163 - Fire Fundamentals pt 11 - Soot in Fire Safety Engineering

I don't see that much research on soot formation in fires and soot effects in fires around , yet it is something very , very foundational to how we assess the fire safety in our buildings . And in this episode let me try to introduce you to the concept of the soot .

And this also kind of ties in with the previous episode where we've discussed the historical experiments on visibility in smoke by gin . So suit formation and suit in fires is something you could even call a follow-up to that previous podcast episode . In this episode we'll try to answer some important and interesting questions related to soot .

First , we're going to try to figure out where does the soot come from in fires and why in different types of combustion you'd find different amounts of soot as a result of that combustion . We're going to discuss what does it do once it appears in your fire . Where does it go ? Which phenomena does it do once it appears in your fire ? Where does it go ?

Which phenomena does it influence in the fires , and which of those are actually important to us for our everyday fire safety engineering ? Then we're going to talk about how do we include soot in our design ? Because , yes , we do . We actually do include that in our design and especially in our modeling . So a lot of important questions .

There's going to be a bit of chemistry . There's going to be some stories in the episode . I hope you'll enjoy them . If you don't like soot at all , feel free to use a magical number for soothield of 0.1 gram per gram . You can consider that as almost conservative assumption for fuels with unknown composition and go on with your life as a fire safety engineer .

If you want to understand why we've proposed this number as something that you could call conservative , then stay with me in the podcast episode , because I'm going to answer that question as well . So , yep , it's fire fundamentals . We're going to learn some fire science .

Really happy to introduce you to the concepts of soot , let's spin the intro and jump into the episode . Welcome to the fire science show . My name is Wojciech Wigrzyń 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 . Okay , let's go soot in fires and smoke in fires . Long before I became a podcaster , and perhaps even long before I became a published author in scientific fire journals , I was already annoyed by the suit .

You see , the reason I became a scientist is to seek answers to questions that were annoying to me , and most of those questions were related to kind of magic numbers and where does stuff come from and when you work as a fire safety engineer when you work as a smoke control engineer , actually , and you do a lot of CFD simulations you very quickly realize that

visibility in smoke is the thing that you're playing with . Like I once put a claim that we should stop calling it fire safety engineering .

We should call it visibility engineering , because , in the end , what we are doing is to try and build systems in our buildings that would maintain the visibility in smoke for the required duration of the time to escape the building .

In essence , that's what we have been doing and that is what we largely still are doing , based on the assumption that we need to keep the evacuation conditions tenable . We have the tenability criteria related to visibility , temperature , heat flux and so on , so on .

The visibility is always the first one that passes , and there's an episode of this podcast I think it's episode three with Gabriel Avigne , where we discussed this at large . Now , as you realize that your visibility in smoke is the driving factor , there are some ways to tune the visibility . Now , perhaps tune is the wrong word in here .

There are ways to fiddle with visibility . You put a lot of stuff into your models and you put some numbers into your models that highly influence how much smoke you're going to have . One of those numbers is the soot yield factor , and that's the number that defines how much soot is produced in your fires .

Now if you start playing with this , you immediately observe that the visibility in smoke changes . Basically , because you change the amount of smoke , it must alter the visibility . The effect is quite profound . Now , when you start doing your due diligence as an engineer and you want to give your best , what value are you supposed to put in your model ?

And as long as you can find values for very simple liquid fuels or gases methane , ethylene , methanol , propanol , stuff like that you can find suit yield values for that . Once you start venturing into complex fuels , you will soon find some crazy numbers . Like for polyurethane foam , you can find values up to , I think , 0.18 .

And yeah , the scatter is just tremendous and there's no suit-healed value for a vehicle or suit-healed value for a kiosk in a shopping mall . You have to figure out those values on your own and that's a heavy burden .

I think that's a heavy burden for an engineer to come up with values of their own when scientific consensus does not exist in a specific field , like we don't have a scientifically sound value of soot yield for a very complex fuel package in a very odd architectural setting .

We have values measured in a small scale laboratory apparatus for quite fewer fuels and yeah , because it's quite a heavy burden for an engineer , I decided that it's a science and I could do this scientifically and in fact I did . Then I've met Gabriel Levinier and he was also researching that . He's also an engineer .

He also came to very similar conclusions , as I did with the challenges related to the use of soothed yield in modeling .

He was doing sensitivity analysis for different parameters that go into the equations that yield the visibility in the end , and he found that soothed yield was one of the variables , or the amount of smoke that's a product of soothed yield was one of the key variables that changed the visibility value in the end in the biggest way .

I saw his presentation in SFP Europe first conference in Copenhagen , I believe , and then we've talked , we've talked and then we've repeat his study together , experimentally and with CFD modeling , with FDS , with ANSYS , and we came to the same conclusion Suitability is extremely important , extremely impactful in our analysis .

I'm going to continue the story later on in the episode when I'll introduce you to the value that we found as something that we believe is fairly safe for complex field packages for engineers to use . But before we get there , I want us all to better understand soot , to better understand what is the thing that we are talking about .

What is the thing that we are dealing here about ? So let's try and first answer the profound question what is soot ? The profound question what is soot ? So answering this question requires some chemistry , and let's start with chemistry of simple combustion . Take a simplest fuel . You can get like methane .

When you write an equation for the combustion of methane , you have methane . You add oxygen , you receive carbon dioxide , you receive water . These are products of perfect combustion . And the more complex fuels you put into those equation , you eventually , if it's a perfect combustion , it's going to end in the same way Carbon dioxide , water .

If there were other species in your fuel , you're , of course , going to get other products . If you add nitrogen , you're going to get cyanide and so on , but at large , the products of perfect combustion are gases and water . Now , the combustion is not always perfect and sometimes you'll get byproducts of that . One of the byproducts is soot .

But that's the short story . You can go here a little bit deeper . If you think about combustion , the first reaction equation I told you you put your fuel , you put your oxidizer , you receive the products . This is actually a lie . Well , it's not a lie .

It's a simplification , a massive , massive simplification , because it's not a simple reaction from your basic fuel to your products . In reality , what happens is there are dozens , if not hundreds , of different chemical reactions ongoing in a combustion zone in which the compounds go to different transitions into radicals , they merge , they break , they change their forms .

There's a ton of chemistry that's happening even for the simplest fuels . Actually , in some papers that are on the subject of chemical kinetics , and there's a paper on modeling studies of formation of aromatics in laminar , acetylene and ethylene flames , and these guys actually identify more than 500 different reactions that happen inside the flame .

That's the complexity of the reactions that are happening in our flames . Now , another thing is those reactions happen with different probabilities and different speed .

They don't happen infinitely fast , they take time to resolve and , as an effect of this complexity , it's not all molecules of your fuel that will go through the complete transition into the final products , into the last final step of that reaction that would yield you this carbon dioxide and water . Some of them will stop somewhere along the way .

Some of the fuel may not react at all .

Even there is , let's say , a competition between the reactions inside the flame , which of the reaction will occur , and , depending on the fuel that you put in and the conditions in which the combustion takes place , different reactions will complete before other ones and sometimes because of that you will receive an incomplete product .

This is why a combustion of very simple fuel such as methane can be incomplete . It's not that we've stopped in the middle of the equation , it's just . The equation is a simplification of a much , much as methane can be incomplete .

It's not that we've stopped in the middle of the equation , it's just the equation is a simplification of a much , much , much more complex phenomena that happen inside the flame . Now why this leads to formation of soot . So the reason is carbon . Of course , carbon is a molecule , is the most social of all molecules , I would say .

Carbon loves to bond with everything , and carbon also loves to bond with itself . Now , another interesting aspect of carbon is that when it bonds with other carbons , they tend to create those circular structures , the rings . We call them aromatic substances . Those are rings that consist of six , sometimes five , carbons .

They close a loop and that's a very stable chemical compound once it forms a ring . This is why those rings are so persistent . Now those rings can combine and form more complicated structures that consist of multiple rings of carbon .

This is where we start calling them polycyclic aromatic hydrocarbons , pahs , and those substances are in a focal light because they're also connected with cancer and stuff like that . So they're pretty nasty stuff .

Anyway , at some point point , if enough of those rings combine with each other , once enough pahs combine with each other , there are four molecules big enough that we start calling them soot . This is either true reactions between pahs between themselves .

There's also a transition called the H-abstraction , c2h2 addition , haca mechanism Very complicated stuff , but there are models , chemical models , that explain us how those particles connect with each other to form a really large particle . That's a collection of those rings and those flakes are called the soot .

This is exactly soot and this is how , chemically , the soot is born . Now , another thing that you must recognize is the fact that soot consists of carbon and carbon is fuel . Right , you can burn carbon . So why doesn't soot burn itself in a fire ?

In fact , it does In most or , in many cases , a lot of soot that's created within the flame itself is burned down in the flame . Well , outside flame is a very small region of space in which the chemical reactions and release of heat happens , but that's not the entirety of what we see .

Then you have regions of very high temperature beside the chemical reaction region or above the chemical reaction region if you are under the gravity of Earth and buoyancy causes flame to go up .

Now for different fuels . We have a property of the fuel which is called the laminar smoke point height , and this is basically a property , a measurement of how high the flame of a specific material or of a specific fuel can be before it starts emitting soot outside of the flame .

So if your soot is created in a reaction but then it enters very hot region of the flame , it undergoes further reactions , it undergoes further transitions , it oxidates , it basically burns out , acts as a fuel and turns into new products which hopefully are the simplest carbon dioxide and water if the combustion is complete . Now for different types of fuels .

This range is different . For methane that's almost 0.3 meters . So you can have 30 centimeters or 29 centimeters flame of methane and still do not see any soot coming out of that flame . For polystyrene , that's only one and a half centimeters the height at after which you're going to see soot coming out of that flame .

For polystyrene , that's only one and a half centimeter the height at after which you're going to see soot .

So if you burn polystyrene and your flame is higher than one and a half centimeter , you're going to see some soot escaping the flame and not oxidizing in time , not being completely oxidated , not being burned out by the flame , and it's released into the environment and when it escapes the flame we start calling that smoke .

There are interesting stories actually about that . I love the story that Jose Torero gives about the candle , and there were multiple lectures by Jose . I'm going to link one that I like the most in the podcast show notes , so you're very welcome to watch it . It's really good . But long story short . Think about candle .

Candle is is a product of engineering design , especially one particular part of candle and that is the wick . So , as I just told you , there's this length of a flame at which the soot will oxidate perfectly and will leave no smoke .

So if you're using candles to light up your castle , you don't want smoke entering your castle because it's going to smell bad and it's going to get dirty and it's going to be problematic . So you don't want your candles to actually release soot .

You can solve that by consulting with your nearby combustion specialist or attending a combustion symposium in the neighboring kingdom . But , assuming you're lighting your castle with candles , you probably don't have access to computational fluid modeling and the achievements of modern times . So you have to sort it in a different way .

And they actually have sorted it out by observing the flames of the candle and realizing some basic properties of the flame . The fact that the candle releases or does not release soot is related to the flame height , and the flame height is related to how much fuel you put into the combustion . Now the fuel in a candle goes through wick .

Basically , you have some sort of wax , some sort of solid fuel that melts . Then you have a piece of wick cloth that is drenched in this molten wax and because of capillary forces , the wax can travel upwards through the wick . Now it reaches a point where it's very hot , so it evaporates and vapors of wax burn out , creating your flame .

So that's basically how a candle works . Now , if your wick is too large , you're going to take too much fuel from the wax and your flame is going to be too high , and if it's too high it's going to emit soot and you don't want that .

So the way to deal with that is to manually trim the wick whenever it starts producing soot , but that would be annoying and a lot of work .

So actually , a specific type of wick was invented , one that it's made from two weaves and those weaves are weaved together in such a way that the wick bends a bit and once it bends , a part of that wick is gonna burn off in the flame region , basically maintaining the same length over the time , because as it grows , as you burn off wax and more wick is

exposed , it's going to bend more and it's going to burn out the tip of it . So , again , it's short enough to not cause sootiness in your flame . And you can explain this with chemistry . I just told you how it works .

Basically , the soot is produced in the flame and it's oxidated before it can escape the flame , and as no soot escapes , then you have a beautiful source of light and you don't create much smoke , and your king is very happy with the cleanliness of his castle . Now you can explain that also with a thermophysical number . We call it the Damkohler number .

Damkohler number is a ratio between the reaction rate and the transport rate . Basically , it's what I've just told you . So reactions , the transport is how quickly the particles fly through the flame before they leave the candle flame and they are not able to oxidize anymore . Reaction rate is how quickly they burn .

So if those two things are in perfect alignment , if the reactions take just this much time as it takes for the particles to be transported through your flame , then you're not going to get any byproducts that you don't want in your reaction .

It's funny because the medieval engineers or Renaissance engineers who were working on the candle wick , they didn't know Damkohler number . Damkohler was not alive yet at that point , but still they figured out . I've also learned about a very interesting thing that's happening in Japan .

So they have those very expensive inks traditional , extremely expensive inks for calligraphy . Calligraphy is a very popular thing in the Far East and there is a very specific black ink that they would sell for a large amount of money . And where that ink comes from is also a story of soot , and actually it's also a story of a candle .

So I've learned that the materials that are used to produce this expensive ink is actually soot from candle flames .

So basically , they have this soot factory , which consists of hundreds and hundreds of candles , and those candles are built in a very specific way so that the wick is a little longer than necessary , a little longer than necessary for the complete combustion or complete oxidation of the soot within the flame range .

What happens then is that those candles start to produce a tiny amount of soot , so a little bit of smoke escapes the flame from those candles , and above those candles there are some surfaces at which this soot would deposit , and the people would actually gather that soot that deposited on those plates , and then this soot is used to create the ink

. So I assume that this soot has some properties that are very attractive to the ink makers . Most likely it has a very specific particle size distribution that works very well when mixed with other substances to create the ink and gives the blackness that they look for with ink .

However , from my perspective , from the perspective of a fire safety engineer and someone who's passionate about combustion and all of those stuff , what these guys did actually is they manually trimmed the wicks to get a damkuller number at which a very specific particle distribution is produced , like they don't even know , most likely , what damkuller number is and they

have no clue . They've done that but through years of experience and this is some generation to generation art of creating of that ink . So the method to do that is very , very old . They found a number the ratio between the reaction and transport that produces soot that they like .

Now the funny thing is they do it with candles and I find it quite dangerous because people are walking through a sooty environment . That's really horrible and we're going to talk about that in a moment .

But I think with fire science , knowledge of thermodynamics and chemical reactions , we could just create a simple burner that would have this particular dump color number and would produce the soot that they need at an industry scale . So they would not have to collect tiny amount of soot from their candles .

However , yeah , it's a craft , it's a handcraft and I guess this gives some magic and gives a justification to the price point for the ink . So here you have learned the story of soot creation from a chemical perspective , from thermodynamic perspective and simply from observing the real world . This is how soot comes into light In fires . It's the same .

You also have chemical reactions in the flame , hundreds of them . The more complex the fuel , the more complex the reactions will be . A lot of those reactions are competing with each other . Not all of them will complete .

Then the incomplete combustion products will start merging to form soot particles and those soot particles will transition into outer regions of your flames .

In those outer regions , if there still is enough oxygen left and some heat that allows for oxidation , they will oxidize and disappear , and if there's not , they're going to transport into your building , creating problems with visibility in smoke that we have to deal with as engineers . One question that you could ask is there soot in the flame ?

And that could be a very interesting question . In fact , yes , there is , and in fact that's a really really important scientific problem how much soot there is in the flame . If you consider effects of the soot , one of the most profound effects of the soot is related to the combustion science .

People who study the flame itself or the combustion processes itself , the sootiness in the flame changes the color of their flame . So if there's no soot in the flame , the flame is going to be blue or almost invisible . That's when you have minimal amount of soot in there . That's why methanol flames are almost invisible .

There's this famous YouTube video of a Formula One car burning and the driver is escaping and he's shaking a flame from himself and you cannot see the flame . That's because he was covered in methanol and methanol produces almost no soot , so the flame is invisible .

Once you start adding soot to the flame , it becomes more and more yellow and actually the yellow color of the flame is a product of soot being present in there . Now why it's important for our combustion colleagues ? Because soot is related to radiation . Clean air has emissivity of zero .

It's clean , it's transparent radiation , which you can test experimentally by going outdoors to a sun and you feel heat on your skin .

This is because the heat can transport through the atmosphere a little bit of it , being taken by carbon dioxide and methane in the atmosphere , thanks to which we have global warming , but most of it passes through the atmosphere and reaches your skin , depending on the region of the world you are in , causing a pleasant warmth or perhaps even burns .

Now , once you start adding soot to your gas , to your air , soot has emissivity of almost one because it's black carbon , so it's almost a perfect black poly . Once you start adding soot to the air , you start getting a mixture that would have some emissivity . What emissivity ?

That depends , of course , on the amount of soot you add , and we'll reach that in a few moments . But you basically added a new property to your air . It starts participating in radiant heat transfer and in flame .

If you have soot in your flame , that flame starts emitting heat , starts radiating heat outside of the flame and this heat goes somewhere , can heat up new molecules , can heat up new fuel , can influence the reactions that are happening within the flame .

So actually , the fact that soot creates radiation , this changes everything in the combustion regime , changes everything in the combustion regime and when you study combustion , when you really try to model those chemical reactions that are happening inside of the flames , the presence of soot changes a lot and is a very , very important condition that you need to consider .

What other thing it influences ? From our perspective it influences toxicity for sure . So one thing is that soot oxidation would lead to carbon monoxide production and there is an observation that with increased soot production there's an increased carbon monoxide production .

You can read on that in Bart Mercier's and Tarek Beji's book the Fluid Mechanic Aspects of Fire and Smoke Dynamics in Enclosures . Actually that's the book I've read when preparing to this episode a very good read and I always recommend this book . So it affects the carbon monoxide production but also the soot itself . You can consider it as a vessel .

It's a very large particle . It's a very large particle of carbon that is in the flame is surrounded by other products of incomplete combustion . If we assume that soot is created from polycyclic aromatic hydrocarbons , you cannot assume that all of them form soot .

Some of them remain in their primal forms and those PAHs do actually sit on the soot particles , so some of them will just land on the soot particle and fly with the soot particle . Those particles are connected with cancer . Those are connected with serious health issues , so the soot itself is a very toxic or hazardous compound .

Another aspect of that comes from the size of those particles . So they have a really nasty size distribution . Most of them would be in the range of , let's say , less than a few micrometers . For most of the smokes , less than one micrometer .

And this is a really , really nasty size that you get , because if the particles are less than 100 micrometers , you can inhale them , so they're going to pass through your nose . You can inhale them , so they're going to pass through your nose . If they're smaller than 10 micrometers , they can enter the orthorhactic duct .

So this is when stuff starts getting into your lungs and then the epa classifies the dangerous pollutants actually into groups of , of course , which are larger than one , three micrometers , fine , which are larger than 1-3 micrometers , fine which are smaller than 1-3 micrometers and ultra fine that are smaller than 0.1 micrometer .

The fine ones , the few micrometers and less they can actually penetrate your blood system . So with this size of particles you get all the worst effects you can get you can inhale them , they can get into your lungs and then they can go beyond your lungs . This is why soot particles are so dangerous to humans .

And actually , if I think about my firefighter friends many years ago , you would imagine a firefighter as you know this guy dirty , sweaty , who just rescued someone from a fire , carrying an axe , and this is how the firefighter would look like . And the firefighters I know today are the cleanest people I know Like .

These guys are obsessed with cleanness and there's a reason for that . Cancer is a horrible thing in fire departments and it's such a horrible thing that people who risk their lives to save lives of others are at such a large exposure to cancer against substances and sometimes pay a very heavy price for their willingness to help .

Now they realize this risk comes from soot a lot and they do everything they can to separate themselves from soot . That includes cleaning the clothes , that includes specific ways of taking down your clothes , making sure that no soot particles is in contact with your body , any part of your body .

It's not just respiratory , it's also soot that gathers on your skin that also can penetrate and do harm to your organism . So definitely there are toxicological effects of soot .

Also , if I recall correctly , in my episode with David Purser , when we were discussing the toxicity in fires , david told me that he was able to approximate the amount of toxic gases that a person has inhaled by the amount of soot found in the lungs of a person who has perished .

So you can actually make this direct connection that if you start inhaling soot in fires , you start inhaling all other stuff that comes with the fire smoke and that can be fatal at some point . Finally , the aspect that we are all aware of of soot in fires is the obscuration effects .

So soot mixing with air , just like when we discussed radiation , it has emissivity factor and because of that , light passing through this , soot starts to be scattered , starts to be absorbed and on the other end of your smoke layer or the smoke plume , you see less light . Basically , less light passes through the layer of smoke .

How much of that passes through layer of smoke ? That depends on the amount of soot that you had in your smoke . Actually it's quite interesting and many people don't realize that , but there's a specific depth of a smoke layer for a specific amount of soot in it , which would make that layer non-transparent . It's optical depth of the smoke layer .

So if there's more soot and the layer is big enough that you basically scatter or absorb all the light , you see black smoke and you don't see anything behind it . If there's less than that , the smoke layer is a bit transparent and this property also influences the heat transfer . So how much heat does the layer emit ?

So if you have a very diluted smoke , it's going to emit less heat than a very dense smoke layer at the same temperature . That's an interesting property of smoke

. Also , if you have light sources within the smoke , you either see them or not , depending on how dense the smoke layer is . We can approximate that . So there are ways you can calculate if you're going to see the light or not .

The simplest way to do that is through the use of Lambert-Beers' law , which basically correlates the light intensity with an exponential relation of the amount of smoke that there is and the length of the optical path . So that's one way that would be commonly used .

It's not a perfect way because it does not account for scattering , it just accounts for absorption , and it accounts for absorption of parallel light lines , and technically it should be used for homogeneous aerosols , and smoke is very heterogeneous . We have different particle sizes . So it's not a perfect thing , but it's a thing . We use it .

We use it every day in laboratories to measure the smoke obscuration . So Lambert-Bierlo is the reason why we can measure smoke . If you see lasers in laboratories , those lasers are meant to measure the light obscuration . Basically , you shine a very powerful light source through a layer of smoke and measure how much of that light is gone to the smoke .

Once you measure that , you have transmittance . You can have extinction coefficient . This is how Gene measured the smoke layers in his experiments in 1970s , and this is exactly how we measure those today as well .

Now , this is very easy when you're measuring it directly in a physical experiment , but it's a little more complicated when you're trying to calculate that in a computer , because there's no easy way to shine light through smoke layer in a computer . You would need a very sophisticated numerical models for that , which we actually do not really use that widely .

We did some of those based on Lambert-Beer ray tracing and stuff like that , but it's not something I would call mainstream in fire science . Yet we do have a very clever trick to account for that . So in our smoke simulations we know how much mass of the soot there is in the air .

We know the mass density of smoke in the air and this is a very simple product of a transport equations in your CFD software . So your numerical model basically calculates how much soot there is in every part of your model at every given time . Now to go from this into extinction , we need to know basically how much area of the soot particles there is .

This is a way how we present that and it's called the extinction coefficient . Now to calculate the extinction coefficient , you need to know how much extinction coefficient does a gram of a soot bring right . So if you know how many grams you have , you need to know how much extinction coefficient does a gram of a soot bring right .

So if you know how many grams you have , you need to know how much each gram gives , and this is a value that was approximated by George Mulholland in NIST , and he has calculated that this value is basically similar for many , many substances that are releasing flaming smoke at 8.7 square meter per gram , and this value became an FDS default and it's something that

you're using every day , even if you do not know that . What Mulholland did is he took results from multiple , multiple experiments done in multiple laboratories for multiple substances and he basically found the best fit for that value . An interesting thing is that Mulholland did this for red light , emitting laser at 633 nanometers .

This is the wavelength at which this value was approximated . And now we enter another interesting property optical property of smoke . So I said that you'll have particles lesser than one micrometer and there's going to be distribution . In fact , yes , there will be a distribution of those particles .

They will have different sizes , they will have different diameters , but smoke is not spheres , it's not perfect spheres . It's like flakes of random shapes . They can be very long and very thin , they can be flat and almost square , all kinds of shapes and distributions you can get .

And because you get particles of different shape and actually those particles are in the similar scale as the light wavelengths . They interact with light in a different way . So different particles of different size will affect the wavelengths in a different way and actually it's kind of profound . You can see and measure that .

Jin in 1970s already has realized that and has provided us some knowledge about the ratios of extinction in different colors of light on the smoke , and he noticed that the differences are profound . So when you have the light that is most absorbed by the smoke would be the lower length .

So let's say the blue light , and we've measured that In fact , yes , the blue light is much more absorbed by the smoke . The green light , which is 520 nanometers , is not that much different than blue light . The least absorbed is the red light . So the red gives you much higher visibility if you think about it .

And then if you go into infrared spectrum , less and less is actually absorbed by the smoke . So the smoke becomes more and more transparent to those photons . Now how big is the difference ?

When we were doing our experiments with heptane , we ended up with something like 19 meters of visibility for the blue light and 30 something meters of visibility for the red light . So , yeah , that's actually quite a lot . That's like twice larger visibility in the red light than in the blue light through the spectrum of the visible light .

So the difference is actually quite big . Now you can actually account for that in modeling if you want it . There was another person in NIST her name was Whitman , and they've analyzed the relation of specific smoke extinction coefficient with wavelength and proposed the best fit .

That gives you an exponential low that allows you to estimate the value of your extinction coefficient specific extinction coefficient at different lights . So you could use that . Could use that .

Another thing that you could use this property is because different fuels will produce different smoke distributions and different particles will create different patterns of how light is absorbed . You can actually use multicolor densitometers to check out what the smoke is composed of .

Or you could even and there is research that I know of that you could actually use this combination of how different wavelengths are absorbed to figure out what is burning in your building . You could use this to remote sense what is the chemical that is actually being burned in your building . Pretty fascinating , isn't it ?

Just to shine the light through a smoke and you can figure out where did the smoke come from , even though for our lay eyes , every smoke looks the same it's black and annoying . Now , finally , I promised you some explanations for engineering . So all we've talked about so far is pretty complicated . And no , we're not combustion scientists .

We do not go that deep into suit modeling in fires , in flames . In fact , that's a huge part of combustion science and there are hundreds of people that are working on that worldwide , perhaps even thousands . As I said in the intro , in fire science we don't put that much love into soot and we don't go that deep into that .

Now we need to have something in our everyday engineering and that something is a soothill factor . You can , okay , if you go into combustion modeling , if you apply some finite rate chemistry at the dissipation model , at the dissipation concept , you can account for sooth creation . We use Brooks-Moss model to account for creation of sooth in turbulent flames .

That's pretty reliable and that's something that we would use . But in most of the engineering you would rely on very simple assumptions and you would just put a soothill value . Soothill basically represents how much of your fuel turns into soot in your fire . A very simple measure .

And now , as I mentioned in the intro , to find the soot yield value for complex fuels is almost impossible . There's no sources that will tell you how much soot is produced from burning a car . It's simply too complicated .

You cannot shrink a car , put it in the tube furnace and measure the specific extinction coefficient and the rate of production of soot in that fire . It's way , way , way too complicated , so we have to assume something . Now a lot of people would assume something crazy like a value for polyurethane foam , a very high number . That's probably quite safe .

A lot of people would assume something very bad like a value for heptane 0.035 . That is a very low value of your soot yield and you're going to start seeing a much less soot production that perhaps you would in a fire .

We were very concerned about that with my friend Gabriel Vigne , but because we're not combustion scientists , we're not material scientists , we looked at that problem from the opposite side . So how much does that influence our engineering ?

And because of the exponential nature of the visibility in smoke , because how Lambert-Beer correlation works , the more soot you put into your model , there's a diminishing effect of putting more soot into your model , like , you eventually lose all your visibility and there's no more visibility to be lost , so you're not losing anything .

We've actually found that the more soot yield you put at the scale of a compartment fire , the sooner you reach your visibility threshold . But at some point you stop seeing tremendous changes in your tenability criteria . And we found that points to be somewhere around 0.1 sooth yields , so 0.1 gram per gram At this level .

If we put that much soot into our models , if we put more soot than that , you would not see a significant diminish of your evacuation time available evacuation time because it would be already lost because of how much soot we've put in the first place .

However , if you put much less soot , if you use , let's say , 0.03 value , then you would extend your available safe evacuation time tremendously . Now we've also thought that 0.1 value . That place is somewhere in the middle of the range . So if you start with zero for perfect combustion , the suitiest fuels like toluene would give you like 0.18 .

This 0.1 value lies somewhere in the middle . We thought it's a nice simple value that proven to be quite robust . You're not doing a huge error by incorporating that value in your everyday engineering . So in reality , if your fuel had lower soot yield than what you've chosen , your results are quite conservative . For that case Perhaps it's an error .

I also hate when people call errors conservative assumptions . At some point they just are errors . But I think you don't make that big error . I don't think you will . In most of the cases your combustibles in your compartments would yield something very close to 0.1 . I really believe that .

And if in reality your fuels are much moreotier than what you've assumed so let's imagine you've assumed 0.1 , but in reality they're 0.13 , then the error you've made is pretty small . It's a few percent in the tenability rate . So if you assign any margin of safety to your calculations , you've covered that and , worst case , you're conservative .

You're not making a huge error . However , if you assumed your suit yield of 0.03 , and then your fuels turned out to be 0.13 , boy , the difference will be tremendous . It's a massive , massive error , massive underestimation of smog production in your compartment , and this is going to have consequences for engineering .

So this is what we've chosen , this is what we've proposed . There is a fire safety journal paper that I will link in the show notes , in which we've described our thought process and how we have achieved those numbers . It's not yet the final answer that's marked . There is much more to discover related to suits and fires .

This is a part of the grant visibility prediction framework that I'm developing with Bergeshire University , wuppertal , led by Professor Lucas Arnold , my co-supervisor of this grant , and Lucas with his postdoc , christian Berger . They're doing some great , great work out there . They are using MIE model instead of Lambert SPUR to figure out smoke properties .

They see really good results with that and they've also identified some errors that are related to how we calculate visibility compared with real-world measurements .

Those actually quite well match what we've observed with Gabriela in 2017 , a massive discrepancy between the value of visibility measured in the laboratory and something that we get in our numerical modeling based on Jin's model . I think we've talked about that a bit with Lucas 100 episodes ago .

But yeah , with the new findings , perhaps we're going to record a new episode together in some time in the future treating about the uncertainties in measuring visibility in smoke and the errors that you introduce with the common models . Perhaps , when we're one step forward in the project , we'll be ready to share that with you .

Anyway , this is one hour of rambling on soot problem and I think that's enough for you , my kind listener . That's enough soot for you . At one end , we're very primitive in the way how we treat soot in fire science . On the other hand , it's fundamental to us . It's so , so important . It's smoke that we deal with in the fire safety engineering problems a lot .

Visibility is the first sustainability criterion to pass its critical value . It's something that defines the design of our building , so I guess it's nice to understand it how soot is created , what does it do , what it consists of and what's the physics of how it interacts with light . I hope this was a fun episode for you .

There will be more links in the show notes . So if you want to check the hostess video , if you want to check our papers on visibility in smoke , our paper on multi-wavelength measurements of smoke obscuration , they're linked in the show notes . Be my guest and do some more reading . I recommend Bart Mercy's and Tarek Beji's book .

It's generally very good for any fire scientist to get familiar with that book , even though some parts of it are very high level and very challenging to read . But it's quite comprehensive . I like it . So , yeah , there are resources , you're armed with knowledge and now go out there and do some really good fire engineering .

And thank you for your willingness to learn and your willingness to spend time with me . It was a pleasure to have another episode of Fire Fundamentals in the podcast . Next week we're back into interviews . We're going to talk about engineering , or maybe we're going to talk about AI , I don't know yet We'll see .

I'm sure it's gonna be a fun episode for you in the next week , so make sure to be here with me next Wednesday . Oh , and I'm releasing Uncovered Witness season two on Monday , so you perhaps want to check that as well . Thanks for being here with me . Cheers , bye , cheers , bye .