Experimental of Fire Behavior in Compartment - Case Study Example

Student Name: Tutor: Title: The Experimental Study of Fire Behavior in a Compartment Date: ©2016 Table of Contents 1.Literature Review 3 1.1Overview 3 1.2Background Information 4 1.2.1The Burning of Thermoplastics 4 1.2.2ISO 5660: Cone Calorimetry 5 1.3Definition of Parameters 8 1.3.1Rate of Heat Release 8 1.3.2Mass Loss Rate, 8 1.4Previous Studies 9 1.4.1Compartment Fire Experiments 9 1.4.2Models and Relationships developed 14 1.5Conclusion 14 1.6References 15 The Experimental Study of Fire Behavior in a Compartment 1. Literature Review 1.1 Overview A lot of literature has been presented regarding the behavioral combustion of polymers in enclosed compartments. Both natural and synthetic polymers have been studied and their flame performance published. In most of these experimental studies, cone calorimetry (ISO 5660) has been the technique of choice as it offers a realistic model for the measuring of the rate of mass loss, rate of heat release and effective heat release among other parameters of a sample under study(Quintiere, 1998). The data obtained by the researchers of those studies form the basis of understanding the features of compartment fires. Furthermore, analytical models explaining crucial burning parameters could be developed (Utiskul, 2006). The literature reviewed in this section of the dissertation focuses primarily on the rate of mass loss, total heat release and the rate of heat release for different colors as it may not be possible to reference all the studies relating to the burning parameters of plastic. The colors of plastic studied include black, green, red and white whereby two samples for each colour are used. The exposure conditions for each sample are 25kW/m2 and 55kW/m2 respectively. The paper aims to develop a relationship between the aforementioned exposure conditions and the parameters under study. 1.2 Background Information Zhang et. al (2005), asserts that the concern on the hazardous potential of polymeric materials on fire has increased in the past decade due to the increasing number of plastic materials being used in populations’ daily lives. Sure enough, the burning of fires involving synthetic plastics involve specific behaviors with the melting behavior being the most featured burning behavior for a wide variety of polymers. This occurrence is commonly acknowledged by the fire safety community (Beyler, 1988) but despite this recognition, a few attempts of a detailed and in depth study of specific behaviors have been made. 1.2.1 The Burning of Thermoplastics Analytical and theoretical plastics’ burning models that are currently in place (Quintiere, 1994), (Stelker et. al, 1991), (Moghtaderi et. al, 1997) and (Staggs & Whiteley, 1999) tend to ignore the effects of melting behavior of plastics or fail to illustrate it completely. Zhang et al (1996) proved that there is significant relationship between the melting behavior of thermoplastics and their burning rate. They achieved this by comparing the melting behaviors of plastic polymers in both small scale cone calorimeter tests and large scale systems. Thermal aspects and the smoke emissivity of the samples under study can be assessed using a cone calorimeter. The outcomes of the assessments of the rates of heat release of the plastic materials would be useful in in determining the rates heat release in real live set-ups such as a department store housing plastic containers. There are several theorems existing in literature that can be applied in the aforementioned analysis. For instance, in the study of furniture fires, the convolution theorem (Thomas &Heselden, 1972) would be applicable through the use of the heat release rate per unit area curves determined using a cone calorimeter. The results from the study would be helpful in developing systems such as fire and smoke detection systems through smoke and gas analysis (Chow & Wong, 1994). Such measurements could also be useful in the estimation of the toxicity index of smoke in actual fires and fire tragedy investigations as shown in Babrauskas (2000). Once the samples are burnt, the hazardous species among the smoke constituents can be isolated, quantified and studied further. However, this technique requires advanced measuring equipment such as a FTIR equipment (Bulien, 1996). Chow (1998), observed that there is a significant difference between fire fuels after flashover fires and before. For this reason, there has been rising interests in the study of the materials of such fire fuels with thermoplastics being one of such. It is important to note that samples tested under conditions of high heat fluxes, in this case 55kW/m2, would portray an image of how actual fire hazards take place. It is therefore an important fire assessment tool. For instance, 20 to 25 kW/m2 may be taken as the exposure condition for a room’s floor, 30 to 35kW/m2 for the normal wall materials and 45 to 50 kW/m2 for ceiling materials (Duggan, 1997). Furthermore, overall fire hazard toxicity can be can be determined through the use of measured smoke data (Babrauskas, 1991) especially the burning behavior of plastic products that make up quite a lot of household goods. Such materials ought to be assessed through cone calorimetry technique. In the experimental study of compartment fires on non-standard cases (Lindholm et al, 2012), keen attention needs to paid to the inherent relationship between the heat release rate and rate of combustion i.e. total heat release. In such cases, the surfaces of the samples in question do change as a result of melting during the burning process. Furthermore, the tendency of the plastic samples tested to ignite, ought to be complimented by heat release rate and oxygen consumption vs time records of cone calorimeter tests under a radiant exposure of 25kW/m2 (Rychlý et al, 2014). 1.2.2 ISO 5660: Cone Calorimetry 1.2.2.1 Device Description It refers to a reliable technique for the measurement of data relating to sample surface ignition and penetrative burning tests. As opposed to being one bulk device, a cone calorimeter is made up of crucial components that work together to measure and analyze burning parameters under study (Lindholm et al. 2012). As stated in the previous sections, a cone calorimeter test is the ideal small-scale test for the measurement of heat release rate and the total heat release from a burning polymer sample under controlled heat source (ISO 5660 part 1). Being a primary cause of growth and fire spread, the rate of heat release (HRR), its sub-parameters including the magnitude and timing of the peak rate of heat release (PRHR) and the average short term rates of heat release are the single most important factor in predicting fire growth rate. The cone is established as the primary technique for the determination of parameters relating to fire and flammability under ventilated fires. Part 2 of ISO 5660 provides that the cone calorimeter device can also be used in the determination of smoke generation. A cone calorimeter is applicable in the measurement of the following parameters as per the provisions of ISO 5660; i. heat release rate (kW/m2), ii. total heat release (MJ/m2), iii. rate of mass loss (g/s), e iv. effective net heat of combustion (MJ/kg) v. smoke production rate (m2/s). As mentioned in the previous section, FTIR analysis can also help to analyse the levels of toxic components in smoke. For a complete test series involving 3 tests, five samples of 100 by 100 mm and maximum width of 50mm get tested. However, this dissertation is concerned with single test series, as a result, two samples of 100mm by 100mm each and a thickness of 50mm are tested for each colour of plastic studied. Figure 1.1: Components of a cone calorimeter device. 1.2.2.2 Test Procedure. The tests performed herein must comply with the provisions of ISO 5660. A sample of the product being tested measuring 100 mm by 100 mm, gets subjected to particular radiance level. As the surface of the sample gets heated, the sample starts to produce pyrolysis gases that are then ignited by a spark igniter as shown on figure 1.1 above. The exhaust hood then collects the gasses emitted and are then channelled via the ventilation system whereby, gas analysis, smoke measurement and flow measurement are simultaneously conducted. From the gas analysis performed, further parameters including heat release can be derived. Data on oxygen concentration in the emitted smoke is used in this derivation. A laser system installed within the device enables the measurement of smoke throughout the entire process. Furthermore, the data collected herein can be used to model actual fires and perform the basic pass or fail tests. 1.3 Definition of Parameters 1.3.1 Rate of Heat Release The heat release rate of a material is a vital fire characterization tool (Biteu et al, 2008) It refers the product of a material’s rate of mass loss and the heat of combustion. Most techniques that have been advanced to estimate it are based on mass balances. In the case of his experimental study, calorimetric technique is used to assess the heat release rate of plastic samples of the various colors chosen. Below are illustrations of possible relationships of this parameter with time; Figure 1.2: Example of HRR vs Time for a sample radiated at 50kw/m2 1.3.2 Mass Loss Rate, As a material burns up and pyrolysis gases get released, there is an apparent loss in mass. The rate of change in the materials mass is therefore referred to as mass loss rate and is important in the study of fire behavior as it proportional to the amount of energy released by material 1.4 Previous Studies 1.4.1 Compartment Fire Experiments Ingburg (1928) was among the earliest researchers to try and explain fire behavior. Since then, a wide variety of experimental data has been obtained for partially and fully developed compartment fires. His paper focused on the severity of building fires and in doing so, he dealt with the behavior of compartment fires. However it was not until Kawagoe (1958, 1963)’s researches that aimed at developing a relationship between the burning rate and the ventilation parameter that the true potential of the cone calorimeter in the study of compartment fires got understood. Furthermore, Gross (1965) studied the burning behavior of fiberboard crib under three varying enclosure sizes and determined that the mass loss rate is proportional to the ventilation parameter expressed as . He however experienced data shifts due to various scales, a fact which necessitated the used of scaling factors. Individual factors represented a ratio of the samples linear dimensions to the respective enclosure size. This then enabled him to normalize the rate of burning (and effectively the total heat release) and the ventilation parameters. The rate of mass loss parameter was however deeply dealt with by Thomas and Heselden (1970). They are responsible for the presentation of the results and analysis of the commonly known CIB test which was a joint test program on fully developed single compartment fires carried out in 1958 by eight registered laboratories. The CIB test explored the burning behavior of wood cribs inside different compartment shapes. The results of which were used to develop and introduce several empirical and analytical relationships between radiance intensity, average rate of burning, smoke and gas temperatures and the respective fuel and compartment characteristics. Not all of the relationships however were empirical in the realistic sense as some got developed through heat balance considerations. The use of small scale testing compartments such as a cone calorimeter was justified by the fact that the researchers identified minor scale effects on the tests. Heselden et al, (1970) then empirically presented the mass loss rate parameter as on average amounting to 5.5 kg/min-m5/2. For the various ventilation controlled experiments, Heselden et al. (1790) then proposed the empirical equation below; (1.1) The value of the constant in 1.1 above was determined to be roughly. It is important to note that that the value of the constant obtained by Heselden et al. (1970) matched the works of Kawagoe and Sekine (1963). In both tests, the exposure condition for the above value to be determined was maintained to at least 150 kg of fuel per square meter of open ventilation. Both papers found a correlation between the value of constant k and the geometry of the enclosure. An experimental study on behavior of compartment fires using pool fire and crib fire (including plastics) was published by Tewarson (1972). He divided the compartment into four zones based on the zone temperature, rate of combustion and ventilation characteristics. He then developed empirical expression for the determination of toxic gas concentrations, a process that required regional interfaces among boundary layer zones. Harmathy (1972) intended to present a new dimension in compartment fires and thus extensive reviewed wood crib compartment fires. According to his proposition, the school of thought that air shortage in a compartment limits the burning rate was somehow unrealistic (Harmathy, 1978). He based his revolutionary new idea on several experimental studies and presented the logic behind it as follows. First, he presented a new parameter of ventilation empirically represented by and normalized by the area of burning fuel. This represented the rate of air flow per fuel available. The analysis of various forms and range of data from experimental studies helped Harmathy to show that the burning of samples under study, explicitly depend on the size and geometry the compartments. However this preposition, would only be applicable in regimes whose ventilations are controlled. He then clearly defined the value of the empirical formula relative to a specific ventilation parameter. He also developed a linear relationship between the specific ventilation parameter and the rate of burning. For each sample. In his subsequent studies, Harmathy also determined the ventilation level effects on uncharred fuel material. He also showed that the rate of burning of charred material increase with rate of air flow to a maximum value. Bullen and Thomas (1920) furthered Harmathy’s research and clearly distinguished between charred and uncharred materials. They achieved this by use of the expression; (1.2) Other studies have tried to explain the relationship between ventilation parameters and heat radiance required (Santos and Delichatsios, 1982). The results of which have showed a relationship between surface flame temperature of heat source and the tendency of soot formation. Tewarson et al. (1981) were also in agreement with Santos and Delichatsios in that they determined that the flame radiative heat flux increases with the concentration of oxygen for plastics as well as liquid fueled fires. However, few studies have tried to analyze the effect forced ventilation has on the behavior of compartment fires. Peatross and Beyler (1997) is among the few full scale studies on this phenomenon. While using a range of fuels including ordinary diesel, wood, and slab of polyurethane, they found the mix conditions for ventilations tests that they carried out. The fuel mass loss rate reduced with decrease in the concentration of oxygen, a phenomenon that enabled them to propose a direct proportionality between the two parameters. The only limitation to this study was the non-inclusion of thermal enhancements. The effect studied is subject to overstatement as a result of the presence of scaling effects on the emissivity of the flames. Large scale liquid fueled experiments were conducted by Fleischmann and Parkes (1997) using samples as large as 200mm. This however required thermal enhancers and the study was such that ventilation could be varied as the experiment progressed. They determined that the mass loss rate was nearly seven times that in non-enhanced burning. Results of this research were then compared with other studies and used to develop a behavior prediction model of compartment fires. Other studies have focused on the effect of material surface area, type of fuel and shape of compartment on the rate of mass loss. The analysis of the Ohmiya et al. (1998) study by Delichatsios et al. in 2004 aimed at illustrating the aforementioned relationship. It was also a regime whereby ventilation was controlled and they found that there was a close dependence on the air inflow rate by the rate of mass loss that changed to a stratified from its uniform sense. A significant finding was that the exposed fuel area is not fully involved in the burning process. This study however did not explain why mass loss rate was dependent on the rate of air inflow during ventilation. Kumar et al. (2006) studied the gas temperature effect on mass loss rate by means of cross ventilation. The results of the study showed that the temperatures were higher in the cross ventilated experiment compared to the ordinary singly ventilated when dealing with huge fires. This phenomenon could easily be applied in the assessment of fire hazards such as forest fires where the level oxygen concentration and fuel availability is quite high. Academics at the University of Maryland, recently carried out large scale experiments using 400mm cubical compartments fueled by heptane in different pan diameters. This was a joint effort by different individuals dealing with respective burning parameters On that regard, Wakatsuki (2001) was responsible for analyzing single ceiling vents, while Rangwala (2001) examined a single zone model of the wall vent case. Furthermore, Ringwelski (2001) dealt with sample areas that were equal in area both at the bottom and top of the aforementioned walls. The ventilation effects at the region of low air supply were dealt with by Utiskul et al (2005) to characterize and explain the low ventilation phenomenon studied by Utiskul, Hu et al(2006) applied the FDS technique. Other researches such as Williamson et al. (2006)further explains the FDS phenomenon with regard to flame dynamics in the determination og grid cells. These studies and others clearly show that the field of compartment fire behavior is wide and largely unexplored. It is therefore the aim of this dissertation that studying plastics as the form of crib material would enhance these studies and perhaps come up with new hazard models. 1.4.2 Models and Relationships developed Temperature vs time graphs developed by Kawagoe and Sekine (1963) were developed via the integration of compartment energy balances and the time parameter. The technique applied in this case got limited to ventilation controlled regimes having a uniform rate of heat release. In support of this proposition, (Lie, 1995) advanced an expression that was parameter based. (McCaffrey et al, 1981) also proposed the MQH correlations based on compartment energy balances. Assumptions had to be incorporated though in the hope of determining energy losses to the compartment walls and flow vents (Friedman, 1992) presented a detailed report for computer developed fire models. His work was recently reviewed by Olenick and Carpenter (2003). Friedman emphasized the need for thermal feedback for a more realistic fire model including responses to vitiated oxygen. 1.5 Conclusion The fact that most of the current consumer products are dependent on plastic for packaging and transport, there is need to develop efficient and safe plastics without compromising on the consumption levels of consumers. It is therefore important to understand the burning behavior of the plastics involved so as to develop means of fore retardation. This will enable manufacturers to develop materials meeting safety specifications where fire retarders are incorporated into polymers. 1.6 References Babrauskas, V., 1982. Will the second item ignite?. Fire Safety Journal, 4(4), pp.281-292. Babrauskas, V., 1984. Development of the cone calorimeter—a bench‐scale heat release rate apparatus based on oxygen consumption. Fire and Materials, 8(2), pp.81-95. Beyler, C., 1988 “Thermal Decomposition of Polymers,” The SFPE Handbook of Fire Protection Engineering (1st ed.), DiNenno, P. J. (ed.), National Fire Protection Association, Quincy, MA 02269, pp. 1/173-176. Bullen, M.L. and Thomas, P.H., 1979, January. Compartment fires with non-cellulosic fuels. In Symposium (International) on Combustion (Vol. 17, No. 1, pp. 1139-1148). Elsevier. Chow, W.K., 1996. Simulation of tunnel fires using a zone model. Tunnelling and Underground Space Technology, 11(2), pp.221-236. Delichatsios, M.A., Silcock, G.W., Liu, X., Delichatsios, M. and Lee, Y.P., 2004. Mass pyrolysis rates and excess pyrolysate in fully developed enclosure fires. Fire Safety Journal, 39(1), pp.1-21. Fleischmann, C.M. and Parkes, A.R., 1997. Effects of ventilation on the compartment enhanced mass loss rate. Fire Safety Science, 5, pp.415-426. Harmathy, T.Z., 1972. A new look at compartment fires, Part I. Fire technology, 8(3), pp.196-217. Ingberg, S.H., 1928. Tests of the severity of building fires. NFPA Quarterly, 22(1), pp.43-61. Kawagoe, K., 1958. Fire behavior in rooms. Building Research Institute, Ministry of Construction (Japan) (No. 27). Report. Kumar, R. and Naveen, M., 2007. Compartment fires: CALTREE and cross-ventilation. Combustion Science and Technology, 179(8), pp.1549-1567. Quintiere, J.G., 1981. An approach to modeling wall fire spread in a room. Fire Safety Journal, 3(4), pp.201-214. Quintiere, J.G., 1998. Principles of Fire Behavior. New York: Delmar publishers. Tewarson, A., 1972. Some observations on experimental fires in enclosures. Part I: Cellulosic materials. Combustion and Flame, 19(1), pp.101-111. Thomas, P. and Heslden, J., 1972. Fully Developed Fires in Single Compartments, A Co-operative Research Program of the Conseil International du Batimen (CIB Report No. 20) (No. 923). Fire Research Station Note. Thomas, P.H. and Heselden, A.J.M., 1972. Fully-developed firess in single compartments: A co-operative research programme of the Conseil International du Bâtiment (CIB report no. 20) Fire Research Station Fire Research Note 923. Building Research Establishment, Borehamwood. Utiskul, Y., 2006. Theoretical and experimental study on fully-developed compartment fires. Read More

The results from the study would be helpful in developing systems such as fire and smoke detection systems through smoke and gas analysis (Chow & Wong, 1994). Such measurements could also be useful in the estimation of the toxicity index of smoke in actual fires and fire tragedy investigations as shown in Babrauskas (2000). Once the samples are burnt, the hazardous species among the smoke constituents can be isolated, quantified and studied further. However, this technique requires advanced measuring equipment such as a FTIR equipment (Bulien, 1996).

Chow (1998), observed that there is a significant difference between fire fuels after flashover fires and before. For this reason, there has been rising interests in the study of the materials of such fire fuels with thermoplastics being one of such. It is important to note that samples tested under conditions of high heat fluxes, in this case 55kW/m2, would portray an image of how actual fire hazards take place. It is therefore an important fire assessment tool. For instance, 20 to 25 kW/m2 may be taken as the exposure condition for a room’s floor, 30 to 35kW/m2 for the normal wall materials and 45 to 50 kW/m2 for ceiling materials (Duggan, 1997).

Furthermore, overall fire hazard toxicity can be can be determined through the use of measured smoke data (Babrauskas, 1991) especially the burning behavior of plastic products that make up quite a lot of household goods. Such materials ought to be assessed through cone calorimetry technique. In the experimental study of compartment fires on non-standard cases (Lindholm et al, 2012), keen attention needs to paid to the inherent relationship between the heat release rate and rate of combustion i.e. total heat release.

In such cases, the surfaces of the samples in question do change as a result of melting during the burning process. Furthermore, the tendency of the plastic samples tested to ignite, ought to be complimented by heat release rate and oxygen consumption vs time records of cone calorimeter tests under a radiant exposure of 25kW/m2 (Rychlý et al, 2014). 1.2.2 ISO 5660: Cone Calorimetry 1.2.2.1 Device Description It refers to a reliable technique for the measurement of data relating to sample surface ignition and penetrative burning tests.

As opposed to being one bulk device, a cone calorimeter is made up of crucial components that work together to measure and analyze burning parameters under study (Lindholm et al. 2012). As stated in the previous sections, a cone calorimeter test is the ideal small-scale test for the measurement of heat release rate and the total heat release from a burning polymer sample under controlled heat source (ISO 5660 part 1). Being a primary cause of growth and fire spread, the rate of heat release (HRR), its sub-parameters including the magnitude and timing of the peak rate of heat release (PRHR) and the average short term rates of heat release are the single most important factor in predicting fire growth rate.

The cone is established as the primary technique for the determination of parameters relating to fire and flammability under ventilated fires. Part 2 of ISO 5660 provides that the cone calorimeter device can also be used in the determination of smoke generation. A cone calorimeter is applicable in the measurement of the following parameters as per the provisions of ISO 5660; i. heat release rate (kW/m2), ii. total heat release (MJ/m2), iii. rate of mass loss (g/s), e iv. effective net heat of combustion (MJ/kg) v.

smoke production rate (m2/s). As mentioned in the previous section, FTIR analysis can also help to analyse the levels of toxic components in smoke. For a complete test series involving 3 tests, five samples of 100 by 100 mm and maximum width of 50mm get tested. However, this dissertation is concerned with single test series, as a result, two samples of 100mm by 100mm each and a thickness of 50mm are tested for each colour of plastic studied. Figure 1.1: Components of a cone calorimeter device. 1.2.2.

Read More