Archive for April, 2015

by: Noah Ryder and Jason Sutula

This topic was presented at the International Symposium on Fire Investigation Science and Technology (ISFI 2014, September 22-24), which took place at the University of Maryland in College Park, MD.

ABSTRACT

Over the last several years, a continued push to design and build “green or sustainable” buildings has accelerated throughout the United States.  According to the U.S. Environmental Protection Agency, the term “green building” is defined as “…the practice of creating structures and using processes that are environmentally responsible and resource-efficient throughout a building’s life-cycle from siting to design, construction, operation, maintenance, renovation and deconstruction.” The goal of green building construction is to reduce the overall impact of building and development on human health and the environment.

While the concept of green building construction and green materials is important from a global perspective to ensure minimal impact of growth and development on the environment, little thought has been given to the impact of green building construction and green materials on fire initiation or fire growth within green buildings or with green materials.  The risk of using these materials in building construction has been highlighted by a number of recent fires in which firefighters were killed or injured, property losses have been excessive, and unexpected fire damage has been observed.

In the event that a fire incident occurs within green building construction, the type and installation locations of the green materials can result in enhanced heat release rates and pathways of fire spread not typically observed with standard construction materials.  The composition, construction, and placement of green materials in newly constructed structures runs the risk of creating fire damage that may be misinterpreted by investigators after a fire incident.

Thus, a need exists to develop a methodology that can be used to evaluate and compare the potential fire growth risk associated with green materials.  This paper proposes the use of the cone calorimeter as a standard test method that would allow for relevant material properties and material performance data to be obtained on green materials.  Example data is presented and linked to the potential consequences of fire growth on green materials as well as the probability of a fire occurring or spreading.  A simple formulation is then proposed that can be used to compare the relative performance of green materials and the risk associated with them.  Finally, the example data is extrapolated within a limited example of a post-fire scene reconstruction to assess resulting damage patterns in the context of green materials.

INTRODUCTION

The modern green building movement has roots from the rapid increase in oil prices that occurred in the 1970’s in the U.S.[i]  A byproduct of the gas crisis was increased funding in research to improve energy efficiency in all aspects of building development.  This included research into construction methods, more efficient energy consuming devices, and green materials.1 While the concept of green building and green materials is important from a global perspective to ensure minimal impact of growth and development on the environment, little research has been conducted on the impact of green construction and green materials on the potential risk associated with fire initiation or growth within green construction or with green materials.  In the event that a fire does occur, green materials may release significantly increased quantities of carbon dioxide above standard material.  Thus, the sustainability of the structure is called into question.

The goals of this paper are to summarize some of the current fire data available for green materials, present various test methods available for assessing the flammability hazard of green materials, and offer a potential methodology for an effective comparison of the relative risk associated with the materials.

Risk Associated with Green Materials

The risk associated with green materials has been highlighted by a number of recent fires in which firefighters were killed or injured.  These fires most often occurred in newer construction where engineered I-joists were used and the floors collapsed in short order.  This has caused a number of fire departments and municipalities to call for an increase in understanding of how green materials perform under fire conditions in order to determine whether it is safe for fire fighters to enter a structure involved in fire.

Risk is an inherent part of life and takes on a special role when examined from the perspective of fire.  Risk must be identified, assessed, and mitigated to the greatest extent possible within a reasonable cost.  This is true whether a process is being assessed within a chemical plant or if a green material is being evaluated for sale in the commercial market.  To quantify it, risk is typically presented as[ii]

Risk = (Consequence of Incident) x (Frequency at which Incident Occurs)

The process of assessing the risk begins with identifying hazards associated with the process or the product.2  For example, if a green material candidate is being evaluated for sale in the commercial market, the testing can be done to determine if the material is combustible.  Additionally, if a particular material is a combustible hazard, standardized and custom testing can be used to determine environmental conditions that could drive the green material to ignition.

Once various hazards are identified with a particular candidate green material, the risk equation can be used to evaluate the magnitude of the risk.  This is the second step of the risk analysis process and is referred to as Risk Assessment.2  Available statistics related to the incidence rate of fire in the U.S. show that fire is a rare event.  Unfortunately, the consequences associated with a fire event can be astronomical when accounting for loss of life, injuries, and potential property losses.

Once the first two steps have been adequately addressed, Risk Management can be implemented in an effort mitigate the analyzed risk.2  When taken in the context of a candidate green material for commercial sale, risk management can include the determination of a test or suite of tests that the material can undergo to determine its relative hazard in comparison with other non-green materials in the marketplace.

Very little research has been done specifically on the risk associated with green materials or even on their performance from a fire perspective; however, FM Global has produced a report that addresses the potential risk to green materials due to fire and the impact on total carbon emissions.  Unfortunately, this does not directly address issues related to safety of the occupants and first responders.

Cone Calorimeter

One of the more prevalent bench-scale test apparatus that exists for determining the ignition and flammability parameters of a particular material is the Cone Calorimeter.  A test program using the Cone Calorimeter is typically conducted using a series of samples of the investigated material (i.e., square shape with four inch sides) that are exposed to different heat fluxes in the range 1-50 kW/m2.  Figure 1 shows an overall picture of the Cone Calorimeter in a laboratory setting.

Figure 1

Figure 1 Overall View of the Cone Calorimeter Apparatus

The theory behind the testing is based on an ignition model for a liquid as adapted to a solid combustible fuel.  Ignition in solid bodies can be physically modeled as a function of ignition temperature, as described by Quintiere.[iii]  The theoretical approach rigorously applies to liquids, but can be easily extended to decomposing solids, assuming that decomposition chemical reactions do not cause any significant variation of the surface temperature.  This approximation is considered to be reasonable for most materials.3  The underlying concept is that the surface temperature is sufficient to generate enough vapors to reach the lower flammable limit (e.g., the flash point).  Achieving the lower flammable limit allows a piloted ignition of the pre-mixed flame, which leads to a sustained diffusion flame.  As an additional assumption, the surface temperature is equal to the auto-ignition temperature of the vapor mixture at the surface.  This concept may also be applied to thermally thick materials.[iv]

When conducting each test, the time to ignition is recorded for every heat flux as the time interval between initial exposure of the sample and its ignition; however, the igniter is activated almost immediately after the insertion of the sample.  The critical heat flux is then determined through 1-2 kW/m2 steps, starting from 10 kW/m2.  The threshold is given by the heat flux to which the sample is exposed and has not undergone ignition after a 30-minute exposure.  A typical curve reporting the time to ignition as a function of imposed heat flux is shown in Figure 2.

figure 2

 Figure 2 – Time to Ignition as a Function of Heat Flux for ABS Thermoplastic

 Figure 2 shows a common thermoplastic ABS (Acrylonitrile Butadiene Styrene) chosen as a representative example.  Very low times to ignition (i.e., less than one minute) have been measured for heat fluxes in the range 30-50 kW/m2.  The critical heat flux was evaluated as 9.5 kW/m2 because no ignition was observed at 9 kW/m2, while a time to ignition of 670 s was measured at 10 kW/m2.

Once a test is used to determine a measure of flammability, the thermal inertia of a material can be determined.  The last step before the ultimate evaluation of the thermal inertia is to determine the b parameter, which pertains to the relation between normalized heat flux and the square root of time to ignition.  This function is presented in Figure 3 for the considered case.

figure 3

 Figure 3 – Normalized Heat Flux as a Function of Square Root of Time to Ignition

As shown in Figure 3, the data points collapse onto a line, which leads to a straightforward evaluation of b as its slope.  The heat flux as a function of time to ignition presented in Figure 3 only serves as a way to linearize the dependence of ignition on the imposed heat flux, while the physical trend is actually shown in Figure 2.

Using the above formulation, the k(rho)c thermophysical property is calculated to be equal to 0.16 kW2 m-4 °C-2 s. Using a similar analysis and test data, thermophysical properties can be determine for any green or non-green material.

TESTING AND DATA FROM LITERATURE

Based on the definition of green materials, there are a large number of materials that are currently being used, or considered for use, in green building projects.  Some of these materials include bamboo, straw, Linoleum, sheep wool, panels made from paper flakes, seagrass, cork, coconut, lignin, and wood fiber plates.  The available test data on these materials is limited, but some data does exist in the scientific literature.

One of the byproducts of commercial paper production is lignin.  This byproduct has been deemed by industry as a natural, renewable, and biodegradable material,[vi] which fits the definition of a green material.  Due to its ability to char when exposed to heat or flame, it is mixed with other components to create hybrid materials.[vi]  In one particular study,[vi] lignin was blended with styrene–acrylonitrile–butadiene copolymer (ABS) to produce a hybrid material.  Cone calorimetry testing was performed on the resulting hybrid and was compared with standard ABS.  The results of the testing indicated that the addition of the lignin reduced the peak heat release rate of the sample and slowed the combustion process.  While these results were promising, the data indicated that the peak heat release remained at an extremely high value, 526 kW/m2, when subjected to an external heat flux of 35 kW/m2.  This provides a good example of a green material in development that attempts to reduce the flammability hazard of a building material currently in use, but must be carefully evaluated to determine the actual benefit from a fire risk perspective.

Another potential green material that has been tested in the literature is Linoleum.  Linoleum is most commonly used as a flooring material.  In a study designed to investigate the flammability response of various flooring materials,[vii] Linoleum was compared with particle board, rubber, PVC, and polypropylene using the cone calorimeter.  The results indicated that the Linoleum floor covering behaved moderately from a flammability perspective when compared to the particle board (i.e., peak heat release rate of 247 kW/m2) and rubber (i.e., peak heat release rate of 754 kW/m2).  Linoleum produced a peak heat release rate of 400 kW/m2.[vii]  These values were obtained with an external heat flux of 50 kW/m2.  From a relative analysis, Linoleum appears to be a better actor from a flammability perspective than rubber when comparing this data and a worse actor than particle board.  The data from this type of study could be used to better assess the fire spread and growth rate of Linoleum when analyzed during a post-fire investigation.

In cone calorimeter testing conducted at the University of Maryland, a number of green materials were tested and a summary of the results is presented below.  In some cases, the green materials performed well in the cone calorimeter (e.g., high critical flux for ignition values and low peak and total heat release rate values), while others performed very poorly.  This clearly shows that it is important to test individual products in order to better understand their fire growth and spread performance.

Bamboo Flooring

Bamboo flooring produced a critical flux for ignition of 13 kW/m2, a higher value than that associated with many typical wood flooring products.[viii]  Bamboo is being used in an increasing number of products because of how rapidly it can be grown, and, thus, it is deemed to be a sustainable material.  The material appears to be comparable to or to have a slightly better fire performance than other comparable products such as oak flooring.  Figure 4 depicts the data for the critical heat flux of Bamboo Flooring.

figure 4

 Figure 4 – Heat Flux versus Ignition Time for Bamboo Flooring

Foam Spray Insulation

Several different varieties of spray-in foam insulation were tested and each yielded the same basic results.  The fire retardant foam did not appear to be significantly different in behavior to the non-fire retardant varieties.  All three foams that were tested performed identically.  Figure 5 shows the data for the critical heat flux for the various spray-in foams.

figure 5

 Figure 5 – Heat Flux versus Ignition Time for Spray-in Foam Insulation

As shown in Figure 5, the foam insulation achieved ignition at very low heat incident heat fluxes (< 10 kW/m2) and did this in less than two minutes of exposure.  In comparison, cellulose insulation will tend to smolder and not actively flame and fiberglass insulation is typically noncombustible, though the paper facing may ignite.

Data gaps that current exist must be filled with a sound testing protocol that can determine the relative potential for fire growth and spread of new and current green building materials.  The above-presented results were generated using the cone calorimeter.  This test apparatus is just one of several test apparatuses or methodologies that could be considered for developing a complete understanding of the fire performance of green building materials.  Other test methodologies could include both bench-scale and full-scale testing.

COMPARISON OF SMALL-SCALE TESTING ON FULL-SCALE RESULTS

The main issue associated with full-scale testing is the cost of the test.  It is much more expensive to conduct testing on large scales for many reasons.  A larger amount of sample material is needed, an increase in testing personnel may be required, and the support facilities may be significantly larger.  These and many other factors can increase the cost substantially over small, bench-scale testing.

Additionally, a larger test does not guarantee better results.  A classic example of this is through analysis using ASTM E84 – Standard Test Method for Surface Burning Characteristics of Building Materials, which is also more commonly referred to as the Steiner Tunnel Test.  This particular test was developed prior to the common use of thermoplastics as building materials.  Samples are placed on the “ceiling” of the test apparatus and ignited from one side to observe the time of flame spread.  Between various woods and rigid combustibles, the test may provide a fair assessment of their relative flammability.  Unfortunately, the test becomes completely inappropriate for materials that deform and may melt or fall down during the test.  Depending on the material, the use of the ASTM E 84 test can be of very little value in comparing fire flammability test performance of a particular material with actual performance of the material in the field.

The known deficiencies with using the Steiner Tunnel Test have been well-documented.  Thus, other large-scale tests have been put forth as possible alternatives.  One in particular is the ISO 9705 Room Corner Test.  Fire conditions that develop in this particular test are much more in-line with the characteristics of real fires.

Multiple studies have been conducted examining a potential link between small bench-scale testing such as the cone calorimeter and large-scale tests such as the Room Corner Test.  Two studies in particular by Hansen and Hovde[ix] and Quintiere and Lian[x] provided strong analytical data and concluded that the cone calorimeter test was sufficient to predict the results of the larger-scale Room Corner Test.  Based on the known deficiencies of some large-scale tests and literature studies that suggest the small-scale results of the cone calorimeter can predict the useful results of the ISO 9705 Room Corner Test, the cone calorimeter can be considered to be a simple, small-scale test that can be adequately used to assess the flammability properties and relative fire risk of all combustible materials, including green building materials.

POTENTIAL IMPACT OF GREEN BUILDING MATERIALS

As technology changes and new building materials are introduced, it is important to be able to appropriately identify the fire risk that is associated with them.  This requires that a standard approach to risk be adopted so that the risk can be appropriately addressed.  While large-scale testing is often thought of as the best way to characterize a materials behavior in a “true” fire scenario, it may be more useful and practical to utilize a smaller-scale test method, such as the Cone Calorimeter to obtain ignition properties and heat release data.  The data derived from these small standardized tests materials can be indexed against each other and can be used to quantitatively predict large-scale performance.

The risk that green materials pose to structures and to personnel is no different than standard materials with regards to the variables of interest, namely ignition characteristics and heat release.  The main difference is that of performance.  The key variables that can be explored to quantify the risk that a material poses are:

  • Time to ignition (as a function of heat flux)
  • Thermal inertia (as derived from the cone calorimeter or other device)
  • Peak heat release rate
  • Total heat release rate
  • Products of combustion (yield and species)

Risk to Occupants

The risk to occupants posed by green materials is most clearly brought to light by documents detailing the ease of ignition, heat release rate, and toxic products of combustion.  As occupants need to be made aware as early as possible regarding a fire, any reduction in available egress time will be critical.  When looking at the difference in a green foam insulation product vs. cellulose or fiber insulation product, the differences can be examined simply using a standard t2 fire growth.  The foam insulation, even when treated with a fire retardant as shown above, will ignite readily, will tend to expand toward the fire, and, in general, can be treated as a fast or ultrafast fire, whereas the more traditional insulation materials may be non-combustible or slow fire growth.

The other major area of risk to occupants is the toxic products of combustion.  As additional plastics are used in products and building materials, the toxicity of the smoke is increasing.  Thus, in addition to the potential increase in the ease of ignition and the increased heat release rates, toxic smoke can more readily affect occupants as they attempt to egress.

Lastly, the mere presence of these products has been shown to potentially alter a firefighter’s decision making process regarding when and if they should enter a home or structure.[xi]  The concern is that if the materials are in use it may be an unacceptable risk for the firefighters to enter, and, in turn, this affects the ability of the occupant to be rescued.

Risk to First Responders

It has been well-documented[xi],[xii] in the news and in other locations that new lightweight building materials and insulation products produce an elevated level of risk to first responders.  Insulation products have created fires that have developed more rapidly, and structural flooring components have a reduced ability to withstand a fire.  This has led to an increased number of fire deaths and injuries from first responders and a national push by several groups[xiii] to further examine the use of green materials and to ensure that their response to fire is well understood.

Michigan saw two fires involving manufactured OSB I-joists that resulted in firefighter deaths within a short period of time.  These deaths occurred largely due to the failure of the flooring, which resulted in the firefighters falling through the floor a short time after the fire initiation and into an actively burning basement.11  Similar incidents have occurred in Wisconsin, Tennessee and elsewhere around the country.  In Pennsylvania, these and other incidents have caused a change in the law that essentially requires dimensional lumber to be used in new construction or requires that gypsum board be installed as a fire barrier below the OSB joists.

In full-scale testing comparing standard dimensional lumber with an engineered floor assembly, the engineered floor assembly collapsed within 15 minutes while the dimensional lumber was still structurally sound for a considerable period afterward.[xiv]

Risk to a Structure

The risk to a structure largely rests with the ease of ignition and the fire growth.  Part of making a building green and energy efficient is ensuring that there is a tight seal to minimize energy losses.  In the process, it also creates an ideal condition for containing the heat during a fire.

There have been a number of high profile fires in which the building materials played a key role, most recently the 2009 Monte Carlo fire in Las Vegas, the Borgata Water Club fire in Atlantic City in 2007, and the Mandarin Oriental Hotel fire in Beijing in 2009.  These are of particular note because each of these fires involved the use of foam materials on the exterior of the building, primarily for energy efficiency reasons.  Structural Insulated Panels (SIPs) are being used with increasing frequency and their use is increasing the fire load and combustibility of structures when compared to other materials.  In addition, lightweight concrete construction and other forms of construction are susceptible to early structural collapse during a fire.

There have also been an increasing number of fires associated with the spontaneous ignition of spray foam insulation during the installation process.  Since the curing process is exothermic and if the chemical mixture or application is incorrect, a fire can occur during or after the installation.[xv]  In this case, the product not only is a contributor to the fuel load, but can act as the ignition source and is often located in an area that is not covered by automatic fire suppression.

METHOD FOR ADDRESSING RISK

There have been several studies that have attempted to characterize the risk by using scaled data to correlate to large-scale results.[ix],[x],[xvi]  These have each offered a perspective on how this can be done.  The method outlined in Quintiere and Lian[x] shows promise as an effective way to characterize the risks associated with materials.

The method uses data derived from the cone calorimeter to predict time to flashover in the ISO 9705 compartment and has shown good success with a range of 54 different materials.  The methodology relies on four parameters:

  • Heat Release parameter, obtained from the heat release rate measurement
  • Thermal Response parameter, obtained from the slope of the time to ignition1/2 results
  • Critical Heat Flux, minimum heat flux at which ignition occurs
  • Available Energy parameter, the total energy released by the material during the test

These properties can be used to generate a curve to predict the measured time to flashover.  The empirical correlation that results from this analysis is:

equation 7

The results show that this equation can predict the phenomena of flashover fairly well.  Thus, the propensity for a material to cause flashover can be used as a quantitative means for evaluating the risk the material poses from a fire behavior standpoint.  If a material has the ability to produce flashover conditions in a shorter time period than other materials, then it can be viewed as increasing the risk.

FIRE DAMAGE PATTERNS IN GREEN MATERIALS

The push to include green materials in building construction has already resulted in an impact in fire investigation.  Knowledge of the fire risk and fire growth properties of green building materials will continue to take on importance in the field.  Due to the lightweight nature of the fuels, rapid fire spread will destroy physical evidence and can mimic the rapidity of burn that is commonly seen with liquid accelerants.

One example of this was the August 13, 2006 fire in Green Bay, Wisconsin.[xvii]  The fire began in the basement of a single family residential home.  Fire department personnel made entry into the structure and within minutes, the first floor collapsed into the basement, injuring one fire fighter and resulting in the death of another.  Post-fire analysis identified the construction of the floor slab to be lightweight pre-fabricated wooden I-beams.  The resulting damage would have been difficult to account for if fire department personnel had not been on the scene early enough, and the fire had been allowed to continue to burn without intervention.  Due to the rapidly burning nature of the lightweight pre-fabricated joists, entire portions of a residential structure could be lost during a fire.  This large extent of damage can make origin determination difficult and could result in the misinterpretation of the damage patterns.  Consideration must be given to the materials involved at a fire scene to ensure proper interpretation of the fire growth and spread.

SUMMARY

Green materials are being introduced into the marketplace at a rapid pace, often with minimal fire testing.  When tests are required, they are often poor predictors of how the materials will respond under real fire conditions or at a larger-scale.  As buildings become more elaborate and move toward increasing sustainability through the use of green materials, the materials utilized may present an increased fire risk.  There needs to be a methodology that can be used to evaluate and compare the potential risk associated with a particular green material.  The cone calorimeter is a simple standard test that allows for relevant material properties and material performance data to be obtained.  This data pertains to both the potential consequences of a fire as well as the probability of one occurring or spreading through the ignition delay curve.  Quintiere and Lian[x] and others have provided several good attempts at using the cone data to predict full-scale fire behavior.  The simple formulation can be used to compare the relative performance of materials and, indirectly, the risk associated with them.  This could be used as a means of regulating what the minimum performance characteristics of a material need to be in order to be acceptable.  In addition, it provides flexibility as the measure is not based on a single material property but on a range of material parameters.  Finally, understanding and quantifying the risk associated with green materials will allow for better understanding of the resulting damage patterns at a fire scene when green materials have been involved.

[i] “Green Building,” http://www.epa.gov/greenbuilding/pubs/about.htm. [Online]. [Accessed: 22-Dec.-2011].

[ii] N. Hyatt, Guidelines for process hazards analysis, hazards identification & risk analysis. CRC Press,  p. 474., (2003).

[iii] J. Quintiere, “A theoretical basis for flammability properties,” Fire and Materials, (2006).

[iv] J. Troitzsch, International plastics flammability handbook, p. 500, (1983).

[v] ASTM 1321 Standard Test Method for Determining Material Ignition and Flame Spread Properties.

[vi] P. Song, Z. Cao, S. Fu, Z. Fang, Q. Wu, and J. Ye, “Thermal degradation and flame retardancy properties of ABS/lignin: Effects of lignin content and reactive compatibilization,” Thermochimica Acta, vol. 518, no. 1, pp. 59–65, (2011).

[vii] P. Johansson and J. Axelsson, “The influence of floor materials in room fires,” Brandforsk Project 300-061.

[viii] M. Spearpoint, “Predicting the piloted ignition of wood in the cone calorimeter using an integral model — effect of species, grain orientation and heat flux,” Fire Safety Journal, (2001).

[ix] A. S. Hansen and P. J. Hovde, “Prediction of time to flashover in the ISO 9705 room corner test based on cone calorimeter test results,” Fire and Materials, vol. 26, no. 2, pp. 77–86, (2002).

[x] J. Quintiere and D. Lian, “Inherent flammability parameters—Room corner test application,” Fire and Materials, vol. 33, pp. 377–393, (2009).

[xi] “Lightweight Building Materials Cause Greater Fire Risk – Milwaukee News Story – WISN Milwaukee.”

[xii] “BuildingGreen.com LIVE: Massachusetts Fires Tied to Spray Foam Incite Debate by Tristan Roberts on 07/14/2011,” buildinggreen.com. [Online]. Available: http://www.buildinggreen.com/live/index.cfm/2011/7/14/Massachusetts-Fires-Tied-to-Spray-Foam-Incite-Debate. [Accessed: 23-Dec.-2011].

[xiii] J. Tidwell and J. Murphy, “Bridging the Gap: Fire Safety and Green Buildings,” firemarshals.org. [Online]. Available: http://www.firemarshals.org/pdf/NASFM_greenfire_guide.pdf. [Accessed: 23-Dec.-2011].

[xiv] “Common Building Material Poses Deadly Threat To Firefighters – Milwaukee News Story – WISN Milwaukee,” wisn.com. [Online]. Available: http://www.wisn.com/news/17968429/detail.html. [Accessed: 23-Dec.-2011].

[xv] “BuildingGreen.com LIVE: Fire Risks Not Limited to Spray-Foam Insulation by Tristan Roberts on 11/01/2011,” buildinggreen.com. [Online]. Available: http://www.buildinggreen.com/live/index.cfm/2011/11/1/FirefighterBuilder-FIre-Risks-Not-Limited-to-SprayFoam-Insulation. [Accessed: 23-Dec.-2011].

[xvi] R. Petrella, “The Assessment of Full-Scale Fire Hazards from Cone Calorimeter Data,” Journal of Fire Sciences, (1994).

[xvii] http://www.nfpa.org/newsandpublications/nfpa-journal/2009/july-august-2009/features/lightweight-construction [Accessed: 01-August-2014].

Advertisements