Fretwork | Are fires more toxic when FRs are present? (long version)
definitions-template-default,single,single-definitions,postid-766,single-format-standard,ajax_fade,page_not_loaded,,qode_grid_1300,footer_responsive_adv,qode-content-sidebar-responsive,qode-child-theme-ver-1.0.0,qode-theme-ver-10.1.2,wpb-js-composer js-comp-ver-5.1,vc_responsive

Are fires more toxic when FRs are present? (long version)

Are fires more toxic when FRs are present? (long version)

We have been asked “do flame retardants really increase the toxic load of fires and create an additional risk to anyone in a fire scenario”.

This seems to be based on the theory that gas phase flame retardants (GPI) remove oxygen from the vicinity of a fire and thus create oxygen depletion which in turn causes the fire to be more toxic. It is not certain if the claim also adds the breakdown products of the fire to the toxicity or simply delays the point where fire breaks everything down to very simple chemicals.

This sounds like a case for STUART (Science, Technology, Understanding And Reasoned Thinking – see

We have discussed this idea within our Network and received contributions that are represented here. Not all are directly from those involved in the treatment of textiles – and it was surprising how much the science and technology crossed over between different professions.

STUART is concerned because any answer will by definition be long and involved and who will read it? He is reminded about the argument over diesel cars and the concerns over the mortalities they can cause through traffic pollution. You can find a very well-reasoned answer to this question at:

The author will, we hope, forgive the comment that the answer is also a bit long and tedious. We can be certain it is totally correct and moving quickly on to the last few paragraphs you can find a sort of executive summary of his take on the issue that (probably) very few people have taken the trouble to read. Many may have read it and simply ignored it because it did not meet their expectations or the type of result they were looking for. It is fair to say that the 40,000 deaths figure was really only a scare headline and a more reasoned approach to the data was possible. The issue was described as “zombie data” in the Daily Telegraph i.e. you can kill it by detailing the answer to its ‘headline fear story’ but it still comes back to life.

The reason this is so relevant is that the engineering behind the burning of fossil based fuel to avoid pollution involves trying to find exactly the sort of burning process that is theoretically possible in any fire scenario and that is where we started! Obviously a ‘clean burn’ is some kind of ideal to be achieved.


You may even recall that some years ago the Government started to advocate diesel fuelled cars because they were ‘cleaner’ only to change direction when it became clear that nitrogen and carbon particles could be involved, so diesel is now ‘dirty’. We should all recognise a so-called clean burn and the opposite which is (surprise, surprise) a dirty burn and the use of the extremes of definition.

We have been asked “do flame retardants really increase the toxic load of fires and create an additional risk to anyone in a fire scenario”.

FRETWORK is concerned with the fire risk presented by textiles in everyday use.

The UK Upholstered Furniture Fire Safety Regulations (aka FFR) were drawn up in response to the risk assessment that severe consequences could result if upholstered furniture is ignited in the home.

(see ).

The approach taken was to try and influence potential fires in terms of where and how they are likely to start and at exactly the point that they start. The use of flame retardant treatments for the foam filling and textile covers have proved extremely effective and there is overwhelming data to support this approach. The whole subject has been very intensively researched.

We have been working with ignition controlling measures for this type of fire since the introduction of the FFR. (see ).

It was a significant factor in the considerations leading to the introduction of the FFR that the consequences of fire scenarios initiated by upholstered furniture had several clear features:

1.   It was unusual for someone to be present when the fire actually “started” – the delayed ignition problem. The data was quite stark in recognising that fires “broke out” in the early hours of the morning when the house occupants would be asleep but the source of the fire had been created some time before.

2.    Many deaths were particularly distressing to firemen attending them because the fire started in one room but the smoke and toxic fumes developed and spread through the dwelling ahead of the fire. Intervention by the Fire Service was often timely for the fire but too late for the occupants because of the smoke and toxic fumes generated by the way the fire burned. Occupants were more likely to be poisoned than burnt to death.

3.    We use 2 different types of ignition tests for the FFR, one with a smouldering and the other a so-called open flame source. These have become commonly known as the Cigarette and Match tests respectively, reflecting the fact that smokers’ materials were often implicated in these fires.

The FFR is based on a strategy of trying to control ease of ignition. Questioning of the role of flame retardants in adding to the toxic load of a fire requires that we look beyond the ignitability and look at the type of fire we are dealing with. We must now look at burning behaviour rather than ease of ignition and we must recognise the difference between clean and dirty burning behaviour.

“do flame retardants really increase the toxic load of fires and create an additional risk to anyone in a fire scenario”

Clean burning can be seen in many households, gas fired central heating is a good example. The fuel supply is simple and clean (propane and butane gas), nearly completely carbon and hydrogen and thus relatively pure chemically. It is a gas so requires little heat to burn. Modern boilers ensure any available air is well regulated so burning can be efficient. If a boiler is faulty and has too little air supply then we know that it can produce carbon monoxide and be dangerous. Working efficiently, it will turn all the fuel into very simple molecules – virtually completely breaking down everything and leaving residues that are not toxic. Your gas boiler does not normally smoke!

So where does dirty burning come from?

If you take the case of carbon and fully combine it with oxygen then we have carbon dioxide. If we have a less efficient process with less oxygen available we will find carbon monoxide or carbon combined with one molecule of oxygen (written as CO) rather than the 2 found in the dioxide (written as CO2) – it’s simple really!

Except this is where it gets complicated in a way that some people, not surprisingly, find hard to understand. Both CO2 and CO are toxic to humans. We breathe in the oxygen from the air and breathe out CO2. Our bodies manufacture CO2. We have systems to deal with it but, as with many chemicals, too much and we cease to function. CO, however, is highly toxic and causes a narcotic effect that quickly leads to confusion of the senses and death. It blocks the systems we use to take in oxygen and we have difficulty overcoming its presence in our bodies. It is therefore possible to say it is much more toxic than CO2. Clearly, toxicity can have a wide range of effect so we should be careful in using it as an unqualified expression of risk.

The other common substance resulting from an incomplete and less than efficient ‘dirty’ burning process in fires is hydrogen cyanide (also written as HCN) and that is also very toxic. In fact, it is quite normal for anyone exposed to the smoke and toxic fumes in a house fire to require treatment for cyanide poisoning.

Most chemicals and substances that go outside the limits that our bodies can cope with are toxic but there are very important issues with the amount we take in and the way it affects us and this is sometimes called the dose and overdosing we know is a serious risk to the way we function.

So, we may recognise that restricted oxygen will produce more toxic chemicals and that is at the heart of the dirty burn concept. A combustion engineer would describe its lack of efficiency, a firefighter would be concerned with higher toxic gas levels and temperature.

When we take our knowledge of real fires in buildings and more particularly fires in people’s homes we can look at the situation and easily recognise 3 things:

  1. We are not considering a simple fuel source but all the things that homes are made of.
  2. Modern homes are meant to be warm and draught free.
  3. The lack of a free and copious air supply is limited by modern housing design – which is all part of the thermal efficiency of the building – but that is another story.

Firemen arriving at such a fire scene would expect there to be occupants at risk and a possibility for rescue.

The Fire service will always assess a fire prior to entering a building.

They tend to only enter a burning property if people are believed to be inside.

They wear breathing apparatus and have a strict time limit whilst in a building.

They have a potentially dangerous job but do not take unnecessary risks.

The key to understanding this is the Fire Triangle. We use it to explain the use of flame retardants with textiles. Interrupting this triangle is the model we use to show how to inhibit or stop fires developing. That is textile flame retardancy in action.

Here we can see the relationship between fuel, oxygen and heat that is at the heart of the science of burning. All 3 components are needed for a fire to burn. The same rules apply to all fires.

Gas phase inhibitors are particularly effective flame retardants for coated textiles because they can deal with the complexity of modern textile design used to make furniture attractive for the consumer. They act to disrupt the fire triangle by breaking down when heated to release free radicals that disrupt the chemical process that allows oxygen to join in the chemistry of a fire. It is possible to then to describe them as “oxygen depleting”. They are very effective flame retardants for textiles. They act locally where the flame is applied as an ignition source during a test. You will see examples of tests on this site where a flame has been applied to a sample in a different place to repeat the test. The flame retardant is just waiting to do its job.

Another picture we use to describe fire is a graph of fire temperature plotted against time.

( see  ) The fire graph does not show any restraints on fire growth, in fact it may be OK to demonstrate your barbeque in action but not a real fire. The problem is that fires in buildings are not very easy to typify i.e. it all depends on the circumstances. The graph shows the fire developing through ignition, growth, flashover, fully developed and decay stages. Clean burn can only occur when there is heat to turn sufficient fuel into combustion in the presence of an excess of oxygen. We are considering here that the fire will burn without intervention and will consume all available fuel. It will be allowed to burn out. It could be that this will be a chosen course of action by the fire service.

The science of what happens when materials burn is complex in the extreme as the chemistry of a fire shifts and changes as fuel and conditions change. We have described the toxic fumes that will be found but many of the products of such a scenario will be highly toxic – unless they have been broken down completely by the fire to very simple chemicals that are less toxic.

It is easy to describe the difference between an incinerator in action and a real fire and man’s intervention in an incinerator is designed to bring the chemical reaction as far towards completion and full chemical breakdown as possible. This is the way to create clean burn conditions and is the same idea as the car engine development.

There is, however, another way to look at this. Every method of fighting fire is based around interrupting the fire triangle. Adding lots of water through sprinklers or hoses removes a lot of heat to turn the water into steam and the steam dilutes the air supply thus reducing the available oxygen. Even the prevention of fire spreading by compartmentalisation (or fire breaks) will lead to reduced fuel and air access and will slow a fire and produce a dirtier burn condition. They deny the fire a continuous supply of fresh fuel by making compartments and can effectively create a limitation on the amount of air available. They tend to be made of inorganic materials that do not burn readily. They are normally the invisible but highly effective and important parts of fire-safe building design.

It is a simple fact that every man-made intervention in a fire is designed to disrupt the fire triangle and every approach in effect makes the fire dirty. Putting out fires makes them “dirtier” and that means more smoke and toxic fumes.

This leads us to a simple conclusion: fires in the home are dirty and dangerous because of the conditions in which they will normally occur. They involve as fuel a wide variety of materials.

Smoke detectors between the starting fire and the sleeping occupants are crucial to provide warning and possible escape time.

Smoke detectors respond best to the smoke and toxic fumes generated by a fire.

Intervention may be necessary to prevent the fire spreading to adjoining buildings or occupancies.

Textile flame retardants are used to try and stop fires starting or rather developing from relatively small ignition sources. Modern electronic devices seem to be capable of replacing the smokers’ materials at the heart of the original FFR. (see ).

Most of the data we have seen has been based on assessing the effect of gas phase inhibitors in a fire that has reached a clean burn state i.e. a fully developed fire with notable increase in heat supply and of course with a good oxygen supply to give a more efficient burn condition. These conditions involve temperatures well beyond survivability and a fireman’s normal protective equipment would not be suitable to allow entry to the fire. Survival in such a fire is impossible.

Unfortunately we are left with questions rather than answers.


Are the clean burn conditions that the tests describe found in a domestic house fire?

Are the GPI’s still present or would they have been activated at much lower temperatures in a real fire and thus be part of the dirty loading of the fire, suggesting that their presence in the tests was made artificially?

Are the conditions described purely found by researchers in a laboratory?

Are we considering more the subject of incineration rather than the  circumstances of a real fire?

The effects of having flame retardants in materials put through an incinerator has been the subject of research work and practical experiment. It has well understood impacts on end-of-life assessments for goods and also recyclability.

Are we being told that intervention in a fire should not happen at the risk of producing dirty burn conditions? – of course not.

Some conclusions – or afterthoughts(?)

Flame retardants are used to treat textiles to provide performance to criteria identified through an assessment of identified risks.

The amount of smoke and toxic fumes produced if flame retardants do their job will be very limited in comparison to a fire developing under the same conditions and far less than a fire allowed to simply burn out.

The data we have seen seems to be based on the effect of certain chemicals being present during the so-called clean burn phase of a fire.

We should be very careful in relating dirty burn conditions to the toxicity of chemicals.

Fires are dirty and dangerous and the advice to “get out and stay out” is worth remembering.

It is not unusual for it to be necessary to treat anyone exposed to smoke and toxic fumes in a fire for cyanide poisoning irrespective of what was burning in the fire.

Or are we being told that stopping fires is bad and we should look to make them burn cleanly?

This sounds like Science, Technology and Understanding Put Into Disarray. (STUART thinks there is another acronym there?).

Peter Wragg
Peter Wragg
No Comments

Sorry, the comment form is closed at this time.