141 - Smouldering in Mass Timber with Harry Mitchell

Hello everybody, welcome to the Fire Science Show. It has been a while since we did the last episode on Mastimber, so it's time to come back to Mastimber Fires and let's put a different spin on it this time, because in this episode we're gonna talk about smoldering fires in Mastimber compartments. Smoldering for timber is quite an obvious phenomenon, but in the world of Mastimber buildings it's gaining more and more attention as more and more research is carried on the topic and my today's guest pretty much brought it into the mainstream science. My guest is Harry Mitchell and because I think as we publish the episode it's his Viva week, it's Mr Harry Mitchell when the episode airs and hopefully it's Dr Harry Mitchell by the next week after the episode. Good luck, harry, crossing my fingers for you. Anyway, harry has done some very interesting experiments with Arup. They were called Code Red and you have heard Panas Kotsuvinos in the podcast talk about them in length. Harry's role was to observe the smoldering happening in those experiments and draw some conclusions, and that's what we are talking about today. I brought description of smoldering phenomena in Mastimber. What has been observed in the Code Red experiments and how does that translate to real life, full scale, everyday engineering? And yeah, it's very interesting. I mean, it's obvious that smoldering happens to Mastimber. If you have a log in your fireplace and it just ended flaming and you go to sleep, you wake up in the morning it's gone. That's smoldering. It happens in Mastimber as well, but what does it mean to the structure? What does it mean to the people? What does it mean to the firefighters and fire investigators? That's what we're gonna cover in this podcast episode. So please join me in welcoming Harry and let's spin the intro and jump into the episode. Welcome to the Firesize show. My name is Vojtche Vinczinski 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, welcome to the Fire Science Show. I'm here today with Harry Mitchell from Imperial College London. Hello, harry, hi.

Vojtche, thanks for having me.

Thank you for coming again. Thank you for bringing the issues of smoldering fires to the mainstream science on timber buildings. Thank you, harry. You really needed another complicated phenomenon to account for. Good job, man. It's something very interesting and something most of us doing full scale experiments on timber buildings have seen at least once in their lives. It's kind of obvious that if you even take down a large fire, there are these places of the building where the fire is concealed and can go for a long time. And I don't know if I can generalize, but I think all of us have seen fires burning through the walls for the next multiple hours, eventually making their way through the wall. It's great that someone has studied this scientifically. Before we go into the consequences of smoldering fires and smoldering as a process for mass timber buildings, let's try and get the vocabulary right. So maybe let's try to define what smoldering fires are, how they behave.

So the majority of humanity has seen smoldering at one point or another. I mean, all you need to do is light a campfire in your back garden and, yes, it will claim for a long period of time, lovely for marshmallows, but then at some point the flames die down and you end up with this glowing of the char that covers the surface area. What's left to do? Your wood? And that's what smoldering is. So smoldering is essentially when the char that is left over from your fire reacts with oxygen in the air, produces heat, energy and other gases and ash. So that in itself is a singular process. So that's what is also typically known as char oxidation. But that doesn't happen in isolation. So in order for smoldering to occur, there needs to be char present for that to happen. So that's the charring or pyrolysis process, where essentially you have your virgin fuel, so that could be timber. In the other research by our research group we look at peatland, which can also smolder. So any cellulosic permeable fuel can potentially smolder.

But the presence of char is not the only prerequisite, because I often had like entire slabs charred, like the char layer forms in the presence of fire and that's the normal mechanisms of timber fires. What are the other prerequisites besides the char being present?

Well. So smoldering tends to operate quite well under conditions where heat energy can be conserved, so areas where heat losses are minimized. So, for example, in areas where you've got insulating materials, such as like, for example, if you have mineral wool around the timber slab, that can reduce heat losses. Or, for example, if you have two slabs coming up against each other, if you have both of those surfaces smoldering, those can not only reduce heat losses but allow re-radiation between your smoldering surfaces, so that kind of thing can allow smoldering to sustain itself, whereas in a lot of cases where you, for example, just have a plain slab of char timber that is continuing to smolder, that might reach a critical point where heat losses are too excessive, to the point that the smoldering just dies out, which is what that's ideally what happens. But you do see, in some more challenging geometries that doesn't occur.

And is being exposed to flaming combustion directly like a prerequisite, like I imagine a scenario where I have, let's say, some sort of protection board between the fire and the timber, can the smoldering trigger behind that board, even if the timber has not went through flaming combustion on its own?

I would say so. So if there aren't sufficient conditions for the volatile gases produced by charring to a ignite, then no flake combustion can occur, but then smoldering can potentially still occur in those elements, just at a much, potentially a much slower rate. So the rate of reaction needs typically much slower in smoldering than compared to flaming, but in some cases that doesn't necessarily mean that the effects are less severe.

Okay, cool. So that's on smoldering, and if you want to learn more about smoldering, I think episode two of the podcast has all about it with Gero, so you're very welcome to join that to catch up. Now let's put the timber in our large structure. So in your paper, which I will link in the show notes, you reviewed existing literature that touches smoldering. So there has been a lot of experiments in which smoldering was observed at some point, but not that many in which it would be explicitly studied. Please tell me how you came up with the idea to actually do that, to study it in particular, what was your assumptions and how did you want to measure it?

So I suppose all the experiment that I was involved in just for a bit of clarification, the experiments that I was involved with it's the Code Red Experiment Series. This was an experiment series that was led by the Arab fighting. It's the largest mass timber compartment fire experiment to date, but by no means is it the first mass timber compartment fire experiment that observed smoldering. So one of the first that's mentioned in the literature is from the Eponomic Experiment Series led by Felix Weissner where essentially they saw that they had a timber ceiling in their compartment and I think from memory it was about 36 hours after the experiment the ceiling was observed to continue smoldering and then eventually collapsed. So essentially the objective of our experiment one of the objectives of the Code Red experiments was to try and characterize how these areas of localized smoldering can initiate, develop and then potentially have structural implications hours and days after the fire. And we decided to study that first of all without any form of suppression and after the end of flames in the experiment, because this means that we could just observe how the smoldering develops unimpeded by any suppression means, and decided to do this using just robust and simple instrumentation, being visual cameras and infrared cameras, as everyone has access to either their eyes or a camera. It's quite easy to be able to use those to be able to look at various points in your mass timbre element, because try and visually observe if they're smouldering. In some cases that isn't sufficient and we end up using infrared cameras to try and find these areas of elevated temperature, which is indicative of where smouldering might be occurring.

I love having a library of podcast episodes now If you're interested in learning about the plans and ideas behind the experience before the Code Red, that's X1, x2, we call them obora experiments. There was a podcast episode 27 with Guillermo on Travelling Fires. You're very welcome to join that one to see the background of the background and the experiments that you were mentioning. Carrie, I had the privilege to interview Dr Panos Kacavinos from Arab back then episode 111, and in that episode we discussed all about the background of Code Red, how it came to life, what were the points, and indeed smouldering has seen one of the very interesting initial points to be studied within the Code Red. So now, if you can briefly introduce us, how did the layout look like? In what elements of the structure in the Code Red experiments you were looking for smouldering and in what elements did you find it actually?

The compartment itself was a 352 square meter compartment, so the entire ceiling of that compartment was mass timber, so I think it's about a 150mm thick CLT slab with five lamella. And also inside the compartment were two glulam columns stationed at, I think, the midpoint and then two-thirds of the legs down the compartment as well. So these were the three main mass timber components in our compartment experiments. So these were the areas where we wanted to study smouldering.

But these are not the only complexities in the geometry, because on the pictures I also saw large beams that were insulated throughout the room, so it was not just a fully flat ceiling. Can you give us a brief explanation on those complexities in geometry?

Absolutely. So obviously the ceiling itself needed to be supported, so there was an insulated concrete beam that spanned the entire length of the compartment, that went through the mid-bathway along the width of the compartment itself. So essentially the entire ceiling sat on that insulated beam, which was insulated in renewable. And then we also had two concrete columns as well in the compartment to study how thermal penetration could progress through the thickness of those concrete beams and just to see how they performed in the fire as well.

Very interesting. I'm asking about these complexities because, as I mentioned before, you would often have an entire slab exposed to fire, an entire timber slab exposed to fire. You would have the entire timber contribute to the fire char, but it's usually those little ends of the slab where those smouldering hot pockets would form. I think that hotspots, that's how you refer to them. So where did they appear in your experiments? Were the locations aligned with what you'd expect before the experiment?

Yes, so, as you've alluded to, the smouldering hotspots that we observe tended to occur in interfaces or edges in our mash timber elements. So the timber ceiling itself was comprised of, I think, 13 individual mash timber slabs that were just secured against each other, and we saw that the majority of the hotspots either happened along the timber slab to timber slab connection lines or around the perimeter of our timber slab where they were structurally supported by the compartment wall. Also, we observed in the glulam columns that smouldering occurred, I think in two or three instances, at the base of the column itself, so where the column rested non-structurally secured to the concrete floor.

And you did not have exposed timber wall sizes in your experiments. No, no, just In some of the experiments we have done previously we had exposed walls and we always observed this pile of char accumulating against those walls. So I guess this also creates quite good conditions for charring to occur in your structural element. I assume for you that must have been the case around the columns, where you must have a lot of char falloff. Let's focus on columns for now. How did columns behave? How did the hotspots form in them? What was the final outcome of those hotspots after your prolonged observation?

Yes. So first of all, the movable fuel load that we used inside the compartment was a woodcrip. So after the end of flaming the majority of this woodcrip had charred, for example, around the columns. We did see that there was quite a bit of accumulation of char from, potentially, the seeding and the column, but also char from the woodcrip. So as I mentioned earlier, with insulating materials, char itself has a very low thermal conductivity so potentially not only could those remaining chunks of char continue to smolder but they could actually end up insulating any smoldering back around the base of your timber column. So particularly in one case in our second experiment we saw that a timber column continued to smolder at its base for I think it was around 32 hours after the end of flaming to the point where the timber column started to teeter away from its ceiling connection. In the images you could actually see, I think over the period of about two minutes the column started to pivot to a point where it was no longer structurally secure and it just collapsed. To what's law? So just also just a couple of clarification points on that. The column itself wasn't structurally loaded and wasn't structurally secure to the ceiling or floor. But potentially, if it had been structurally loaded, it could have potentially failed earlier or the consequences may have been more severe.

Yeah, the consequence of timber elements charring is obviously that their cross-section is reduced structurally cross-section. So perhaps you could comment on the impact of the charring process on the load bearing capacity of columns, for example, is it only the char layer that, to best of my understanding, is not just the char layer that reduces the structural load bearing capacity of the columns.

Absolutely. There are two components in that. First of all, preheating of your timber material Before even charring occurs. If your timber reaches anivate increase temperatures, it will typically degrade in its structural capacity. I believe even at 60 to 100 degrees there is a severe reduction in the structural capacity of timber. Then also the drying process as well Approximately 100 degrees. Any latent moisture in your timber element will typically evaporate and then permeate out of your timber element through exposed surfaces. This can also have implications for the integrity of your timber element as well. For example, if there's a build-up of pressure from your moisture leaving, then that could potentially lead to cracking of your timber element. I think there are a lot of considerations even before your timber element starts charring.

If we talk about columns, for a second step away from this prolonged concealed effects of small ring in elements. If you had an exposed column, the char layer must have formed all around it. Even then, if your movable load has went away, there's a good chance you still have significant sources of heat and radiation in your compartment, whether being it the charring remains of whatever was your fuel load, be it charring surfaces of your walls or ceiling if they participated in the fire, or just simply the accumulation of heat in the compartment. As we know the decay phase. It's not that you turn off the fire and immediately there is no heat anymore in the compartment. In these conditions, the columns would be well exposed to air, I assume, because they would be free standing surrounded by air. There must be oxidation processes happening in those. Had you a chance to measure how much you've lost of that column after the movable fire has burned out? Perhaps?

Unfortunately, we weren't able to measure how much the column degraded after the fire went out, because after it collapsed onto the floor, it didn't stop smoldering. It continued to smolder for, I think, one or two days after that, to the point where nearly all of the mass of the column are being completely consumed.

Okay, I'm asking because, for design purpose, someone could try to estimate what's the reduction in their cross section due to char buildup and usually would assume some kind of charring front velocity that you calculate during the duration of the fire. I'm wondering where one should put a stop on their calculations, like is it the end of the fire? Is it the end of the decay phase? Whatever the decay phase is defined as? Any thoughts on that?

Yeah, I think it's what you said is tricky to define. I suppose during flaming is when charring is going to be occurring most rapidly and in the largest area of your mass timber elements. But then in these localised areas of smoldering, even though the charring doesn't occur at much slower rate, it doesn't necessarily stop. It does make it tricky in these awkward locations to define when is the end of the structural implications for our structure.

I assume that it's also not easy to define the sharp critical conditions at which it would stop, because it would depend on the spatial location. How is position in the corner? Is it in a concealed gap between, like you hear you say, in case of your column, something that was standing on the floor? So there must have been a gap between the column and the floor, which perhaps made quite a convenient space for a fire to conceal itself. How did you observe that? By the way, was it visual observations?

It was infrared, so we've seen that infrared imaging was significantly more effective in being able to detect smoldering. So in the majority of cases we did see that localised smoldering was visible, either by the emission of smoke or by just glowing of exposed smoldering. But in some cases we can see it by visual observation and needed infrared cameras, and the column was one of those cases.

Okay, so now let's move from the column to the slabs. Were observations in the slab and a different like where have you been finding the smoldering hotspots and how you would assess their progression in this slab setting?

Yeah. So primarily we saw that smoldering occurred at the interfaces between, either between two different slabs or between the slab and the concrete wall that the slab was sitting on. And that's for a couple of reasons. So, first of all, reduction of heat losses so if you've got an awkward corner where you're not losing as much heat by convection radiation, then that can potentially allow smoldering to perpetuate more. And also re-radiation. So in cases where you've got two slabs sitting up resting next to each other and both of them are smoldering at the interface, you can end up with re-radiations from one surface to the other, which allows smoldering to perpetuate and keep going. So we decided to, in two instances, measure the rate of aggression of smoldering. First of all, we didn't read imaging in one case to just measure now that hotspot was progressing along the width of the tinder slabs, and then in another case, we just measured, simply measured the size of the hole, Because in tinder we saw that smoldering mostly occurred at the shore surface, especially along the width of the timber. We just decided that the size of the hole is approximately analogous to the progression of smoldering. So we just measured the rate of progression of the hole size over time.

To measure the spread rate and how big. The holes ended up like a size of a fist, a size of a chair in a slab.

No, not the size of a slab, luckily enough. So I think one of the larger ones was about 700 millimeters in length by the end. So that's about. I'd say that's about foot size, which is a little bit worrying for a ceiling. I suppose, if you imagine you have a post-fire inspection in an actual building tire, if you have a hole like that that could be a potential hazard for anyone trying to inspect the structural damage to the compartment or the building. So yeah, I'd say not a massive size, but not insignificant either.

And that was after how many hours? One, two days.

So that was, I believe. Last measurement was at 42 hours, so just under two days in that case, before smoldering ended at that particular hot spot.

From this data can you establish some sort of a smoldering velocity, However you would define it, how quickly it is of the timber.

Yeah, absolutely so. In the two cases that I mentioned, I believe we measured smoldering, the progression of smoldering out 1.3 millimeters a minute in one case and 0.42 millimeters in another case, and I think those values are very different. So I think it goes to show that the progression of smoldering is really impacted by where the smoldering is actually developing and the conditions under which it is. It's not a clear cut problem where you've got one value and it's always going to progress. About value, it's quite a complex phenomenon to characterize.

So what you've done. You have focused on smoldering. Before your experiments, you've observed it during your experiments, you have characterized some of the properties of that smoldering. Now let's try to get something practical out of it, because I still don't understand fully to what extent this is something that's considerable in my design of timber buildings. First of all, how often did you find those concealed locations, those hotspots forming? I even find it difficult to find the correct means of quantifying how often they happen. You seem to quantify them per meter of your edges. That sounds like an interesting way to quantify it. So can you give me quantification of how often did they happen?

Yeah, absolutely so. Over the three experiments that we studied, we found that there were around 19 hotspots in total, so three in the first one, seven in the second one, nine in the last one. Since we observed that most, if not all, of these occurred at the edge of our timber slabs, it makes sense to characterize that frequency of hotspot occurrence in terms of her meter of slab edge, because then that can provide a rough order of magnitude if you wanted to extrapolate to a larger timber compartment. So if you've got not a 352 square meter compartment like code red, but you've got a 1000 square meter compartment, how many hotspots might you anticipate seeing? So we found that that was between 0.012 and 0.58. Hotspots per meter of timber edge. So basically, for every two meters of timber edge you could potentially have one hotspot.

How many of those have survived till the final observation in the 42nd hour of your observations? I assume they were going down eventually, not all of them at the same pace.

Yes, so the majority of them had gone out by 42 hours For a couple of reasons. Either there was a change in conditions, where the heat losses were too great and they went out by themselves, but also in two experiments we did observe that there was a brief period of rainfall, around 30 to 60 minutes a few days after the fire, and these we did see in some overnight infrared imaging. Those did contribute to the extinction of some of the hotspots.

Okay, let's now talk about some practical challenges in managing this behavior. So one thing you have a massive fire in your building. Let's say you have a flash-overed fire of 200 square meters of your space, or traveling fire in that space, whatever we choose to call it. Have you attempted any extinguishing actions? Or, from your experience, how challenging it is to take out those hotspots by the fire department that arrives to the scene, because I assume they don't want to leave the scene with an ongoing combustion which is bothering fires of the knees.

So first, of all, suppression of hotspots wasn't the focus of our research, but we were still interested in it. I know there's been a lot of work on a previous experiment series that rise by Daniel Brandon looking at suppression of smoldering behind encapsulation boards of, I think, tipped-ing mashed timber walls. So we decided, okay, we'll have one case study to see how easy is it to suppress a smoldering hotspot. So we found one of these smoldering hotspots at the wall interface of a timber-sealing slab and suppressed this with a fire hose and that's all we could manage to suppress it. It's, at the end of the day, they are suppressable. But we did see that it is slightly more challenging visually than putting out a flaming fire because it's harder to observe if it has actually gone out or not, because they are harder to detect without infrared imaging.

Have you been observing that particular location afterwards, like eight, 12 hours after did it reappear, or were you successful by just applying water from the outside?

Yeah, so we were successful in suppressing it in that initial suppression attempt, but it did take a couple of periods of pointing the hose at it for a few minutes, taking the hose away, looking at it with an infrared camera. Has that worked? It could use a bit more, so we're applying it again. So it does require a bit of tenderloving care, as it were, to make sure that it has actually gone out properly. And also, we did have to remove a bit of protective encapsulation board around the perimeter of the slab, and I can imagine this could be the case in a realistic scenario where you can have smoldering behind something and you need to remove it in order to actually access it to suppress it successfully.

From my own account, when we were doing experiments in large open land compartments not 352 meters, but I think ours was like 70 or 80 square meters, quite large actually, with CLT slabs and quite complicated beams and columns we, as most experiments on such settings have we had opening on one side and there were two beams. At the end of the experiment we started extinguishing it and of course we got down the flaming combustion in the compartment, as you could expect. But you could still see smoke coming up from behind those beams. You could see there is something happening behind them and I would say we did it on purpose. We sat there and we said, okay, we need to take down this fire at any cost. Let's simply see how difficult it is, and it really was difficult. We had to eventually cut out openings on the side walls of that building to give us an unobstructed access to this particular edge between the beam and the ceiling to apply water in that location, and only after that, 15, 20 minutes later, we did not see any more obvious combustion happening. At that point we didn't have a thermal camera at hand to monitor that. I can see where you were going with your explanation that it's much easier with the thermal camera and it would have been for us. The lesson learned was it was really challenging to reach that place and we are talking about the fire experiment that's happening in a controlled environment, a free-standing compartment where you can literally cut an opening to a wall because you can stand next to that wall. It's not 100 meters up outside of the building to cut that hole. And, on the other hand, if I have such a phenomenon ongoing, you have no idea what's the load bearing on the ceiling, what state the structure is, whether it collapse or not. It's not giving you a lot of hints before it collapses. So I would assume going inside that compartment and just spraying above my head, that would be a ridiculously dangerous thing to do. So my first-hand observation is that these things are really persistent and it's not the challenge with putting the medium. If you put water on it, it's going to do its job better or worse. But to be able to put that water in that particular location very precisely and controllably and safely, boy, that's a challenge.

I definitely think it really does bring how tough the job of firefighters is because, also, we're talking in these experiments here about relatively simple structures compared to what an actual building is. So in an actual building you have cavities, you have penetrations, you have all of these really complex design elements that potentially make a smoldering hotspot that much more inaccessible for a firefighter who's already had to go up 20 flights of stairs. So they are potentially quite challenging to suppress if you've got all these complex design components and this inaccessibility for your timber elements.

And also you have mentioned the smoldering ongoing behind and encapsulation. We also had some things like that happen, especially that when we were doing other timber experiments you would eventually like the ceilings are created from larger slabs that are interlocked with each other. There are these edges in the middle of a slab. If there's fire, obviously we access the fire by bending, moving. Some of the gaps open up a little bit, creating this areas that can be penetrated by the fires. We had instances in which behind encapsulation, like we have a fire experiment, we terminate the experiment, we would extinguish the building. All the obviously visible hotspots are removed because we put water over them. Then the next day you would find significant damage behind the encapsulation, literally the smoldering fire eating out the walls. My challenge in here I understand it can happen. I don't really understand the severity of it. Like, is it threatening to the building or is it just an inconvenience, a consequence, echo of the fire that's dirty, annoying. What's your experience with Code Red? I guess if you had a fire in the entire compartment and you really ended up with 0.5 hotspot per meter of the edge, you could compromise the entire structure along the perimeter of it. I guess in that case it's kind of obvious, but it also meant that you went through a significant fire in that building. I'm not sure if persistence smoldering fire after such a huge fire is the biggest challenge in that building. I wonder for a small building, let's say compartment fire in a bedroom that's contained to that bedroom. With the amount of load bearing pathways you would have in such a building, I would put a risky statement that it's not going to threaten the entire structure. I'm not sure, not the structural engineer. Don't trust me on this one.

No, I think it's a valid statement. I think it would be challenging, considering the localized nature of these hotspots, for it to impact the entire structure, but by virtue of the locations that we saw that smoldering intended to occur more frequently. Those are structurally speaking areas where stresses are going to be more concentrated, which means that if they do continue to smolder unaid it, then the implications are going to be greater than if the smoldering was occurring somewhere else. So I think that in terms of the magnitude of impact of a flame and compartment fire, they're probably lesser, but by virtue of the localized nature and where they're occurring, there could still potentially be some damage there.

I guess this also would go to what part of your structure was actually prone to smoldering If it was just a slab. There's a completely different issue than if it was slab columns and load bearing walls, for example. It's going to be a completely different case. Very interesting Going forward with the potential consequences of a prolonged, considered smoldering fire. Have you observed them going back to flaming by any mean, or perhaps no conditions under which they can go back to a flaming combustion? Yes, absolutely.

So in a good few cases, in a good few smoldering hotspots in our experiments we found that after a change in conditions the smoldering could actually transition back to flaming. What change of conditions? So typically it would be an increase in airflow. So, for example, in one of the hot spots we saw that a hole formed between the slab and the wall into place, which meant that air could flow from the outside through the hole to the inside of the compartment. And this increase in airflow feeds the smoldering sees that char oxidation more, which means that enough heat and enough ferrolesate is generated to transition back to flamin' combustion.

And this flamin' combustion. We're talking about a different floor, literally a different compartment, where it can re-occur. That's very interesting. And how about the production of products? Like, I mean, it's combustion, so it must lead to production of water vapor, products of combustion. Most likely it's inefficiencies, so I would assume the production of carbon monoxide here would be higher than flamin' combustion in well ventilated compartment. Have you done any measurements? Or perhaps, though, from literature? How does this relate to the flaming fires or what one would expect?

No. So I do wish that we had measured some of the gas composition, so some of it, like the, for example, the carbon monoxide and carbon dioxide during the smoldering. But I think smoldering tends to occur at lower temperatures and tends to be less efficient, which means that it does tend to produce more carbon monoxide than in flaming charge. So if you had smoldering in a poorly ventilated compartment occurring for a long period of time, that could lead to an accumulation of carbon monoxide which potentially could have safety implications.

My biggest concern is that a person could be present in a room that is technically not going through a fire. Then the fire is extinguished. There's an unknown hotspot going in the wall somewhere between that person's compartment and the compartment next door. The person comes back and it can be potentially a victim of this accumulation of CO. And if that's the case, then actually it also means that we perhaps need to have some precautions, like in allowing people going back to their homes after a fire has been present in a building with timber structure, because those checks must be on a completely different level than what you would have with a concrete building. That's an interesting consequence of smoldering being a phenomenon. Any other consequences that are not very obvious that you can see with presence of smoldering fires in mass timber buildings?

I would say that transition, as we mentioned previously, transition to flaming is one of the most significant. So, as I mentioned, with holes forming in the ceiling and transition to flaming occurring through those holes, there is the potential to lead to secondary fires in laboring compartments long after the initial fire has gone out. So, whereas you might not see that in a conventional compartment, I'd say that there are safety implications there in mass timber buildings, where a smoldering hotspot could potentially lead to a secondary fire. It was in the hours or days after the original.

So one more scenario that goes through my head, and I'm not even sure if it's physically possible like I outlined it, but I can imagine a scenario where there's an unattended compartment let's say someone went to their vacations and there's a kitchen appliance that's in the corner of a CLT wall, unattended, and for some reason it catches fire. It's not big enough to transition into a flash over to create a large fire in that compartment, but perhaps it was enough to trigger a hotspot in that particular location. This means you had a fire about which no one knew because the room was unattended. There's no. Usually, at least in Poland, there would be no automatic smoke detection in that compartment. It may be different in the UK. So you may have a fire that's ongoing that no one knows about, and then two, three days later their laborers has an unexpected fire occurring from a wall. And perhaps I'm exaggerating. Perhaps such small, localized fire source would not be enough to trigger it. But again, it's based on the knowledge I have today as a scenario I could not exclude completely, whereas if I'm based in my concrete block I am perfectly protected against the stupidity of my neighbor. No fire from their side can penetrate to my side, at least not easily. Is it something perhaps too far? How do you view such scenarios?

Yeah, I agree that it is a possibility. So it's a question of whether it's that better or worse than if that initial hazard had caused the flaming fire. So I suppose, by virtue of sluldering being able to occur through the full thickness of a timber element, it could lead to ignition of a neighboring structure, a neighboring compartment, whereas it had just been a flaming compartment, that might not have been the case. So I think there are different implications than if it had just been a flaming fire. I think also, as you mentioned previously, if that family had been gone for a few days and that sluldering had continued to release carbon monoxide, could that have implications for halve back of monoxide accumulation inside the compartment. So when the family comes back, what's the implications there? I think it's something that has to be considered when you have all these complex design components next to master elements, for example, you could have a short circuit of an electrical element.

Yeah, exactly, and you would most likely put them inside your wall, even to conceal them from being directly visible. And that's the most common or obvious case or the reason of firing Half of the fire. Investigations that I know with short circuit would be called by. That's actually an interesting direction. And do you know any research on electrical arcing, on the ignition of sluldering fires in timber? Perhaps a specific, perhaps a subject for a new PhD in the Imperial?

Interesting, or a new postdoc? We never know. No, I don't know is my honest answer. I don't know if there is any research out there on it. I know that there's been some research in the context of wildfires on the implications of electrical arc-ing of power lines and that kind of thing, but I don't know if there's any research on how an electrical wire could cause smoldering of a timber surface. I know that, for example, in our lab when we research smouldering of peatland fires, we tend to use heating wires to get that process started. So from the analogy of smouldering peatland fires, it may occur in a similar way in a smouldering timber building.

That's again for the end. So it seems that there is still a lot of research to be done on this subject and definitely you've opened a whole bag of issues with the mass timber, with putting the smouldering fires on the forefront. You have done, or you have participated in the world's largest experiment on mass timber, I assume to study this we don't all have to do that. Do you see a scalability ability to test this or experiment with those phenomena in small scale? I know Hazelab is well known for their peat fire experiments, including land mines and stuff like that. So are you trying now to go from the field scale to lab scale and study them, or perhaps modelling? I hope?

so. So that's answering the question of modelling. Timber is an exceptionally hard material to model from a chemical kinetics perspective and my predecessor from my group knew very well. So I think from that perspective it's very challenging. But in the future, with a very keen PhD student who knows, but I think in terms of experimental reducing the scale and reducing the complexity from an experimental perspective, I think by the virtue of the fact that all these hotspots are localised, I think it would be a really interesting research path to try and understand because we've had these massive experiments where they've located the boundary conditions for where these hotspots typically occur. Let's now try and characterise that at the bench scale, try and characterise what parameters actually affect how fast smouldering spreads and if smouldering will go extinct by itself. So yeah, I think we've had the big scale and now, if people progress to a small scale, we can parameterise these problems and come up with some design changes from there.

And I think this would be very, very welcome, because here we're touching the subject of detailing, like really the details of how you finish your surfaces, how you overlap all these boards in those gaps. How do you build those tiny details In here? It really matters a lot and it's difficult to build 350 square metre compartments every time you want to check a different gap orientation or detailing. So definitely we need to figure out a reliable way to go, a smaller scale and, through this research, perhaps figure out ways that perhaps simply reduce the risk of those hotspots appearing or their severity or longevity. Perhaps that's a way to go. Anyway, ari, an hour has passed. Thank you so much for bringing the smouldering of Mastinbury elements to the listeners of FireScienceShow. I see there's a lot more. We need to learn about them, but it's great that we are already aware that this problem exists, is persistent and, thanks to your research, we understand some scenarios in which it can unravel and what it can cause.

Brilliant. Yeah, thank you very much, bocek. I'm Bocek Mastin. I've published to be on the podcast and I hope everyone's learned something.

Any resources I should link to people. I'm going to put the IFSS paper, but perhaps there's something else. Code Red Papers.

Yeah, I think the Code Red Papers provide a good broad summary of the experiments themselves. Also, if people want to learn a bit more in depth about smouldering processes as well, I can wholeheartedly recommend my supervisor's chapter in the SFPE guide. That was very useful for me learning about the topic.

Fantastic. Thank you, ari, and see you around Brilliant. Thank you very much, bocek. And that's it. Boy oh boy. Smouldering in Mastinbury. If we didn't have enough issues with Mastinbury buildings so far, it's kind of interesting. I still don't fully know how consequential it is. I mean, it's obvious that smouldering happens in Mastinbury buildings. It's obviously that we should take it into account. At the same time, there's not much we can do. There's no magical pain that you can apply to your intersection between the slab and the wall to prevent smouldering from happening. I guess it's something you need to acknowledge and live with. And now the question whose responsibility is that? Is it on designers? Is it on the firefighters? Is it on the building owner? I don't know. This episode leaves me with a lot of questions in my head. How do we approach that and whose problem is it? Because it's obvious that it can happen, that it can have consequences, it can progress, it can lead to damage, perhaps destruction of the structure. Anyway, I highly appreciate Harry's insight into that and opening this view for us. I cannot conclude whether this changes my view on Mastinbury or not. I guess not. Really just gives me another angle at which I need to view my structures and at which I need to provide some additional help to my clients. That's it, something to consider. So far, some great reading materials in the show notes of this episode, so please refer to those. Please cross your fingers for Harry's Viva. I hope it is Dr Harry Mitchell. Before the next podcast episode I'll upload the update, the episode, with congratulations if he succeeds. I'm sure he will. He is such a great scientist and a great speaker. That's it for today. I'm not going to answer any more questions. I leave you with the questions in your heads and if you have a different angle in this topic and you have some of your own thoughts on this, share them with me. I'm really interested. It's a really, really interesting subject that we brought it today. Anyways, hopefully next week, another interesting subject. As always, as every Wednesday, fire Science Show is here for you to deliver you interesting fire science, to deliver it in the easiest possible form and help fire safety engineers grow and develop in an easy and fun way, and that's what I'm doing in here by interviewing all those smart people, and I hope you benefit along me. Thank you for being here with me and see you here next Wednesday. Thank you, bye. This was the Fire Science Show. Thank you for listening and see you soon.