Thursday, 8 November 2012

Fire protection and polymers: natural flame retardancy

While investigating the limitations of the theory that explains the ignition behavior of a polymer, we discovered something unexpected, a new natural flame retardancy mechanism.

In the UK, a fire is started every 3 min, and over the course of a year, the cost of fire totals approximately £7 billion [1]. It is a major threat, and continues to be the leading cause of property damage worldwide according to the insurance company FM Global. In the modern world, polymer materials are ubiquitous because of its technological, manufacturing and commercial advantages. But they also fuel flames, and are a prime actor in accidental fires. Better understanding of how polymers burn is a necessity if we are to save human lives, protect infrastructure and the environment, and improve businesses.

Ignition is a key process in the initiation and growth of fires. The risk of fire is associated to the ease of igniting the materials present. This is true for the initiating event but also for the subsequent spread. For example, the flames and the hot smoke transfer heat to nearby fuels, igniting these too, thus leading to further growth of the fire.

Pyrolysis is the thermochemical process by which a solid (or liquid) decomposes and produces the gaseous fuels that feed a flame. When a solid fuel is heated it eventually reaches a temperature threshold where it begins to break down chemically (typically around 200 to 300 C). Pyrolysis is similar to gasification but with two key differences, i) pyrolysis is the simultaneous change of chemical composition (e.g. long hydrocarbon chains to shorter chains) and physical phase (i.e. solid or liquid to vapour); and ii) is irreversible. It is an endothermic reaction, meaning that it needs an external supply of heat to continue because the products carry more chemical energy than the original fuel. It does not involve oxidation reactions.

Watch this accelerated video to see the pyrolysis of a block of PMMA, a synthetic polymer used in plexiglass, when it is exposed to a strong source of radiant heat (arriving from the top).

Since World War II, laboratory experiments performed with radiation heat sources have provided a basic understanding of ignition. It has led to what is called the classical ignition theory. This theory allows to calculate the time it takes to ignite a solid fuel when it is exposed to heat. It was developed from experiments conducted at low levels of heat (in the range below ~70 kW/m2). The theory says that the time to ignition decreases with the square root of the incident heat. These calculations have been used extensively in fire science and in fire protection engineering for decades. Although the expression has been altered slightly many times as research developed, it has kept pretty much the same mathematical form (=inverse square root for all heat flux levels).

But in 2006, researches at Worcester Polytechnic Institute conducted experiments at high heat fluxes, up to 200 kW/m2 on a range of polymers (PMMA and wood, for example). Their experimental data could not be predicted correctly by the classical theory. The error at high heat levels was large. Instead of the expected continuous square root behavior, the measurements were diverging from theory with a gradual flattening towards a constant ignition time for heat levels above ~80 kW/m2. The researchers could not explain the phenomena but reported their measurements [2]. This data posed a challenge to the scientific community.

This is an important failure because in large accidental fires, most of the radiative heat arriving to nearby fuel items is above 100 kW/m2. Thus, if this error is not corrected, predictions of the pattern and the rate of spread of a fire will be erroneous. Also, this failure of the ignition theory marks a limitation of our understanding and hinders the development of new fire protection technologies.

Measurements of the time to ignition of PMMA samples found in the scientific literature. The cloud represents experimental uncertainty. Top: Inverse of the square root of the delay time to ignition vs. heat flux (inset: zoom for heat fluxes up to 60 kW/m2); Bottom: Delay time to ignition vs. heat flux. Figure from [3].

So, in 2008 we picked up the challenge and tried to solve this riddle. We conducted a detailed investigation [3] of all the experimental data in the literature for the polymer best studied in fire science: PMMA. Data extended from low to high heat levels, see figure above. We then used a comprehensive numerical model of pyrolysis to revise all the assumptions cast in the classical ignition theory. We interrogated the experimental data using the numerical model to tell us why the failure. We wanted to identify the assumption and the mechanism (or mechanisms) responsible for the unexpected failure at high heat levels. All possible physical and chemical assumptions were systematically studied, one-by-one and combining them. We found it at the end. The classical ignition theory makes a wrong assumption and misses an important mechanism, a physical one, related to radiation heat transfer (and related to optics as well).

The problem is that the classical ignition theory assumes that the radiation is absorbed at the exposed surface of the material. We found that this is a good approximation to all materials at low heat levels, but it is not a valid assumption for many materials at high heat levels [3]. The assumption breaks down at high heat, with PMMA for example, because this material is translucent to some radiation. We could correct the ignition theory by taking into account that a fraction of the radiation penetrates directly in-depth, into the material, such that the surface heats up less. This leads to slower ignition and slower fires. The elusive mechanism is called in-depth radiation absorption.

This discovery was also reached simultaneously and independently by researchers at FM Global [4], although using an analytical approach and a smaller experimental data set. We learn about FM Global's work after presenting our findings at an international conference (BCC 2009 Recent Advances in Flame Retardancy of Polymeric Materials), so we were lucky to be able to cite them too and include their data in our final version published in 2010 [3].

The discovery is important because many polymers are known to exhibit some degree of transparency to radiation. PMMA is just one example, the example for which most fire data exists. Due to in-depth absorption, a material delays ignition because heat reaches directly deep into it thus leading to lower temperatures at the surface and hence taking longer to reach ignition. The work shows that in-depth radiative abortion acts as a natural fire retardant in polymers; it helps to 'cool down' the surface when heated.

This mechanism could be exploited by the plastic industry to design new polymer formulations that favor materials that are transparent to radiant heat and absorb less at the surface but more in-depth. It might help to formulate physical flame retardancy, whereas currently the plastic industry relies mostly on chemical retardants.


[1] An Introduction to Fire Dynamics, 3rd Edition, 2011, by Dougal Drysdal, Wiley. 

[2] Flammability characteristics at applied heat flux levels up to 200 kW/m2, by P Beaulieu and N Dembsey in  Fire and Materials, 32(2), pp. 61-86, 2007

[3] Numerical Investigation of the Ignition Delay Time of a Translucent Solid at High Radiant Heat Fluxes  by N Bal and G Rein in  Combustion and Flame 158, pp. 1109-1116, 2011.

[4] Absorption of thermal energy in PMMA by in-depth radiation by Jiang, De Ris and M.M. Khan, Fire Safety  Journal, 44 (1), pp. 106–112, 2009

Wednesday, 31 October 2012

Writing in the Sciences: a keystone of the future

I had a skype teleconference with Oriol Rios last week.

Oriol is a bright MSc student at Ghent University to whom I had the pleasure to teach fire dynamics last year. He did great in my course. He was telling me how much he is enjoying a massive online course on 'Writing in the Sciences'. This is an open and free registration course to "train scientists to become more effective, efficient, and confident writers" taught by Prof Sainani at Stanford University. I immediately agreed with the objectives of the course and praised Oriol's initiative. Hope more scientists would be taking it. Indeed, I will be taking the course myself soon, and will be strongly suggesting it to my students and postdoc. We can only improve science by learning to write better.

A bonus to our conversation was Oriol's first homework in the course, which involved writing a short summary of a 'hot paper' in each students' field of expertise. He chose one of my papers (nice!). The text is below. I was honored by his summary and understanding of what we attempted to do.

A hot paper in a hot science field; Fire Safety
 by Oriol Rios
Thinking about “hot papers” in the fire science field is inherently funny; the scientific approach to fire safety is a “hot field” –it's main journal was first published in 1977- and so is the object of study. "Forecasting fire growth using an inverse zone modelling approach" (W. Jahn , G. Rein, J.L. Torero, 2011) stands out due to its innovative and revolutionary approach to forecast building fire dynamics. Jahn et al. explore and validate a novel forecasting technique based on data assimilation and inverse modelling. Sensor observations of an enclosure fire (e.g. fire in a bedroom) are gathered during an interval of time (assimilation windows) to estimate the invariant parameters using an inverse modelling approach; evaluating the parameters involved in an equation knowing the result in advance. These values are inputs for a two zone model -a simple model that just considers a hot upper layer of smoke and a cold lower layer of fresh air- that finally forecast the temperature of the upper layer, the heat released rate, and the smoke layer height. The forecast is delivered with positive lead time, that is before the predicted event takes place. The method was validated using a Computational Fluid Dynamic (CFD) program to prove that 30s of observation leads to a successful 100s forecast. The ultimate aim of this new technique is to assist emergency response -particularly fire fighters crews- by giving them a schematic description of the situation and the expected fire development before they enter the scene. This paper stands out as important because it is the first to provide a method of delivering a reliable forecast with positive lead time. Although the envisioned tool is still far from operational and more research must be conducted regarding complex fires, the authors suggest that the necessary data to run the model could be obtained just by tweaking the sensors that are already present in many new buildings -smoke detectors, temperature sensors and so on. This provides a powerful tool with a simple set up and low computational cost; A keystone of future fire safety engineering.

Note by G Rein: A minor comment is that the paper uses CFD simulations as sensor data, not for validation.

Friday, 28 September 2012

Imperial Haze Lab

Today was my last day at the University of Edinburgh. I left my office this afternoon. I look forward continue collaborations with Edinburgh. It has been a real pleasure and privilege to work there since 2006. I now move to Imperial to continue the growth of the Edinburgh school of thought.

In my Edinburgh office in 2007
My former office in Edinburgh, on the day I left. Empty!

With my arrival to London, a new fire research group is created: the Imperial Haze Lab. It starts small, with one PhD student (plus four IMFSE students). We expect to gather pace and size in the coming years with topics of research on fire dynamics and reactive solids, in both the built or natural environments. Some examples of planned research are forecasting fire dynamic (applications both to buildings and to wildland), pyrolysis modelling, travelling fires for structural design, fire threats to renewable energies, smouldering wildfires (the largest fires on Earth) and carbon sequestration in char.

My new full contact information is:

Dr Guillermo Rein
Mechanical Engineering Bldg, room 711
Exhibition Road, Imperial College
London, SW7 2AZ, UK
G.Rein at
 Let me know when you are in South Kensington; I have already found a few places with excellent cafe.
My new office in London. Empty!

Thursday, 6 September 2012

Chemistry of peat fires

Proceedings of the Combustion Institute
NOTE: This paper received the Distinguished Paper Award on Fire Research at the 34th International Symposium on Combustion by The Combustion Institute.

We have just published a paper in the Proceedings of the Combustion Institute on the chemistry of smouldering peat:

 "Study of the competing chemical reactions in the initiation and spread of smouldering combustion in peat"

Smouldering is the slow, low-temperature, flameless burning that represent the most persistent type of combustion phenomena and which leads to the largest and longest burning fires on Earth. Smouldering megafires in peat and coal deposits occur with some frequency during the dry season or eventual droughts in, for example, North America, Siberia, the British Isles, the subartic and South-East Asia.

In this work, we use an experimental methodology to study the smouldering combustion of samples of peat under a wide range burning conditions. By varying the oxygen concentration and the ignition conditions we investigate the competing pyrolysis and oxidation reactions.

We focused on the three main solid species involved in smouldering fires: peat, char and ash . It shows clearly how pyrolysis concentrates carbon in the char while a large fraction of the hydrogen is released, while the oxidation releases most of the carbon and concentrated the minerals in the ash which H, C and N contents are negligible. The fraction of carbon in char is ~1.5 times higher than in peat, and ~35 times higher than in ash. The change is even greater in terms of carbon density, it increases from 77 kg-C/m^3 in the peat to 133 kg-C/m^3 in the char, to then sharply drop to 0.7 kg-C/m^3 in ash.

The experiments clearly show that there are pyrolysis and oxidation reactions. Char is formed by pyrolysis and consumed by oxidation. So at the beginning of a test there is no char, and at the end only a small amount of char remains, but in between substantial amounts of char (~50% of initial weight) were momentarily formed. Smouldering produces and consumes its own char: it initially produces char through pyrolysis before being consumed by char oxidation reactions. The competing nature of the production and consumption char reactions has been experimentally shown (see figure below).

 Evolution of peat and char fractions through an experiment.

The virgin peat reacts during the first 15 min to produce char and ash. Thereafter, only char reactions take place producing ash. Tracking the amounts of peat and char at any given time shows that first char is formed. It reaches a maximum fraction (~50% of the initial mass) in 20 min and then slowly the char is consumed down to ash (10% mass). At the end of the experiment, 90% of the initial mass has been released as gases, leaving a void and a thin layer of ash.

By varying the oxygen concentration and the thermal conditions we investigate the competing pyrolysis and oxidation reactions at a fundamental combustion level. The figure below shows infrared images of the surface of samples at different oxygen levels (21% is normal air) during the early burning stages of ignition (5 min after first heat exposure). As the oxygen level is increased, the temperature of the sample surface increases (indicated by brighter colour) showing that although pyrolysis dominates in this early stages of spread, oxidation reactions also play a role.

 Infrared images of the sample at [O2] of 17%, 21%, 25% and 35%just 5 min after first heat exposure.

The results presented here can be used to advance our fundamental knowledge of large-scale smouldering wildfires which are currently not well understood.

Title: "Study of the competing chemical reactions in the initiation and spread of smouldering combustion in peat"
By: Hadden, Rein and Belcher In: Proceedings of the Combustion Institute (in press), 2012.

Friday, 24 August 2012

Two new doctors in fire science

I am delighted to announce that my two PhD students Freddy Jervis and Nicolas Bal have recently defended their theses successfully and are now Doctors of Philosophy in Engineering from the University of Edinburgh. Congratulations. See below the announcements circulated.  

Dr Bal. Email sent by Dr Welch on 21/08/2012 -----
Dear all,
It is my pleasure to inform you that Nico Bal has successfully defended his PhD thesis in the viva exam today, subject to minor editorial corrections. His studies were sponsored by the BRE Trust and supervised by Guillermo Rein, the thesis title was:

"Uncertainty and complexity in pyrolysis modelling"

The external examiner was Chris Lautenberger, Principal Engineer, Reax Engineering (Berkeley, USA); Tim Stratford chaired the exam committee and I was again the internal (fourth time for Guillermo's students!).
By the end of the viva Nico had filled the white board with equations and graphs and we were in no doubt that there is a lot of both uncertainty and complexity in pyrolysis modelling. But out of this complexity Nico has established significant new insights, a great achievement and a platform for lots more interesting work in the field.
Many congratulations to Nico!
Stephen ---

Dr Jervis. Email sent by Dr Welch on 02/07/2012 -----
Dear all,
It is my pleasure to inform you that Freddy Jervis has successfully defended his PhD thesis in the viva exam today, subject to minor editorial corrections. His studies were supervised by Guillermo Rein and the thesis title was:

"Application of fire calorimetry to understand factors affecting flammability of cellulosic material: Pine needles, tree leaves and chipboard"

The external examiner was Prof. Xavier Viegas, forest fires expert from Coimbra University (Portugal)*; I was the internal.
Freddy did an amazing job investigating the intricacies of pine needles combustion using the Fire Propagation Apparatus, with extension to assessment of the effects of leaf morphology on flammability (which has implications for historical changes in fire activity arising from climate-driven floral changes!). At the other extreme his study also looked at the impacts of oxygen and heat flux on burning of chipboard, of relevance to fires in buildings.
Well done Freddy!
 Stephen ---

Friday, 10 August 2012

Forecasting fire dynamics: tomorrow's infrastructure protection

Imagine a technology able to forecast fires. It would lead to a paradigm shift in the response to emergencies and provide the Fire Services with essential information about the ongoing blaze with some lead time (i.e. seconds or minutes ahead of the event). It would also allow for the future of infrastructure protection to be implemented in smart buildings.

316 s after ignition. could this be forecasted ahead of time? [Rein 2012]
But despite advances in the understanding of fire dynamics over the past decades and despite the advances in computational capacity, our ability to predict the behaviour of fires in general and building fires in particular remains very limited. The state-of-the-art of computational fire dynamics is not fast or accurate enough to provide valid forecasts on time...

Paleofuture: forecast made in 1900 of the fire-fighting in the year 2000.
By Villemard, 1910, National Library of France
But we found a way to solve this problem. In a recently finished PhD thesis and set of published  papers, we show the technology is possible. We  propose to use sensor measurements of the ongoing fire to steer and accelerate computer simulations. This takes advantage of the concept of data assimilation (similar to what meteorologists do to forecast the weather).

Our method consists on combining a simplified spread mechanism with a fire model, and use sensor data to find the fire parameters that dominate the spread. This way, the model automatically recovers information lost by approximations in the physics, chemistry and the maths.

Concept of data assimilation and the sensor steering of model predictions [Cowlard et al 2010]
 A series of compartment fire cases haven been studied this way, and we investigated two different fire models. First a simple two-zone model, and then a state-of-the-art computational fluid dynamics (CFD) model.
For the simple two-zone forecast model, the firepower and the growth rate were estimated correctly up to 30 s ahead of the event: the model was faster than the fire. This was the very first time a fire forecast technology was demonstrated and the first time positive lead times were reached. The results show that the simple model is able to deliver fast and useful information about the ongoing fire thanks to the sensor data. This initial work demonstrated that the new methodology is effective, and allowed us to move to the next level of complexity. 

Computational domain of the fire
compartment [Jahn et al 2012]
For the CFD forecast model, we use a coarse grid that provides short computation times. Spatially resolved forecasts were obtained in reasonable time. It is even possible to estimate the growth rates of several different spreading fires simultaneously. Although actual positive lead times were not reached here with CFD, it is shown that the use of relatively coarse grid size in the forward model significantly accelerates the assimilation (up to 100 times faster) without loss of forecast accuracy. Actual positive lead times with CFD are possible by reducing the computational time by at least another order of magnitude in the near future using high performance computing techniques.

Dalmarnock Fire Test One conducted on July 25th.
Our latest bit on the topic was a test case using the measurement data  from a real fire. We forecasted in near real time the Dalmarnock Test One, conducted in 2006 inside the 3.5 x 4.7 x 2.4 m living room in a high rise building in the city of Glasgow. It was possible to find a good fit between the observations and the forecast using CFD.

The results are a fundamental step towards the development of forecast technologies able to lead the fire emergency response. The work opens the door to forecasting fire dynamics, but it is an on-going research topic.

We are happy that the work has been featured in the media and  people is being exposed to this novel idea:

Our research resources on the topic (in reverse chronological order):

1) Jahn, Rein and Torero (2012), Forecasting fire dynamics using inverse
Computational Fluid Dynamics and Tangent Linearisation, Advances in
Engineering Software 47 (1), pp. 114-126. doi:10.1016/j.advengsoft.2011.12.005

2) Rein (2012), Plenary Keynote: Numerical forecasting of fire dynamics: tomorrow's infrastructure protection - Young Investigators Conference of the European Community on Computational Methods in Applied Sciences (ECCOMAS), Aveiro. See below.

3) Jahn, Rein and Torero (2011), Forecasting Fire Growth using an Inverse CFD Modelling Approach in a Real-Scale Fire Test, Fire Safety Science 10, pp 1349-1358, doi:10.3801/IAFSS.FSS.10-1349 

4) Jahn, Rein and Torero (2011), Forecasting Fire Growth using an Inverse Zone Modelling Approach, Fire Safety Journal 46, pp. 81–88. doi:10.1016/j.firesaf.2010.10.001. Paper shortlisted for 2010 Lloyd's Science of Risk Prize.

5) Jahn (2010), Inverse Modelling to Forecast Enclosure Fire Dynamics, PhD Thesis, School of Engineering, University of Edinburgh.

6) Cowlard, Jahn, Empis, Rein and Torero (2010), Sensor Assisted Fire Fighting, Fire Technology 46 (3), doi:10.1007/s10694-008-0069-1

Numerical forecasting of fire dynamics (Plenary YIC ECCOMAS)