160 - Fire Fundamentals pt 10 - Flame Spread with David Morrisset

It was a very popular episode of Fire Science Show and I enjoyed it thoroughly . So David is a PhD student at Edinburgh and almost a doctor , so he's just a few steps away , crossing my fingers for his viva and brilliant future . I actually love interviewing PhD students and postdocs . You know young researchers .

It's really fun when you get to talk to legends of fire science and , of course , that's something that's also my dream to interview the biggest names . But interviewing , you know , the new generation , the ones who will become the legends of fire science , is a huge pleasure .

Anyway , with David , we've chosen to jump into pretty deep water , and this perhaps is the most challenging , most difficult episode of the Fireside Show . Yet we're talking about flame spread . And boy , flame spread is a really complicated matter , so it's a friendly piece of warning .

This episode is not easy , but here I would love to convince you why it's worth your time to listen to it . And simply , flame spread is something that drives so much of the fire safety in our buildings .

Even the simplest way how we create design fires based on alpha t squared relation , you can tie it back to flame spread , the catastrophical fires that surprised us in some way . You could actually explain those surprises by studying flame spread .

A lot of stuff goes down to flame spread and yet , because it's so difficult , it's very rare that people really understand it and in some aspects it's counterintuitive , which makes it even harder . So these are the points that we really like to highlight with David In this episode .

I usually do the summary at the end of the episode , but this time I'll try to give you a brief overview of the episode beforehand . In this episode we discuss what makes flame spread so important for fire safety engineers .

We discuss two modes concurrent and opposed flame spread , which basically means upwards and downwards , if you consider a vertical item burning , and in those two flame spreads you either have flame moving in the same direction as your flame spread or flame moving in an opposite direction than the flame spread .

So these two modes will have completely different driving mechanisms , completely different phenomena that will define or that will influence how the flame spread behaves . And , as david highlights it's . It's not that we can always focus on one and forget the other . It's not that we can focus on one simplification .

It can become pretty complex when you start having lateral flame spread and you really have to understand which mechanisms are driving your flame spread in a very specific application .

When you listen to this episode , I highly encourage you to capture those bits of the conversation what drives the flame spread , what are the differences between the mechanisms and how we can use it for our advantage , how we can use the knowledge about flame spread , this new intuition related to Flamespread in designing more fire safe buildings , and if we succeed with

that , then I'm gonna be super happy . And well , let's at least try . So let's spin the intro and let's do some Flamespread . Welcome to the Firesize Show . My name is Wojciech Wigrzyński and I will be your host . This podcast is brought to you in collaboration with OFR Consultants . Ofr is the UK's leading fire risk consultancy .

Its globally established team has developed a reputation for preeminent fire engineering expertise , with colleagues working across the world to help protect people , property and environment . Established in the UK in 2016 as a startup business of two highly experienced fire engineering consultants , the business has grown phenomenally in just seven years .

With offices across the country in seven locations , from Edinburgh to Bath , and now employing more than a hundred professionals , 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 fire safety solutions .

In 2024 , ofr will grow its team once more and is always keen to hear from industry professionals who would like to collaborate on fire safety futures . This year , get in touch at OFRconsultantscom . Hello everybody , I'm here today with David Morissette from University of Edinburgh . Hey , david , good to have you back in the show .

So today we're going to be talking about that second thing , which is well not PMMA . It's about flame spread and learning about flame spread in general . Now question is I mean , flame spread is some hardcore physics . To be honest . It's like really , really hard when you start digging that .

What's the premise of studying that for you , like , why did you go into the footpath of Fernandez-Peyo , williams and other giants to study this ridiculously difficult thing ?

What sets Flamespread apart , too , is if you look at it in terms of some of the fundamental topics that you see in fire science too , is , if you look at it in terms of the war , some of the fundamental topics that you see in fire science , whether it's ignition and flashovers or those , those compartment fire dynamics , those big core ideas , each of which have

a chapter of their own in the textbook . Flame spread is kind of one of those categories , right , sort of a standalone , quintessential topic of fire science . And so when I was starting my phd , my supervisors , angus , law and rory , had sort of sat down .

We're thinking about some of these fundamental problems that would be really rewarding to sink our teeth into , and so the one that came up was Flamespread . And if you again , like you just said , though , if you look at some of the forerunners of the field , right , you had some pretty heavy hitters , right .

So in terms of the Flamespread literature , the likes of Howard Emmons , former DeRis , I mean the list can go on and on Dibble , drysdale , and these individuals are recognized both within the world of fundamental fire science and combustion science .

So you have these amazing people looking at this problem over the last 50 years , so there's lots of good literature for anyone who's interested . Maybe we can link some stuff in the show notes . Absolutely , but there's still a lot to explore with this topic .

Right , you have to understand what's happening in terms of the heat transfer from the flame to the solid . What's going on in the solid ? The solid is then pyrolyzing , producing pyrolysis gases . Those pyrolysis gases then need to mix , and then they need to ignite , and then that flame needs to then again heat the solid , right ?

So you have this really interesting feedback loop between these solid and gas phase interactions , and then the decoupling of these problems is , you know , extremely intricate , and so understanding how does the gas phase and the solid phase interact across different configurations or different experimental conditions truly complex problem .

And that's really where things get interesting , though , from a fire science perspective .

You need to understand the pyrolysis . You can study pyrolysis on its own . It's a very valid thing to study another fundamental phenomenon , but here you have to couple it with everything else .

Pyrolysis on its own , it's very valid thing to study , like another fundamental phenomenon , but here you have to couple it with everything else in dynamically changing environment and in decaying fuel . So I I also think , uh , the interplay of physics in here makes it interesting , but it's also important , right ?

I mean , when you about it , the world is not built from liquid fuels or PM literature .

Right is actually , you can draw a lot of similarities between you know john derriss's and Quinteri's equations for solid flame spread . There's a lot of really interesting overlaps there . But , like you said , I mean most of the sort of scenarios that we're looking at most fire scientists and as fire engineers is solid fuels .

So I guess we'll probably limit this discussion to solid fuels , which I think makes sense . But a very easy link to engineering is think about one of the reasons that we care about flame spread is fire hazard in general is generally linked to the concept of fire growth .

So if you have a heat release rate right , I remember this you know something we've talked about last time I was on right and it's such an important parameter in fire science All the engineering sort of analysis that that heat release rate is going to grow at a certain rate and that growth rate is therefore a really critical input into different bits of analysis ,

right , whether it's your you know cfd modeling or looking at sprinkler activation or anything , and the concept of fire growth is fundamentally linked to the concept of flame spread , right , I mean , one of it's actually really interesting if you sort of dig back deep enough into our you know , classical alpha t square fire right .

One of the core ideas there is that you have a fire that's spreading radially , horizontally on a flat surface right , and then you , what you do is basically that time constant is associated with the idea of what are you know properties in which result in a certain way to spread .

So there is this really you know intimate link between the concepts of flame spread and fire growth .

And then if you take that and extrapolate that to the idea of fire growth is super important in the context anything to do with the heat release rate then these , the concepts that go into flame spread , are truly embedded in a lot of the things that we do as fire engineers .

And you can see this pretty clearly if you go back to our podcast episode and you have a little flashback to what we talked about last time . If you burn an upholstered chair , you can see that the heat release rate developing over such a complex fuel package can very simply be linked back to some of these ideas of flame spread .

Right , we talked about how there was a certain period of the heat release rate curve where we saw that there was flame spread upward , flame spread up the backrest of the chair that resulted in a very different growth rate than horizontal spread along , you know , say , the base of the chair .

And you know , by linking these observations of flame spread you can actually start to articulate the mechanisms that are driving the heat release rate itself .

I mean , once the fire has flash over , there's not very much you can do besides , uh , fire resistance of your , of your walls , which I struggle to call fire science . Anyway , there's not much you can do .

Right , the things that we can do as engineers are in those first phases , when the fire is growing , developing , building up , and in this period it's all flame spread , faster , slower , dominated by whatever other phenomena that we will talk about , but it's all flame spread . So yeah , I mean , even if the engineers will not understand , fully understand the flame

. That's not the point that you can derive Quintyree's model on your own . The point is to understand what makes one flame spread very different from other , like what circumstances , what variables , what physical , environmental conditions make the difference , and that's the point of doing this interview . So let's perhaps dig into that .

So first let's narrow it down to solid fuels . I think that that's a good framework to work with . So how would you even define the flame spread Like what's actually spreading ? How would you like put it into words ?

So the sort of natural way to first separate the problem is to look at the two major categories that people generally associate with flame spread , and that's post-flow flame spread and concurrent . Now what's really interesting is the physics between these are effectively actually the same , but generally we like to constrain the problems .

Just to you know , make the mathematics work out nicer . Right , and you'll see that becomes clear as we go through this . But if , for anyone who wants to sort of play along at home , you can see both of these orientations just by lighting a match , right , and that's kind of like the nicest way to sort of demonstrate this right .

So if you strike a match and you hold it upright , so that your fingers are at the bottom , holding the base of the stick , and the actual flame is pointing upward right , the flame will slowly creep down the wood of the matchstick and this is an example of opposed flow of flame spread .

Right , flame is moving down in the opposite direction of where buoyancy is entraining air right . So buoyancy is moving upwards and the flame is moving down right now .

If you strike a match and you hold it in the opposite direction so now you're striking and you're putting the head of the stick down towards the ground what you'll see is the flame will spread up the matchstick in the direction in which buoyancy is moving .

So all the hot gases , the flame itself , is all moving in the same direction in which you're spreading flame . Right , and this is concurrent flame spread . And what's first thing to note right away is , for anyone who's done this just trying to light a candle right , you'll notice that concurrent flame spread is much faster , right ?

So if you hold a match in that direction for too long , you're probably going to burn your fingers . So right away you notice there's this big difference between those two different regimes , right ? But what's interesting is the physics that actually drives the problem like doesn't actually change that much between each of these two regimes of flight spread .

You still need , basically , you have your flame and you need that flame to heat fuel ahead of the flame front , then the pyrolysis gases need to react in the gas phase and you complete that feedback loop that we talked about . So the actual physical processes occurring throughout the problem are the same .

It's just the mechanisms by which , say , the heat transfer gets to the solid fuel will change . So that's how we start to distinguish between these two different problems .

this repeats and repeats , and repeats in an endless loop until the flame has moved through the entire fuel . Is that like good enough approximation ? I ?

But in the context of , let's say , a post-flow flame spread , uh , looking at , let's say , start with downward flame spread over slabs of pmma , right , the kind of like time scales that we're talking about here , right , if you have flame slowly creeping down and anyone who does , he works in a fire lab downward flame spread over pmma is usually on the order of

about three mils per minute , right , so not racing down the solid by any means . But then concurrent flame spread , like upward flame spread , for example , is going to be , can be orders of magnitude faster , right , depending on the length scales involved and so on , because it's an unsteady process .

And so in the context of fire engineering , upwards sort of concurrent flame spread and these sort of rapid flame spread rates immediately pose a significant issue in the context of fire safety , not to say , you know , also , opposed flow is another regime of flames where we really need to characterize .

But it's interesting to note right away that there is , you know , in I'm sure we really need to characterize , but it's interesting to note right away that there is , you know , in terms of the implications that has to fire , fire engineering , you know , start to see that right away .

And this is also if you start to take a step back and think about these two conditions , like the matches again hold them in opposite directions . It's clear that the conditions between these two scenarios are different .

Right In one you have hot gases moving in the direction of flame spread , and then the other , the flame literally has to move in the opposition of the direction in which all the products of combustion in the flame is is moved , and so for the . For the topic of my phd , we actually chose to study primarily the opposed flow problem and that was largely to .

Basically , we want to study a steady state process , this sort of canonical sort of condition , and resolve as much of the physics as we possibly could with high resolution measurements , right , you've mentioned .

Because I feel there's a secret to PMMA , actually in understanding these mechanisms .

Well , there's the actual practical implications of why PMMA is useful and then there's sort of the history of why PMMA is used . So let's start with the history , because that's just a fun anecdote , and I think Rory might have even touched on this when he was talking about ignition .

But if you looked at back in the day , in like the late 1970s not too late 1960s , there was research going on , particularly in different parts of the world .

There was some research going on to the east coast of the United States looking into flame spread over PMMA , because PMMA was at the time used as rocket fuel , so it was actually something that was being used as a propellant , and so there was a lot of study of PMMA under high pressure as a propellant , and so understanding the rate of flame propagation was a

surrogate , for how are you actually getting energy out of your propellant , basically ? And so because there was existing data when the fire scientists came around to study the process of flame spread .

Theoretically this is my my understanding of this is there was already existing data and they're like oh hey , let's just crack on with pmma and the nice thing about pmma now for everyone who continues to use it today is that all of , especially with ignition and flame , spread the sort of theoretical thermal models that we use .

These processes require assumptions like you , you have to be a vaporizing solid , right ? Basically you have to have a solid that once it heats to a sort of magic pyrolysis temperature , it just instantly vaporizes . Charring solids , melting polymer stuff like that cause some complexities .

So if you get pma and you get a high quality cast PMA this is sort of a nice little anecdote to add there , because if you , depending on how you manufacture the PMA and depending on the cross-linking of the polymer , you can actually , if you extrude the PMA , for example , and there are other kinds of processing techniques that actually will result in the PMA

dripping quite a bit , but if you get nice high quality cast PMA , then it acts pretty like a nice vaporizing solid . You don't see much melt flow , you definitely don't see any charring or anything like that , because it's polymer , but you end up with a solid that actually behaves pretty well .

Basically it's like 100% efficiency pyrolysis , which leaves not much behind . How does that translate to other substances that we would meet in the real world ? Timber is something that's perhaps most interesting for most fire engineers nowadays . So is studying pmma also relatable to to timber structures .

So there's going to be someone who's going to be listening and be like , oh it does , absolutely true . And there are people who understand the degradation kinetics of PMMA . But on a macro level , we can actually observe these phenomena . But the fault that you're getting at is then let's compare that to a slab of timber or a slab of polyethylene .

So you're looking at polyethylene , right . So you're looking at poly material like polyethylene , which is , you know , burn it . It's a dripping mess , and I'm sure you've seen this with you know many cladding tests that you've done , right . So then , all of a sudden , you start talking about the process of dripping and melting

. Right now , all of a sudden , the entire concept of the heat transfer mechanisms that allow for flame spread to occur are now occurring on a material that's flowing . How does that complicate the problem for charring solid , like we talked about ? This is a feedback loop , right ?

So so you know , in terms of the gas phase , flame is then feeding , you know , you transfer to the solid . The solid is then pyrolyzing as you char , you start to sort of . The char layer effectively regulates the amount of heat transfer that reaches the pyrolysis region , right , um , and then that's going to change your gas phase problem , right ?

It's going to start changing the amount of pyrolysis that gets into the gas phase . It's going to change the amount of heat transfer that can be provided . Most labs of timber can actually support flame spread without an external heat flux because the charring behavior .

So that's why you have to study charring salts in something like the lift or you know sort of like those sort of standardized tests with with radiant panels or radiant heating conditions perhaps we also need to define pyrolysis in this series .

So what we need to do is we need to first take this solid and break it down into a gaseous fuel , and so , again , that's a very simplified version of the process for people who study pyrolysis kinetics .

So , forgive me , but the is that you basically you take heat , transfer into the fuel and you create a flammable gas that then burns the flame sheet it's difficult discussion because it's at the same time hard fire physics and .

So , once again , the loop of having a flame , having a heat transfer from that flame , the fuel to undergo some physical change , be it pyrolysis , be it melting , igniting the products of that physical transition process , and having another flame that can transition elsewhere , that's , that's the beauty of flame transfer .

Now you've said that in case of charring or timber , that's actually the case . That's very often brought up by layman . Actually , you know , timber doesn't burn that much each chars , but the processes in there are much more complex . Can phenomenal natural charring have been like an insulated layer ? Can we repeat those artificially ?

On materials , how do we make materials less prone to fire spread , knowing about those things ?

I think the next step we should actually talk about is we've been throwing around terms like heat transfer and whatnot . But let's actually dig into that a little bit more and let's see how those start to change as you start to change things like charring terms or charring solids or whatever .

But I do think , basically , you need to define this framework for what we call the controlling mechanisms of flame spread right , and what we need to do is we need to very systematically study those under different conditions , right ?

So , um , there has been work on charring solids and flame spread right , and it was done by , say , like the likes of atreia and and colleagues . But there's still , obviously there's still many things . I'm sure we can more information to sort of dig into and more studies to more studies to do there .

But it's basically , if we can tie everything back to this , this framework of what actually dictates this process of flame spread . I think that's sort of the the direction , so I've been thrown around this term controlling mechanisms of flame spread this is a term that was coined .

Amazing paper that was , uh , written by friend , is carlos , france , pale and harana , and basically they outline this framework , or what are the things that actually control the flange red process , right . So I'd recommend reading this paper . It's pretty awesome , but a classic . Largely .

The controlling mechanisms break into sort of like two big categories , being heat transfer , right . So something we've already talked about and we can break that down even further from there , right , but heat transfer mechanisms basically control how is the material ahead of the flame front being heated , right .

How is that heated , being pyrolized and then the flame sort of continues along its way , right , how does that process from a thermal perspective ?

Right now , the other mechanism they talk about in the paper and we will just briefly talk about here , but for completion I think it's worth talking about is other gas phase phenomena associated with the flame itself , right , and this includes things like chemical reaction , kinetics or mixing ahead of the flame front , and so the cases that we're normally interested in

as engineers . We generally focus on what we'll call the thermal regime , right , everything is just heat transfer to the solid at the flame front . Right , the process of flame spread can be described through heat transfer . Right , and that's where largely we'll be focusing the discussion here .

Right , but it's worth noting that there's another regime of flame spread , coined as the kinetic regime by Fernandes-Beyer , that all of a sudden the mechanisms that actually control the process start to change a little bit . Right .

But in the thermal problem , right , this is where we see things like the models presented by Driss and Quateria , you know , it's treated as basically a balance of terms , an energy balance .

You can basically create this model where you say I'm going to take this bit of solid , I know that I have a flame spreading along the surface , I'm going to take this bit of solid , I'm going to raise this bit of solid to a pyrolysis temperature right , and that's the feedback loop we've been talking about .

All of a sudden , now it's pyrolyzing , it's producing pyrolysis gases . From that ratio of how much heat transfer am I providing ahead of the flame front as a ratio to how much energy does it take to actually pyrolyze this solid ?

What balances those two terms is basically your flame spread rate , and so your flame spread rate adjusts to the balance of those two turns , and so that's basically how most thermal models work , is they're just looking at the solid itself and saying if I can approximate the heat transfer at the flame front , I can appreciate how much energy it would take to pyrolyze

the fuel . The flame spread rate balance those two terms .

Oh my God , there's too many variables . I cannot like find a solution . There always will be a solution and it's simply an outcome . And for us the balance is what you get out of the equation . Yet nature will find a way , that's true .

Yeah , the place that I wanted to segue you into was more like practical , or maybe not practical , more more like real world consequences of the very basic phenomena that we're talking about in here , like why fire retarded materials would spread the flame slower than non fire retarded .

Why found material would spread fire faster than the same material , but in a slab form , right . Why porous or or a dust will behave different than than the slab of the same material . Because it's all . You could probably track most of those questions back to the fundamental flame spread equations . Let's try the insulative material , formed materials , because it's .

It's something that marked the change of times in fire safety and fire science when we moved from the classical furniture in your apartments into the modern furniture . Everyone's seen the videos demonstrating a fire in a 1970s house and the modern house full of foamed material and the consequence is striking in the time to flash over and stuff like that .

But you can trace that back down to the flame spread and fundamental physics , right .

. I think everything that you mentioned in that list of potential interesting topics to talk about , I think it absolutely all ties back to everything that we've just said , this feedback particularly . You know the heat transfer mechanisms and you know and how that results in the establishing of pyrolysis and so on .

So I mean for foams , but I mean what's , what is the characteristic that makes foam so useful ?

It's difficult . The fire fundamental series is most difficult for me and my guests because we don't have a blackboard where you can draw the model .

But you need to understand that when you have a blackboard where you can draw the model , but you need to understand that when you have a flame spread and you have heat transfer to your solid fuel , the fuel is not an infinitely thin piece of material . Technically it could be , but that's a different case , a different regime .

Let's say you have a heat transfer from your surface downwards to the fuel . You're gonna have fuels that's transferring that heat very well into its depths . Like if you had a slab of copper , it would be beautiful uniform temperature over a large bulk of the material .

If you have a foam material , you suddenly have extremely hot surface and not that much heat transfer into it .

But if you have something that is like a foam right , all of a sudden that term in your energy balance changes dramatically , and so then you have energy available to go to pyrolysis .

That term you're making fun of calling it a balance , but you go back to the balance there , yeah , and if you remove that loss term , then all of a sudden what happens to your flame spread rate ?

You have more heat being provided for pyrolysis , your flame spread rate adjusts , and so I think actually , yes , all of these practical examples can be linked back to this sort of fundamental idea of looking at you know what are the processes that control okay , now it's kind of funny because it's it's .

It's like it's like almost always the same . Go read forman williams or something . But anyway , in here actually presenting this in a very naive or or physically incorrect but lay understandable way , that this , this leads to more temperature on the surface and this leads to less temperature in in the surface .

You know , having foam material with more temperature and bulk material with less temperature is actually a way to describe that difference , right ?

So if you link it back to the idea that it really was being regulated or controlled when you have something like different solids is how much heat transfer is occurring , how much heat is staying occurring , how much heat is staying in the solid , how much heat is being conducted further , how much heat is being sent ahead of the flame front through either the gas

phase or the solid phase ? It all starts to change the process , right , and I should say too , when we're talking about it , doesn't matter whether we're talking about concurrent flame spread or opposable flame spread , right , if we're talking about heat transfer , right , right , there's two major pathways for this process to occur .

Right , okay , you have gas phase heat transfer and you have solid phase heat transfer . Imagine , let's go back to our matchstick example , right , and let's focus on you . Just the opposed flow start , right , but you have this flame sort of slowly creeping down this matchstick .

Right now , the flame itself is going to be transferring heat both through the gas phase and through the matchstick itself . So there's going to be you know well , the matchsticks probably may be a bad example .

Uh , it's gonna be thin , solid , but if you look at more like practical solid fuels right , like large slats of pma , right , there's going to be a thermal gradient through the solid right and then that thermal gradient is going to result in conduction through the solid ahead of the plant and what's dominant in in that case , when the matchstick is pointing upwards ?

So because of that , the gas phase processes start to dominate the problem . And so in the limiting cases , right under thin fuels , gas phase processes dominate . Now if we get to thicker fuels , practical fuels , practical quotes , you know what I mean or the typical fuels we're used to , then it's not so trivial . It actually gets really interesting , right .

So it depends on a lot of things . It depends are we looking at opposed flow versus concurrent flow ? Are we looking at flows under forced flow conditions ? I mean flame spread under forced flow conditions ? Are we looking at flame spread under external heating conditions ?

All of these variables , right , affect that feedback loop , and so then the balance of how much the solid versus the gas phase plays a role starts to change , right , and that's actually the balance of how much gas phase heat occurs and solid ac transfer occurs has been kind of the entire focus of my phd , okay , and so if we look at let's say , focus a little

bit on .

Now if you think about this in sort of in this context right Again , when we said all of the hot gases in the flame were sort of moving in the direction of the flame spread and so if you think about heat transfer processes that occur , this process is largely dominated by heat transfer in the gas phase , particularly radiation from this flame , right .

So if you can imagine the flame sheet over the surface of this fuel , that's pyrolyzing radiation from the flame plays a large role in this and you can see this in the works like there's a great PhD by Isaac Leventon who looked into upward flame spread over various pollen fuels and this measure of radiation at the surface . That's really interesting work .

You can also sort of see this Now the question of whether or not gas phase or solid phase is dominant in this process for concurrent flame spread can also be demonstrated by saying what happens if we replace a slab of pma with discrete little bits of pma kind of stacked on top of each other with a little air gap .

So you imagine like instead of taking a continuous piece of pma , you add little air gaps and you say basically now I have no mechanism for solid phase conduction through the solid , so let's just look at , you know , gas phase heat transfer and you actually get similar orders of magnitude of flame spread , meaning that in this context heat transfer to solid , for you

know , these sort of homogenous , continuous solids , might not be as important . And so the gas phase process sort of seems to run the show here and there's lots of studies that have sort of looked into this sort of idea . But this also should be heavily caveated with .

All of these are looking at , you know , very nice slabs of non-melting PMMA , non-charring materials . But the idea is , when your plant spread is moving in the direction of induced flow , right , gas phase process , the process at least of gas phase heat transfer , begins to play a significant role in how are we getting heat transfer ahead of the solid .

But that story sort of starts to change when we look at opposed flow of flames , because this isn't as simple right now . If we're sort of looking at , let's say , downward flame spread , we know for upward flame spread you have this really nice orange , sooting , luminous flame , right .

But for , for anyone who's ever seen like a nice downward flame spread experiment , right again , if you just light a match and watch the leading edge for something like a downward flame spread experiment is a sort of very nice blue flame , right . It's just something that is very non-soothing . It's , uh , right at the leading edge .

It's very sort of gorgeous blue right . It's very different if you think about the flame in terms of its interaction with the solid right at that leading edge . That's very different than a big sort of turbulent sooting flame . So for down flame spread , we have to think about how is that flame then transferring heat ahead of the flame front ? What does that ?

Right , because if the flow is going the other way , we can't treat it like convection in the traditional sense that we normally do .

Heat transfer coefficient if it's a nice , not sitting flame , radiation still exists , but it's probably negligible compared to other heat transfer coefficient if it's a nice , not sitting flame , radiation still exists , but it's probably negligible compared to other heat transfer mechanisms .

And so you know , the intuition for the opposed flow problem is very different than looking at upward flame spread , you're left with conduction , but in the gas phase right correct , exactly . So , in terms of the gas phase , this is another cool thing you can do at home light a match and sort of stare really closely at it .

You'll see that at the leading edge of the flame you have an actual gap between the fuel surface and the flame itself , right , and so the process that controls the gas phase , heat transfer , uh , for something like downward flame spread is the conduction across that standoff distance , right diffusion , basically from the flame to the salt , which is something that it's

kind of a sort of a very abstract concept it's super counterintuitive , because you would not think about conduction in gas phase that much .

But I think , I mean , I love that example of just like , let's take a step back and look at concurrent flame spread upward , flame spread over slough of pma , right , we know that , you know heat transfer is the star of the show . How is heat transfer getting ahead of the flame front ?

Right , you see , it's driven by gas phase processes , by , like radiation and convection and those kinds of heat transfer , primarily radiation . But then you look at the downward problem , right , and that main mechanism radiation goes out the window . So now , how the heck is this process happening ?

And so in the gas phase , it's dominated by this process of gas phase conduction , right ?

But equally , if you actually look at the solid phase and this is something that you know , we spent a lot of time measuring if you actually resolve the temperature profiles in the solid , you'll also see that there is a substantial amount of heat transfer going through the solid itself .

Now , these thermal gradients don't exist for most upward cases , because the gas basis is basically dominating the process . You don't see as much conduction through the solid . However , for downward flame spread , you know , and we published this in a paper in Fire Safety Journal .

However , for downward flame spread , you know , and we published this in a paper in Fire Safety Journal what's really interesting is if you actually start to look at the balance between how much heat transfer is being received at the surface from the gas phase versus how much heat transfer is being conducted through the solid , the order of magnitude they start to line

up , not even order of magnitude in terms of total value . They start to more or less balance each other out for downward flame spread in particular . And so , in terms of wrapping your mind around , where is this heat coming from ?

There is heat transfer going through the solvent itself , ahead of the flame front , but also transferring through the gas phase as well .

But if you zoom into the leading edge right , your flame is moving in one direction and then the entrained air that's sort of reaching that leading edge , has to be moving in the opposite direction , right . So if you think about it from that reference frame , there's actually a lot of different kinds of opposed flow flame spread right .

Horizontal spread over , say , like a surface , like a spread over a table right , that's also opposed flow . Lateral spread over a wall right , that's also a post-flow lateral spread over a wall .

So if you ignite a wall and it spreads laterally , so sideways , as studied by you know conteria quite a bit , the development of you know test standards like the lift , um , that's also a post-flow flame spread . Now what's really interesting is is those are , you know , in the category of post-flow flame spread .

But that's actually something that I've studied quite a bit over my phd . The heat transfer mechanisms start to again change though , between , even though these are all post-flow flame spread .

I mean , just think about that in your mind's eye , right , how the leading edge of a downward flame spread , uh sort of problem is this nice blue leading edge , but imagine , you know , a horizontal flame , right ? This is basically , like you know , pyramid of it's no longer the same sort of boundary condition , right ?

All of a sudden , things like radiation start to come back into the picture . For the lateral condition , right , you have a very luminous boy wall flame , and so you start to get a little bit of radiation at the leading edge as well , right , something that you didn't see for the downward case .

So so you can still use these same sort of ideas of like , yes , there's , there's some solid-phase heat transfer to the solid , there's gas-phase heat transfer , but the mechanisms of gas-phase heat transfer start to change a little bit , right , and the relative importance of , say , radiation versus gas-phase conduction really start to play a role , and so , by sort of

changing these orientations , you can start to sort of play around with these different variables and say , okay , well , how much do you think this radiation is playing a role here versus that or flange spread ?

I think the reason why engineers need to realize or know or understand this is because you've previously said that the the flame spread on downwards on pmma could be three mils per minute . But if you rotate it , like if you change the orientation , it could be 10x that or perhaps even more if it's vertical .

Right , if it's exposed to external uh flow of air , that accelerates it even further , perhaps even faster . Now the thing is , if you have a material , the material properties are not the sole thing driving the flame spread or the fire hazard .

If you have a carpet , you can test it with your standard test methods for carpets , which includes basically having a radiant panel over the piece of the carpet and then just seeing how far it will ignite in what time , and the carpet may be perfectly safe for your flooring , but if you choose to put it as a decoration on your wall , you've suddenly changed the

regime from opposed to concurrent flame spread and suddenly in your carpet may be something that kills people , and I see such dangerous cases that are created by people who don't have understanding of this fundamental physics , that this simple law of nature , it matters how you place it . Like you know , the orientation matters a lot .

If you somehow forget about it or choose to ignore it , you may have a very , very dangerous outcome and I believe , like understanding this physics , I mean , I agree it's very difficult physics . If you want to go into models , it's one of the most challenging physics For the reasons that we've said .

It combines so much of fire science , heat transfer , gas phase pyrolysis , all the feedback loops you probably can get into mixing kinetics , you know . You know turbulence , boundary layers , emissivity of all the surface .

You change the orientation or you force a flow over , right , the flame spread properties are gonna , the rate of flame spread is going to change you dramatically , and so that's where I think actually the link that needs to be made is is through things like like understanding how do we model these things , how do we model these processes , and that's what the entire

sort of with an IFSS , it's what you know , macfp , that working group , which I know you've listened to the podcast before . That's something that you know . The working group is looking into and trying to understand how do you actually model the Flansburg process and how do you start to link these different things from the solid phase to the gas phase .

And it's a very difficult problem , right , because some of the things that we've discussed already is , uh , the length scales needed to resolve some of this . Physics are tiny , right , you know . So you need to be able to capture the standoff distance and be able to resolve .

You transfer across that standoff distance of the flame on the order of you know , a mil or less . When you're trying to run that on a cfd model , that's probably for practical purposes . You're not making your grid cell to get five grids across your standoff distance in order to run something from a mathematical engineering scenario .

So how much of the physics are we missing out there ? Are we actually able to reliably predict these heat transfer processes with the limitations that we have ?

. So , because it's a complicated matter , let's try to give one more . Try to overlook the topic , like you know , to give to distill the knowledge that our fellow fire safety engineers could really benefit from , would you say . The best thing to look at is a model how we should do that . Actually , that's a challenge . It's your job .

And there's , you know , there's certain concepts that you know , it's apparent to someone like you know , us , who who's sort of immersed in the world of fire science very regularly , but , like , what it comes down to is the process of flame spread , like we understand the implications of flame spread , right , it's linked to fire growth , fire growth linked to fire

hazard , and this idea of flame spread being this propagation of a flame along the surface , series of ignitions , so to speak , right . So when we've been talking about these things of thermal models , right , or this idea of , like , how do you track the heat transfer ?

When we're throwing around these words models , I think it's also worth clarifying that a lot of these are , you would see them in textbooks as equations , right , and so , in a sense , they are analytical models that predict , you know , flame spread .

But I think some of them , you know , you can open up sort of any fire dynamics textbook and you'll see flame spread is , you know , rate is equal to something that's like a ratio of two things right , whether it's , you know , a heat flux provided as a ratio to usually like a thermal inertia term times , you know what's the temperature rise required to pyrolyze .

You know the solid right , so it's like a delta T change in temp , and so that's kind of what we're talking about with these models , right Is ? That is the balance that I keep referring to , right so now ?

So , now that we've gone through this idea that , okay , well , in the context of , say , concurrent flame spread , that heat transfer process is largely dominated by gas phase heat transfer , right for something like dower flame spread , there's a balance . There is a degree to which solid phase heat transfer is occurring ahead of the flame front .

You are getting energy to the pyrolysis region that is coming through the solid itself right , but it's also gas phase heat transfer , right , and and then , if you look at within the category of the post-flow flame spread , that gas phase heat transfer might change between different configurations because of the way the flame is structured , right , and so all this is ,

you know , sort of intuitive on one level when you see it , but also kind of non-intuitive , right , because you'd think that these sort of categories fall into place really nicely , but there are things that change across these categories .

And also , when we're talking about uh defining , like , controlling mechanisms of flame spread , right , a lot of work , especially sort of , you know , over the last 50 years or so , has been really targeted towards defining what we'd call a dominant mechanism , right , because it's because , in order to sort of , if you rewind back , developing some of these models and ,

like you , had to sort of uh identify which are , which are the ones that we can focus on . What are the , what are the heat transfer mechanisms , what are the you know , the physical mechanisms that we can focus on .

Say , this is the dominant mechanism in this case , right , and so the result of that was we had lots of sort of limiting conditions , right , we had sort of flame spread equations that you'll see for thin solids and you'll see flame spread equations for semi-infinite solids , and very little , very rarely do you see things in between .

But but I mean in terms of , you know , looking at the extremes . Those are generally how we classify it right . Or we have to just look at concurrent flame spread or just like a purely idealized post flame spread , but very rarely do you will .

Are you able to take that equation and then say , well , what if it's lateral spread versus horizontal spread versus , you know , and so on . And so the idea that we have to choose these dominant processes has resulted in the formulation of these you know , these analytical models , these equations .

But I think what's really interesting is some of the research that we've been sort of showing in my PhD and other . You know research Like there was a great paper by Ito and Kashiwagi back in the 80s that used this really slick holographic interferometry technique to look at , you know , the temperature gradients through solids .

Like holographic interferometry technique to look at , you know , the temperature gradients through solids . But with work like that is also shown in combination with stuff that we're doing , is that a lot of cases you can't always just say one is completely dominant over the other to the , to the point where you ignore something like the solid .

So for things like a postal line spread right To ignore the solid phase entirely , and it misses out on a big part of the problem . Right , because there is heat transfer going through the solid . So , from a practical perspective , if we're interested in sort of modeling this process , then how do we explicitly account for both ?

Right , instead of the beauty of saying that there's a dominant mechanism is you could say that well then I can just ignore everything else . Right , and so then you know , this introduces complexities . This introduces the fact that we now have to start thinking about can we resolve , can we couple these problems ?

Right , because that's the whole thing that we've been going in circles about is that this is the complex part is how do you decouple that solid phase from gas phase interaction ? And now that we have to consider both the solid and gas phase , this really increases the complexity of actually trying to predict this problem .

You know there would be this transition regimes In practical engineering . I think focusing on one and just being really good at it , I think that's much more like . It's so much easier and so and probably goes a further way and you could focus on material properties and just just realize that orientation changes everything and your thinking is no longer applicable .

Or you could focus on the way how objects are placed against each other . I mean , if you're designing a ventilated facade , it's kind of always going to be upwards with the cavity right .

It's not that orientation and this part of physics will change that much and in that way you can focus on the physics which seems to be dominant in this case the orientation and cavity being dominant and then just play with the materials a bit .

It's not that you're going to solve a cavity fires by playing with materials , unless you make them non-combustible at all , and it's not like orientation doesn't matter when you're working with materials , but I think fire safety engineers should focus on one and recognize how the other impacts , rather than trying to find themselves in the middle of this chaotic balance ,

which probably would be too hard , I think , in a realistic application

.

Right , there are so many rabbit holes and so many existential rabbit holes , you know , to get to the bottom of questioning , you know life in general , right , trying to understand more of the physical process that controls certain process , certain flange of process .

Again , looking at things like we kind of hinted at earlier , towards , like microgravity , fires and the kinetic regime under really high force flows , some of those things , when you're really grappling with these ideas .

And so you're right , in many engineering scenarios that might not be the problems that we're interested in , but I think it's absolutely critical that , no matter what engineering scenario we're looking at , we do need to tie back our understanding back to the relevant controlling mechanisms , right , so you're . So you kind of hinted at that , right .

So , like , if you can say in certain scenarios that you know , yes , this one is dominant , sure , but you need to know exactly at what point , what is the magnitude of the other heat transfer mechanisms , right , if we say that X mechanism is dominant quote , unquote in this case , at what point is that not true ?

And I think it's actually not as well defined as we would like to think . Right , there are still many scenarios in which you know this balance of you know this gas phase versus solid phase , heat transfer . It's you know it's a battle of time scales .

Sometimes there's a point where , like , actually , if the flame slows down enough , solid phase starts to catch up right , and there's our there's really interesting things that we're starting to sort of scratch the surface with with some of this research that we're doing .

So , yes , but I don't think it's as simple as saying that in all cases , we can just say this is the mechanism that we should focus on , right , I think .

I think , if anything , the point is , hopefully , that even if you're saying , yes , in the case of an upward problem , I'm going to focus primarily on the gas phase , but in the back of our minds , let's think about what's going on in the solid phase , and is that at any point do these processes start actually having an effect ?

Because that's going to change our ability to predict the flame spread rate , absolutely .

Now , when I think about it , the place where I see this be extremely useful , this kind of thought process that you've just demonstrated , is when we're trying to implement new technology , new material , into already existing setting that we know very well . This is brilliant because if you start thinking about that , how does the material pyrolyze ?

What are the kinetics of that Like ? What are the critical temperatures at which stuff will happen ? How quickly will it happen ? How will it react to radiant , convective , conductive heat transfer ? Right , will it conduct heat ? That's critical to some extent .

And even though a material is unknown in a full scale , if you understand the basic properties of the material and let's be honest with ourselves , we don't have a good material database on large scale fire behavior for most of the materials , but there has been someone who's done cone on it somewhere in the world , right and it's it's much more realistic that you

will gain information about those properties and and perhaps could try brainstorm the big scale behavior based on those fundamental concepts .

Right , because one is really simple example in terms of new technologies is you know , there's been research coming out looking at flame spread over CFRPs and you know , reinforced polymers . Looking at , you know whether it's carbon fiber , glass fiber reinforced polymers .

Looking at you know whether it's carbon fiber or glass fiber reinforced polymers , particularly with some . There's a really interesting study that was presented at the combustion symposium I believe it was last cycle , maybe the cycle before .

But looking at the angle of the fibers , right , the orientation of the actual , you know , layers of these fibers , the , the angle between them , will influence the rate of a post-flow flame spread .

Meaning that if you , the directionality of your , of your conductivity along those , those carbon fibers , is going to influence the transfer through your solid , thereby , as an extension , it's showing that your magnitudes , your balance of solid phase versus gas phase heat transfer , are not set in stone . Right , very with very simple .

You know that that's a cfrp like something we use in , you know , in construction and aviation all the time . That's a very simple example what happens if we come out with new materials we don't even , we haven't even thought of yet , right ?

So if we fall into this trap of saying , well , always when we're looking at upward flame spread , these are the mechanisms that control it , always when we're looking at downward flame , these are the mechanisms that control it , we always need to revisit .

You know at what point do those break down and what are the conditions in which we start changing those things .

And I think that's really why I'm very passionate about sort of looking into the balance between these two and that's why we spend so much time looking at these heat transfer mechanisms is who knows what the future holds in terms of how these balances will be affected by different , you know , future materials or future assemblies , or who ?

So let's appreciate the listeners who held with us up to the end of the episode . You guys are stars and we appreciate your passion for fire science . This was a really tough episode , but yeah , that's how the fire science looks like . It's not been intended to be an easy science , right ? Well , thank you for having me . I really appreciate the chance .

Thanks , david , see you again . Cheers , and that's it . Thank you for listening . If you've reached this part of episode , I thank you very much and I applaud you . You're the best , you're the fire safety engineer that the world needs , and I hope this episode was full of interesting stuff for you .

I hope it's a beginning of a lot of reading and research , or perhaps just a nice way , nice introduction to this very complicated subject of flame spread . It's something that influences us on so many levels . We need to appreciate that . I think if we all were really good physicists , knowing everything about flame spread , the Grenfell would not happen .

The King's Cross fire would not happen . A lot of disasters could be avoided if we understood mechanisms that David has shown us in this episode . So , yeah , there's a lot of merit to study flame spread . There's a lot of merit to try and understand it and build new experiments , build new research and grow on top of that .

Thank you , david , for doing that for us . Um , a quick word of apology as well . Uh , the the quality of david's recording was a little sub bar . He had to record from a very small room and when I was carrying the I did not capture the echo that his microphone made . It made my editing a little bit hard .

I tried to get as much as possible out of that , but I understand that this is not the usual quality of the audio in the FireSciencer episode . I can just apologize to you . I hope it did not invalidate the experience for you and the knowledge that David has brought to us . But anyway , that's it for today's Fire Fundamentals episode .

Have a great week , have a great weekend and next week we're back in the 5-Science Show with another hopefully good episode , perhaps not as tough . I'll give you some rest , but it's still gonna be fun . Thanks for being here with me . Cheers , bye .