here , and so the problem when it comes to modeling is , when you have so many different equations and maybe not , and maybe gives you 25 parameters , how do you ensure that what you're designing is not only going to work

for your very specific case , for which you have validation data ? How do you ensure that you choose parameters that let you look at things that are less constrained ? And so what we do , for example , in our group , is so .

Joseen Zadeg is one of our PhD students , our PhD student in the group , and he looks at how , once you assume a certain parallelist's process for these gases , how you can get heat release rates from the flames that generate from these off-gassing and how do they match reported literature blindly .

So then you can extend it to other cases , and he's done this for different chemistries , for example . So what he wants to do is he wants to generate combustion models for a very specific cylindrical type battery and then his next steps would be okay . Now , I've done it for cylindrical . How do I adapt it to different geometries ?

Right , we'll adapt it to different geometries . The chemistry will be the same , but maybe the heat transfer might be slightly different , because I have a different geometric boundary , and that's what we've been working on .

So , starting from the assumption that battery thermal runaway can occur in all lithium ion batteries , as long as you're outside the parameters where you should be , how do I deal with the swelling , the leakage , the uncontrolled temperature increase within a model , so that it doesn't matter what caused it ?

Once I get to a certain point , I can be sure that I can model the fire or the explosion properly so that I know how to detect it and prevent it ideally , and that's what a lot of our work is .

And so what he does is , for example , he uses CFD simulations , having that chemistry just as an input to look at , one , the gas composition and two , the flame generation and flame spread , so that then you can predict will the next cell ignite or not ignite , and that can help you plan propagation .

Suppression a little bit more difficult to model because you have to add all of the different reactions that can happen in suppression . And then the biggest issue , I think , in modeling . Okay , this is my personal opinion , because there's always lots of issues with modeling , like limitations .

I mean , a model is only as good as the constraints you have for the model . And there are some amazing chemical models for combustion .

We can get more and more Detailed kinetics , but they're computationally very expensive , and so we've seen this when we are doing engines right , when we're modeling turbulent combustion engines you can get or combustors gas turbines is a great example .

You can get amazing chemical kinetics at the expense of , you know , it'll take you three months to run the model , and so if we want the system that we can Use either for live prediction or for early detection , we need something that runs fast enough , and so another thing that we do in the group is we look at surrogate fuels .

So can we look at making a surrogate fuel that represents the battery to replace that , you know , venting gas to decrease computational cost of the model while maintaining simulation accuracy . So can you get a good simulation where I don't have to worry about the internal chemistry .

I just worry about the surrogate fuel which I use , a fuel we know what lots about and will they reduce our computational time ?

Well , I want in the end is how much heat release rated to do I have from the battery , how much gases I'm producing and in what ratios more , less how much co versus co2 ?

Because in the end I must not really modeling combustion reaction like I can in my CV , but I can as well , just you know , put my own yields into the reaction and how much exactly gives , and it's fast chemistry Anyways and then you just turn it into a one-step reaction system right , and you say it's one

step .

But also a very welcome thing would be just to know the range , like what actually are the lower and upper probability limits , know how big the fire can be . Yeah , if I roughly know if it can reach a megawatt or 10 or 100 , for me that's already pretty good starting point for any analysis because , as you mentioned , engineers tend to over design .

So if I know it can be 10 ish , I can just , you know , put 15 megawatt fire inside , and if my building can handle that , it's probably good enough .

You want to know how much heat is gonna impinge on your system , and so we spent quite a lot of our simulations on looking how much radiator feet flux will reach my surroundings , for all the different Chemistry's and how do they compare .

Because if they're very similar in amount , then it gives you , as a Modular of the system around the battery , a very good estimate of how much heat am I gonna get from the system , what's the heat load that I'm gonna get , even if I'm not impinged by ? You know Conduction of it , so you know I'm not attached to a heat generation .

But if you want me to be excellent or that's a , that's a few PhDs away .

When they fail , then that gives you a range that you can apply right , so you don't have to care about what the battery was . You know I have this battery pack , this is the range and this is the error curve .

So if I have plus or minus , I can use it , and so that's something else We've been looking at again with those surrogate fuels is how does it then ? How can I use it in a one-step problem in my simulation ? So I don't have to worry about the fuel and what's the error .

So if I can bring heat transfer back in , then I can feed back to the electrochemists what feedback loop it's causing and use it as a holistic model .

So that that's something that we're working on , which I think is exciting , because it brings back our expertise in fire , some of my expertise in heat transfer as well , right , and Bringing in all the electrochemists without having to do the electrochemistry fantastic .

I many times I've said publicly that I don't think fire scientists should work very deeply on Like internals of the batteries if they are not close to the industry , because they are more likely to be like two , three generations behind . Yeah , because what you can purchase from your market is it's not really representative of the things in development .

So I wanted to understand the link in here .

So there's about 50 industry , 50 businesses that the fire institution works with . So there's about 27 universities and 50 businesses and so , yes , they are very Linked with industrial problems for our modeling work . A little bit less , but the advantage is we can get lots of data from them .

So I can go to my collaborators Within these project and say I need this and they will get me that data , which is for me a breakthrough because it means I have access to data which maybe is not our academic work , but it's practical data . This is how the problem is .

This is the data we get and then I can use it in our , you know , slightly more fundamental work .

So very , very interesting and then it actually makes a lot of sense . Thanks very much for for bringing me up to date .

And actually , was the legislative system that drives the design of both . I Would not expect that the legislative system would be that important and I didn't really understand what you can regulate in these terms , but it seems the quality control and Requirements related to the technology inside the battery packs is actually making a big difference between industries .

So it seems there are industries that are dealing better . There are industries that are lacking behind . Now we need to bring them all all up to the same point of of good , high quality safety systems . This , this looks promising .

And another aspect mentioned the Passport for the battery to know where the battery came from when you're using it in its second or third life Very interesting concept . And , in general , the reuse ability , the homemade battery kits from recycled batteries Very interesting subject . And that definitely is the next .

What good episode on batteries , which I'll already start planning now . So thank you , francesco , for this interesting discussion . Thank you for sharing the new tidbit of knowledge and for the listeners . I hope you've also got some new parts of knowledge today that perhaps will let you better understand the battery systems and how to deal with the fires of them .

So that's it for today . Thank you very much for listening and see you here again next Wednesday . Bye , bye , bye . You .