Search

Oriented strand board is a widely used construction material responsible for a substantial portion of the fire load of many buildings. To accurately model oriented strand board fire response, kinetics and thermodynamics of its thermal decomposition and combustion were carefully characterized using milligram-scale testing in part I of this study. In the current work, Controlled Atmosphere Pyrolysis Apparatus II tests were performed on representative gram-sized oriented strand board samples at a range of radiant heat fluxes. An automated inverse analysis of the sample temperature data obtained in these tests was employed to determine the thermal conductivities of the undecomposed oriented strand board and condensed-phase products of its decomposition. A complete pyrolysis model was formulated for this material and used to predict the mass loss rates measured in the Controlled Atmosphere Pyrolysis Apparatus II experiments. These mass loss rate profiles were predicted well with the exception of the second mass loss rate peak observed at 65 kW m−2 of radiant heat flux, which was underpredicted. To further validate the model, cone calorimeter tests were performed on oriented strand board at 25 and 50 kW m−2 of radiant heat flux. The results of these tests, including both mass loss rate and heat release rate profiles, were predicted reasonably well by the model.

The effectiveness of suppression of lithium ion battery, LIB, fires with water mist was investigated experimentally using a previously developed bench-scale wind tunnel. The experiments were conducted on lithium ion cell arrays constructed from twelve cylindrical (18,650 form factor), fully charged, lithium cobalt oxide cells, which were densely packed in a rectangular configuration without gaps. An electric heater was employed to initiate thermal runaway in one cell. Failure propagation was then tracked using thermocouples located at the bottom surfaces of the cells. Heat release due to flaming combustion was computed based on the oxygen consumption method. Experiments conducted at 640 l min−1 of air, 320 l min−1 of air, and 186 l min−1 of nitrogen tunnel purge flow were used as reference points. All reference point experiments underwent complete cascading failure. Addition of 1.0 and 1.6 g s−1 of water mist to 320 and 640 l min−1 of air produced agent concentrations of 14.1 and 11.1 wt%, respectively, which are slightly above the concentration of water mist recommended for suppression of traditional fires. Application of water mist at 1.0 and 1.6 g s−1 prevented cascading failure in 40% and 50% of the tests, respectively, and significantly reduced the rate of failure propagation in the other tests. At both water mist delivery rates, flaming combustion associated with burning of battery ejecta was inhibited, reducing the combustion efficiency below 50%. One of the key findings of this study is that suppression of flaming combustion is not sufficient to stop cascading failure. The array must be continually cooled with water mist until the temperature of the cells decreases below a certain threshold, which prevents chemical reactions between battery materials inside the cell casings.

To characterize the UL-94 vertical burning test (UL-94 V), temperature, oxygen content, and heat flux of the burner flame were measured using an R-type thermocouple, oxygen analyzer, and water-cooled Schmidt–Boelter gage, respectively. To account for sample geometry, mock samples from Kaowool PM insulation board were used during the burner measurements. Temperature and heat flux measurements were used to express the heat flux of the burner as a linear, piecewise function of burner flame temperature and a constant convective heat transfer coefficient. UL-94 V tests were conducted using 0.27 cm thick poly(methyl methacrylate) (PMMA) samples with and without insulated sides to investigate edge effects. All UL-94 V tests were recorded on video using a 900-nm narrow-band filter to focus on emissions from soot for tracking flame length over time. In some of these tests, a heat flux gage was embedded into the sample to measure PMMA flame heat flux to verify a previously developed laminar wall flame heat feedback submodel. PMMA flame heat fluxes of insulated samples confirmed the wall flame submodel, while non-insulated samples had significantly greater heat fluxes; the flame submodel was scaled accordingly. The burner and wall flame heat flux submodels were combined with an independently developed pyrolysis model of PMMA to conduct two-dimensional, time-resolved simulations of the UL-94 V test. Burner flame oxygen content was measured to be about 5 vol% and was found to enhance PMMA decomposition. Inclusion of the oxygen effect was essential to predict ignition. The simulation predicted flame length and spread on insulated UL-94 V samples reasonably well, validating the submodels. However, the simulation underpredicted flame length on the non-insulated samples; discrepancies were attributed to burning and spread on the edges which were not modeled explicitly. The developed model can be applied to non-charring polymeric solids with no to moderate amount of dripping. Its applicability for charring and intumescent materials will be explored in the future.

A quantitative understanding of the thermal transport within a growing intumescent char layer remains largely unsolved. The development of improved techniques to analyze charring and intumescent materials is necessary to investigate the physics of heat transfer within the char. To aid in this endeavor, a systematic methodology to parameterize comprehensive pyrolysis models for charring and intumescent materials is presented. Thermogravimetric analysis, differential scanning calorimetry and microscale combustion calorimetry were conducted on 4–7 mg samples to analyze the kinetics and thermodynamics of thermal decomposition and determine heats of complete combustion of gaseous pyrolyzates. A multi-step reaction mechanism, consisting of sequential first-order reactions, was constructed to capture the physical transformations and chemical reactions observed in the milligram-scale experiments. 0.07 m diameter disk-shaped samples were gasified in the Controlled Atmosphere Pyrolysis Apparatus II to characterize the thermal transport within the condensed phase material and evolving char layer. ThermaKin2Ds was employed to interpret the experimental data using inverse analysis techniques. Bisphenol A polycarbonate and poly(ether ether ketone), widely used intumescent materials, were analyzed within this study. The resulting two-dimensional models of bisphenol A polycarbonate and poly(ether ether ketone) were capable of predicting the experimental gasification mass loss rates with a mean error of 17.8% and 16.9%, respectively. An analysis of the char pore structure was also conducted from which quantitative relationships were subsequently developed between relevant thermal transport quantities and physical descriptors of the char's physical structure. Two prominent linear correlations were discovered: the char's effective thermal conductivity increased as a function of increasing volume-weighted mean pore diameter and the product of density and thermal conductivity increased with an increasing char porosity based on image analysis. Finally, a brief sensitivity analysis of the polycarbonate and poly(ether ether ketone) burning rates was conducted to determine which key material properties were responsible for the observed differences in flammability.

Lithium-ion batteries must undergo a series of quality control tests before being approved for sale. In this study, quality control tests were carried out on two types of lithium-ion pouch batteries, here denoted as type A (with stacked electrode configuration) and type B (with a jelly-roll arrangement) to assess the effectiveness of the tests. Electrochemical tests, which included capacity and impedance measurements, found that both types of batteries met the specifications. However, computed tomography (CT) scan, disassembly, and material characterization revealed quality concerns in battery assembly and material composition. Results showed that, for an A cell, cathode extended past anode at the top and bottom of the roll, and a CT scan revealed inhomogeneities in the electrode near the corners. Similarly, analysis of a B cell revealed gaps in the winding structure and cathode material discrepancies. More specifically, the lithium nickel manganese cobalt oxide (NMC) material specified by the battery manufacturer turned out to be lithium cobalt oxide (LCO). The findings indicate that systematic quality control tests are needed to properly identify defects in batteries before they are used in products., Center for Advanced Life Cycle Engineering (CALCE); Centre for Advances in Reliability and Safety (CAiRS)

Lithium ion batteries (LIBs) are among the most promising technologies for electrical energy storage. However, any exposure of LIBs to abnormal operating conditions may lead to rapid self-heating accompanied by ejection of flammable materials, termed thermal runaway. In a multi-cell array or pack, thermal runaway may propagate to neighboring cells and grow into a large-scale fire. In this work, a new experimental setup was developed to investigate this propagation or cascading failure phenomenon using 12–18 cell arrays constructed from lithium cobalt oxide (LCO) cells of 18650 form factor and 2600 mA h nominal electrical capacity. The arrays, consisting of fully charged cells, were mounted in a specially designed wind tunnel, which provided well controlled environmental conditions. Thermal runaway was initiated in one cell using a small electric heater and observed to propagate through the array using temperature sensors attached to individual cells. The propagation process was studied in both nitrogen and air environments to elucidate the impact of flaming combustion. In addition to the cell temperatures, production rates of O2, total hydrocarbons (THC), CO, CO2 and H2 were measured, and heats generated in chemical reactions between the battery materials and in flaming combustion were computed. In the nitrogen tests, row-to-row propagation speed (SP) showed no significant dependence on the size of the array and was estimated to be 0.08 s−1. When cell arrays were tested in air, SP increased to 0.7 s−1 (about 9 times the nitrogen value) in late stages of cascading failure due to the impact of flaming combustion of ejected materials. Measurements demonstrated that failing LCO cells produced minor mass yields of O2 and H2 in addition to relatively large amounts of CO, THC and CO2. The lower flammability limit of the ignitable portion of ejected gaseous products was determined to be 5.79 ± 0.12 vol.% in air. The chemical heat generation resulting from reactions between battery materials inside and outside the bodies of cells was computed to be 56.6 ± 2.5 kJ per cell. The total amount of heat released from flaming combustion of ejected battery materials during air tests was found to be 60.1 ± 17.5 kJ per cell. The efficiency with which these battery materials were combusted was estimated to be about 56%. The results of this study provide previously unavailable, comprehensive assessment of the failure dynamics and energetics in LIB cell arrays or assemblies. These results are expected to serve as a foundation for effective methodologies for detection, mitigation and prevention of electrical energy storage and electric vehicle fires.

Lithium ion batteries (LIBs) are booming due to their high energy density, low maintenance, low self-discharge, quick charging and longevity advantages. However, the thermal stability of LIBs is relatively poor and their failure may cause fire and, under certain circumstances, explosion. The fire risk hinders the large scale application of LIBs in electric vehicles and energy storage systems. This manuscript provides a comprehensive review of the thermal runaway phenomenon and related fire dynamics in singe LIB cells as well as in multi-cell battery packs. Potential fire prevention measures are also discussed. Mitigating the hazards associated with a growing number of LIB applications represents a significant new challenge for the fire safety engineering community. Some perspectives and outlooks on the future of LIB fire safety research and safety engineering are given.

This work introduces a new method for determination of spatially resolved radiation intensity and radiative fraction for axisymmetric laminar diffusion flames by using a modified DSL-R camera employed to collect monochromatic, 900 nm, images and a Schmidt-Boelter heat flux gauge. A novel instrument, Milligram-scale Flame Calorimeter, is employed to supply the flame with either gaseous fuel or solid fuel pyrolyzate and measure the total heat release rate associated with the combustion process. The high spatial resolution provided by the images of the camera allows for a multi-emitter treatment of the 20–60 mm tall flames. The flames’ shapes and monochromatic radiation intensity profiles are extracted from the images. Using the radiation transport equations formulated for the system’s specific geometry, the monochromatic radiation intensities are related to the total heat flux gauge readings collected at several distances from the base of the flame. Subsequently, spatially resolved total radiation intensity data are obtained. Integration of this intensity over the flame surface yields the radiation heat release rate, of which the ratio to the total heat release rate yields the radiative fraction. In this work, methane, propane and acetylene flames have been studied to validate this methodology (covering a wide range of sooting conditions), while polyethylene-fueled flames above and below the smoke point have been analyzed to demonstrate the method’s applicability to solid fuels. The results show that the method produces values that are consistent with previous studies, particularly when the assumptions of the previous studies are replicated. Additionally, the data indicate that for flame heights that are all below (or all above) the smoke point, the value of the maximum total radiation intensity does not depend on the total heat release rate. This behavior of the maximum radiation intensity suggests that said parameter is a good candidate to correlate with the radiative properties of turbulent flames.

Lithium ion batteries are increasingly used in electrical energy storage systems despite their thermal and chemical hazards on failure. These hazards are magnified when multiple cells are combined together in a pack because of the risk of cascading failure. An experimental setup was utilized to investigate the dynamics, gaseous emissions, and energetics associated with thermal failure of cylindrical 18650 form factor, 2600 mAh, lithium cobalt oxide cathode cell arrays in an inert atmosphere. Cell state of charge (SOC) and arrangement were altered to investigate mitigation strategies and provide recommendations for safer battery pack design, transportation, and storage. Complete failure propagation was not prevented in any test, but decreasing to 50% SOC did result in propagation speeds 8.5 times slower than at 100% SOC. Introducing a 5 mm gap between cell rows also slowed propagation somewhat, but to a lesser degree than lowering SOC. Maximum temperatures, hazardous gas yields, and chemical heat generation were all reduced for cells at 50% SOC compared with 100% SOC, but introducing 5 mm gaps had little impact on these quantities. Limiting cells to 50% SOC was by far the most effective mitigation strategy tested, but no strategy was able to eliminate the failure propagation risk. Installing battery packs in inert environments, limiting SOC, and instituting gaps between certain cells can all be effective strategies to lessen the severity of cascading failure, but none of these strategies are individually sufficient for safe transportation and storage of battery packs.

A quantitative understanding of an intumescent material’s reaction to fire is largely an unsolved challenge in the field of fire science. To advance fire modeling, a systematic methodology to fully parameterize a comprehensive pyrolysis model for charring and intumescent materials is presented. Thermogravimetric analysis, differential scanning calorimetry and microscale combustion calorimetry were employed to characterize the kinetics and thermodynamics of thermal decomposition and heats of complete combustion of gaseous pyrolyzates. A multi-step reaction mechanism, consisting of sequential steps, was constructed to capture the physical transformations and chemical reactions observed in all milligram-scale experiments. Controlled Atmosphere Pyrolysis Apparatus II gasification experiments were conducted on 0.07 m diameter disk-shaped samples to parameterize the thermal transport within the undecomposed material and developing char layer. A recently expanded fully verified and validated numerical framework, ThermaKin2Ds, was employed to inversely model the gasification experimental results. The model accounted for spatially non-uniform swelling of the sample and associated changes in the radiant heat exposure. Rigid poly(vinyl chloride), a widely used intumescent material, was analyzed in this work. The resulting two-dimensional model was shown to reproduce the gasification experimental unexposed surface temperatures and mass loss rates with a mean error of 3.9% and 12.6%, respectively. A preliminary analysis of the char pore structure was also conducted to determine the pore size distribution and char porosity. The resulting porosity based on the density of graphite and the porosity based on image analysis (including only the pores that are greater than 1 × 10−4 m in diameter) was found to be 0.96 and 0.53, respectively.

This work details an approach to the development of a model of polymeric material fire behavior and its relation to flame retardant content. This approach employs a new controlled atmosphere pyrolysis apparatus to measure mass loss rate, back surface temperature, and sample shape profile evolution for 0.07-m-diameter disk-shaped samples exposed to well-defined radiant heating. Interpretation of these measurements using a thermal decomposition reaction mechanism, derived from thermal analysis experiments, and a numerical pyrolysis model, ThermaKin, yields properties that define heat and mass transport in the pyrolyzing solids. In the current study, this approach was extended to the analysis of flame retardant materials and applied to a set of materials comprised of glass-fiber-reinforced polybutylene terephthalate blended with aluminum diethyl phosphinate and melamine polyphosphate. Additionally, this work found evidence of so-called “wick” effect through which the molten polymer, when blended with glass fiber, was observed to be transported from regions of higher concentration to regions of lower concentration. Incorporation of the wick effect into the pyrolysis model was required to correctly capture the pyrolysis dynamics. The resulting pyrolysis model was found to be capable of predicting mass loss rate data as a function of material composition and external radiative heat fluxes ranging from 30 to 60 kW m −2 with an average error of 15%. Using heats of complete combustion of gaseous decomposition products determined in an earlier work, idealized cone calorimetry simulations were conducted to show that, when the gas-phase combustion inhibition effect is excluded, aluminum diethyl phosphinate has a relatively minor impact on the heat release rate, while the impact of melamine polyphosphate is significant. This work demonstrates, for the first time, that it is possible to establish a quantitative relation between the burning rate and material composition and thus, enables intelligent design of flame retardant materials tailored for specific applications.

To enable intelligent design of fire resistant polymeric materials, it is highly important to have a quantitative understanding of the relation between the composition of these materials and chemical and physical properties that control the process of fire growth. This paper presents a methodology that provides a capability to determine the core subset of these properties including the kinetics and thermodynamics of the thermal decomposition of the condensed-phase constituents and enthalpy of combustion of gaseous pyrolyzates. This methodology is based on three experimental techniques – Thermogravimetric Analysis, Differential Scanning Calorimetry and Microscale Combustion Calorimetry – and inverse numerical modeling of the results of these tests. The material system analyzed in this study is polyamide 66 reinforced with chopped glass fiber and flame retarded with red phosphorous. This system shows a complex decomposition behavior that is highly dependent on the phosphorus concentration. First, a semi-global thermal decomposition mechanism consisting of a set of first- and second-order (two components) reactions was developed using three material specimens containing different phosphorus concentrations. Subsequently, the extrapolative power of the developed reaction model was demonstrated by accurately predicting the experimental measurements obtained for several compositions not used in the model development process.

This contribution reports an approximate analytical model to predict transient mass flux and ignition time of translucent fuel, black poly(methyl methacrylate) (PMMA), subjected to a time-dependent incident heat flux, atb, where t is time and a and b are constants. The model can be easily extended to other non-charring translucent solids. The model takes into account in-depth absorption of thermal radiation in the condensed phase, which is typically ignored in the analytical formulations. Both critical temperature and critical mass flux were employed as the ignition criteria to examine their effects on the predictions. The model was validated using exact numerical solutions and experimental data, and compared with earlier analytical models based on the assumption of surface absorption. Linear and quadratic heat fluxes were considered for validation and discussion. The results show that surface absorption accelerates the pyrolysis process and leads to higher mass flux and shorter ignition time with respect to the in-depth absorption case. The discrepancy between the predicted transient mass fluxes of these two absorption modes increases with increasing a. The ignition heat flux increases with increasing a and decreases with increasing b for both surface and in-depth absorption cases. However, the critical energy is independent of heat flux in in-depth absorption scenario. Furthermore, parametric studies of in-depth absorption coefficient and critical mass flux were conducted to investigate their effects on the quality of the model predictions. Also, the equivalent ignition temperature was calculated and compared with the experimental values. It is expected that the developed model will find its use in performance-based design applications.

Flame heat feedback was measured in cone calorimeter tests for several polymers to develop a generalized flame model. Polymer samples were squares with 100 mm sides cut from 5.6 to 6.35 mm thick sheets of high-impact polystyrene, poly(methyl methacrylate), poly(oxymethylene), and glass-fiber reinforced (25 wt%) poly(butylene terephthalate) with 16 wt% of aluminum diethyl phosphinate. Flame heat flux was measured using water-cooled Schmidt-Boelter gauges in two locations: the sample center and 13.7 mm inwards from one edge of the sample and found to be 11–23 kW m−2 and 32–49 kW m−2, respectively. Based on the difference in measured heat flux, a center zone and a side zone were defined. Flame heat fluxes were partitioned into radiative and convective components based on an analysis of the materials’ radiative fractions. An average radiative flame heat flux of 9.5 kW m−2 was determined for the center zone while the side zone was assumed wholly convective. Convective flame heat fluxes were defined by a flame temperature of 2150 K and convective heat transfer coefficients of 3.7 and 20 W m−2 K−1 for the center and side zones, respectively. Based on an estimate for the convective flame heat flux, the center zone was determined to be a square with 54 mm sides while the remaining area represented the side zone. For validation, the flame model was coupled with well-established pyrolysis parameter sets and implemented into a numerical pyrolysis solver, ThermaKin, to produce the cone calorimetry simulations. The final model is an area-weighted combination of the center and side zone simulations. Both average heat release rate and peak heat release rate were predicted well by the final model. Ignition times for low irradiation were not predicted well initially; oxygen was suspected to be responsible. A correction was made to the ignition times to account for the effect of oxygen and it significantly reduced the error in predicted ignition times.

Lithium ion batteries (LIBs) are efficient, high-density electrical energy storage devices utilized in a perpetually increasing range of applications. One of the weaknesses of LIBs is that a small deviation from the normal operating conditions may cause an irreversible failure accompanied by a rapid self-heating and ejection of combustible gases and aerosols. The information on how much energy is released upon failure is critical for design of energy storage systems. In the current study, a novel technique, Copper Slug Battery Calorimetry (CSBC), was combined with oxygen consumption calorimetry to measure the rate of heat generated inside an LIB cell (PIHG) and the rate of heat generated as a result of combustion of ejected battery materials (PFlaming). A short duct equipped with a perforated plate was added to the original design of the CSBC apparatus to collect gases and aerosols ejected from the cell, reduce their flow speed and deliver them to a hot-wire igniter, which was used to initiate a diffusion flame. The exhaust from this flame was collected to measure the oxygen consumed in the combustion process and compute PFlaming. This approach to handling the ejected materials increased their combustion efficiency and eliminated thermal feedback from the flame to the cell, which enabled simultaneous measurement of PIHG and PFlaming. The new setup was employed to investigate thermally induced failure of an 1880 mA h prismatic LIB at various states of charge (SOC). It was determined that, at 100% SOC, this LIB released 33 ± 1.0 kJ of energy into the body of the cell and 113 ± 19 kJ was produced as a result of combustion of the ejected battery materials. The latter value is significantly greater than those previously reported for similarly sized cells, which can be explained by a more complete combustion achieved in this new apparatus.

A pyrolysis model capable of predicting materials’ fire behavior as a function of concentration was developed for an intumescent flame retardant system: poly(lactic acid) (PLA) blended with melamine (MEL) and ammonium polyphosphate (APP). The model was developed through inverse analysis of data obtained from bench-scale pyrolysis experiments wherein a 0.07-m-diameter disk-shaped sample was exposed to well-defined radiant heating in an anaerobic environment. Sample back surface temperature, sample shape profile and burning rate were measured simultaneously. A numerical pyrolysis modeling framework, ThermaKin2Ds, and a previously developed semi-global thermal decomposition reaction mechanism were employed in the inverse analysis to determine material properties that define the heat and mass transport inside the pyrolyzing solids. The final pyrolysis model was found to predict materials’ fire behavior for a variety of thermal exposures and material compositions. The model construction process revealed that a reduction in gas transfer coefficients helped to reproduce certain features of the burning rates profiles. Idealized cone calorimetry scenarios were simulated to study the influence of additives on materials’ fire behavior, and the results demonstrated that the blend with 5 wt % MEL and 25 wt% APP was found to be the most effective system with a 69% reduction in the average heat release rate comparing to that of PLA. A similar significant reduction has been reported in the literature, supporting the accuracy of this model. This work demonstrates a methodology that enables intelligent design of intumescent flame retardant materials tailored for specific applications, where low flammability is required.

Recent research has shown that acoustics can be used to suppress flames from a liquid fuel source. The results of these experiments indicated that acoustics alone are insufficient to control flames beyond the incipient stage. Recent research has also shown that variations in the delivery of water mist to a fire can enhance the mist’s efficiency. Therefore, the addition of acoustics to water mist may be an effective means of enhancing an established fire protection technology. For the first time, acoustics and water mist have been combined and studied as a flame suppression strategy. A series of experiments were conducted that explored the potential for coupling acoustics with water mist as means of flame suppression. Heptane fueled flames were created from two different sized ceramic fiber wicks: 30 mm × 50 mm with 5 mL of fuel, and 60 mm × 100 mm with 20 mL of fuel. The flames were then exposed to water mist delivered at a constant rate, which was found to be incapable of suppressing the flames. Next, low frequency sound waves at 62 Hz and 80 Hz were used to suppress flames from both wicks, with each frequency being generated by a different resonator. Finally, acoustics from both resonators were combined with water mist, and used to suppress flames from both wicks. The results showed that a combination of acoustic waves and water mist suppressed the flames more effectively than each individual technique on its own. This finding opens the possibility of developing more efficient ways to use water mist technology.

A variety of flame retardants and synergistic additives have been developed over the years in order to address the inherent flammability of polymeric materials. Environmental and economic concerns have recently driven industry to replace some of these flame retardants and synergists. Screening for potential replacements can be costly due to a variety of reasons. While some methods have been developed to screen flame retardants in a cost-effective manner, there is currently a need for a method that can also produce high quality data that provide insight into the physics behind the observed behavior during screening. This work proposes the use of Milligram-scale Flame Calorimetry (MFC) as a method to screen flame retardants in a highly quantitative, cost-effective manner. A methodology based on MFC measurements is presented and subsequently implemented using poly (butylene terephthalate) (PBT) as the base polymer and poly (pentabromobenzyl acrylate) (FR-1025) as the base flame retardant. This combination was chosen due to its well-understood behavior and compatibility. Using this polymer and flame retardant as a base, the flame retardant activity of antimony trioxide (ATO), aluminum hypophosphite (AL-HYPO) and a combination of melamine polyphosphate (MPP) and a proprietary formula (SOL-DP) as additives was explored with the proposed methodology. The fact that these measurements were performed with dry-blending of relatively small amounts per composition, as well as the milligram-scale nature of the measurements should represent a significant reduction in testing times and cost with respect to other available screening methods. The results show the outstanding performance of ATO, as measured by reduction in the heat of combustion. ATO performance is shown to be highly synergistic. For AL-HYPO, and MPP + SOL-DP, the performance is lower with respect to ATO. AL-HYPO does not exhibit synergistic behavior, although certain loadings can lead to lower heats of combustion. The complex interactions between FR-1025, MPP and SOL-DP show that some binary interactions are antagonistic while others are synergistic. Insight into the mode of action (gas-phase versus condensed-phase) was also revealed by the methodology with ATO exhibiting gas-phase-dominated activity, AL-HYPO showing a mixture of gas-phase and condensed-phase activity with no single mode standing out, while MPP + SOL-DP being predominantly condensed-phase active. Given a desired performance (for example, a range of desired heat of combustion values), the methodology allows for identification of optimum formulations that lead to said performance.

This study outlines a general methodology to develop a thermo-kinetic model of thermally-induced failure for a lithium ion battery (LIB) using COMSOL multiphysics software and experimental data collected with the Copper Slug Battery Calorimetry (CSBC). This methodology is demonstrated by developing the model for a specific type of LIB - T-Energy ICR18650. CSBC tests were conducted on the sample LIB cells to determine the thermal transport parameters and global reaction kinetics associated with the cells' thermal failure. The model was parameterized primarily via an iterative inverse modeling analysis of the CSBC test results. The parameters include heat capacity and density of the cell, radial thermal conductivity of the cell, and the Arrhenius parameters and heat of reaction representing the process of thermal runaway. The reaction parameters were related to the cell's state of charge using empirical functions interpolating these parameter values. The fully-parameterized thermo-kinetic model was then validated against CSBC tests that were conducted under conditions not utilized in the model parameterization process. Subsequently, this model was applied to predict the thermally-induced failure of LIB cells in a more complex scenario: cascading failure of 6 LIB cells in a "billiard rack" configuration. The simulated time of onset of thermal runaway of each sequentially failing LIB cell was found to be in a good agreement (within 9%, on average) with the corresponding experimental observations.

A campaign of experiments was conducted in a previously designed bench-scale wind tunnel to determine the effectiveness of suppression of lithium ion battery fires with a clean agent, Novec 1230 (CF3CF2C(O)CF(CF3)2). The experiments were performed on twelve 18650 form factor, fully charged, lithium cobalt oxide cells arranged in a rectangular array without intercell spacing. A set of baseline experiments performed at 640 l min−1 of air flow, 320 l min−1 of air flow, and 186 l min−1 of nitrogen flow indicated that, while flaming combustion had a significant impact on the speed of thermal runaway propagation through the array, fully preventing combustion did not prevent propagation. Addition of gaseous Novec 1230 to 640 l min−1 air flow immediately after the thermal runaway of the first cell produced 8.5 vol% concentration of the agent. At this concentration, which is above that recommended for suppression of traditional fires, the agent failed to suppress flaming combustion and did not prevent propagation of thermal runaway through the array. Increasing the concentration of Novec 1230 to 15.2 vol% by reducing the air flow rate to 320 l min−1 reduced the flaming combustion efficiency below 18% and prevented complete thermal runaway propagation in 67% of tests.

A systematic study was conducted to elucidate the effects of acoustic perturbations on laminar diffusion line-flames burning in air, and to determine the conditions required to cause acoustically-driven extinction. Line-flames were produced from the fuels n-pentane, n-hexane, n-heptane, and n-octane using fuel-laden wicks. The wicks were housed inside a burner whose geometry produced line-flames that approximated a two-dimensional flame sheet. The acoustics utilized ranged in frequency from 30 to 50 Hz, and acoustic pressures from 5 to 50 Pa. Prior to acoustic testing, the unperturbed mass loss rates and flame heights were measured. These quantities were found to scale linearly, which is consistent with the Burke-Schumann theory. The mass loss rates associated with hexane-fueled flames experiencing acoustic perturbations were then studied. It was found that the strongest influence on the mass loss rate was the speed of oscillatory air movement experienced by the flame. It was also found that the average mass loss rate increased linearly with the increasing air movement speed. Finally, acoustic perturbations were imposed on the flames from all fuels to determine acoustic extinction criterion. To ascertain if the observed phenomenon was unique to the alkanes tested, flames fueled by JP-8 (a kerosene-based fuel) were also examined. Using the data collected, a model was developed which characterized the acoustic conditions required to cause flame extinction. The model was based on the ratio of a modified Nusselt number to the Spalding B number of the fuel. It was found that at the minimum speaker power required to cause extinction, this ratio was a constant (independent of the chemical nature of the fuel).

A new gasification apparatus has been developed to enable a comprehensive analysis of pyrolysis of charring and intumescent materials. This apparatus provides well defined boundary conditions and highly resolved measurements of mass, temperature and sample profile evolution of a disk-shaped 0.07 m diameter material sample exposed to radiant heat. All measurements are collected simultaneously, in a single experiment, and recorded as a function of time. The oxygen concentration in the pyrolysis zone is controlled and can be reduced below 1 vol% to ensure that the measurements are free of oxidation effects. The radiation from an external conical heater has been carefully characterized to account for changes in the sample surface position, including the surface's angular orientation. Using an empirical expression, the radiation heat flux can be predicted with less than 2% error based on the known surface position and heat flux set point. The NIST Fire Dynamics Simulator (FDS) has been utilized in the direct numerical simulation mode to investigate convective losses from the sample surfaces. The convective heat transfer coefficient computed for the top (radiation exposed) surface has been found to be dependent on the surface position; its space-averaged value has been validated against experimental measurements. The capabilities of the apparatus are demonstrated using poly(vinyl chloride). It is shown that the apparatus provides repeatable data necessary for modeling of transport processes inside pyrolyzing intumescent solids. Non-one-dimensional nature of these processes is discussed.

In this work, experimental measurements of flame heat flux and sample mass loss rate are obtained as a diffusion flame spreads vertically upward (in the direction opposed to the vector of gravity) over the surface of seven commonly used polymeric materials, two of which are glass reinforced composites. Using these measurements, a previously developed empirical flame model specific to poly(methyl methacrylate) is generalized such that it can predict (flame to material surface) heat feedback from 3 to 20 cm tall flames supported by a wide range of materials. Model generalization is accomplished through scaling on the basis of a material's gaseous pyrolyzate heat of combustion, which can be measured using mg-sized material samples in a microscale combustion calorimeter. For all seven materials tested in this work, which represent diverse chemical compositions and burning behaviors including polymer melt flow, sample burnout, and heavy soot and solid residue formation, model-predicted flame heat flux (to a water-cooled heat flux gauge) is shown to match experimental measurements taken across the full length of the flame with an average absolute error of 3.8 kW m−2 (approximately 10–15% of peak measured flame heat flux). Coupled with a numerical pyrolysis solver, this generalized wall flame model provides the framework to quantitatively study material propensity to ignite and support early fire growth in a range of common scenarios with a level of accuracy and reduced computational cost unmatched by other currently available modeling tools.

Carbon fiber laminate composites have been utilized in the aerospace industry by replacing lightweight aluminum alloy components in the design of aircraft. By replacing low flammability aluminum components by carbon fiber laminates, the potential fuel load for aircraft fires may be increased significantly. A pyrolysis model has been developed for a Toray Co. carbon fiber laminate composite. Development of this model is intended to improve the understanding of the fire response and flammability characteristics of the composite, which complies with Boeing Material Specification 8–276. The work presented here details a methodology used to characterize the composite. The mean error between the predicted curves and the mean experimental mass loss rate curves collected in bench-scale gasification tests was calculated as approximately 17% on average for heat fluxes ranging from 40 to 80 kW m−2. During construction of the model, additional complicating phenomena were investigated. It was shown that the thermal conductivity in the plane of the composite was approximately 15 times larger than the in-depth thermal conductivity, the mass transport was inhibited due to the high density of the laminae in the composite, and oxidation did not appear to significantly affect pyrolysis at heat fluxes up to 60 kW m−2.

A standard material flammability assessment method, the Microscale Combustion Calorimetry (MCC), was modified and utilized to examine incomplete combustion of gaseous decomposition products of polymeric materials. The incomplete combustion conditions were attained by changing the MCC combustor temperature, oxygen concentration and residence time. A numerical model of the combustor, taking account spatial non-uniformity of its temperature, was developed and employed in inverse modeling mode to derive the global combustion kinetics for polystyrene and poly(butylene terephthalate) and their blends with gas-phase active flame retardant additives. The kinetic parameters were found to be highly sensitive to the presence of these additives. The speed of laminar flame propagating in a stoichiometric mixture of the pyrolyzates and air was estimated from the derived kinetics and proposed as a quantitative measure of gas-phase flame retardant effectiveness.

We report the discovery that a perhydroxylated dodecaborate cluster ([B12(OH)12]2-) can act as an inorganic polyol, serving as a molecular cross-linker in the synthesis of polyurethane-based materials. We further demonstrate how the inherent robustness of the utilized boron cluster can effectively enhance the thermal stability of the produced polyurethane materials incorporating [B12(OH)12]2- building blocks compared to analogous polymers made from carbon-based polyols. Ultimately, this approach provides a potential route to tune the chemical and physical properties of soft materials through incorporation of polyhedral boron-rich clusters into the polymer network.

Polyisocyanurate (PIR) foam is a robust thermal insulation material utilized widely in the modern construction. In this work, the flammability of one representative example of this material was studied systematically using experiments and modeling. The thermal decomposition of this material was analyzed through thermogravimetric analysis, differential scanning calorimetry, and microscale combustion calorimetry. The thermal transport properties of the pyrolyzing foam were evaluated using Controlled Atmosphere Pyrolysis Apparatus II experiments. Cone calorimetry tests were also carried out on the foam samples to quantify the contribution of the blowing agent (contained within the foam) to its flammability, which was found to be significant. A complete pyrolysis property set was developed and was shown to accurately predict the results of all aforementioned measurements. The foam was also subjected to full-scale flame spread tests, similar to the Single Burning Item test. A previously developed modeling approach based on a coupling between detailed pyrolysis simulations and a spatially-resolved relationship between the total heat release rate and heat feedback from the flame, derived from the experiments on a different material in the same experimental setup, was found to successfully predict the evolution of the heat release rate measured in the full-scale tests on the PIR foam.

Poly (methyl methacrylate) (PMMA) has long been a benchmark material of study in the fire research and fire modeling community due to its simple and predictable thermal decomposition behavior. Two common PMMA manufacturing processes, casting and extruding, produce polymers with slightly different properties and characteristics, primarily a result of differences in molecular mass. A dichotomy has been observed between study and application; while cast PMMA is more commonly studied and modeled, it is extruded PMMA that has wider use and application. This study investigates the differences between cast and extruded PMMA on a property-specific level, and whether the differences justify unique pyrolysis models for each. Experimental results show notable variation in milligram-scale tests (TGA, DSC and MCC), but only slight variation in gram-scale tests (Controlled Atmosphere Pyrolysis Apparatus II or CAPA II). Comprehensive pyrolysis models are developed for each PMMA type with the help of hill climbing optimization algorithms to automate inverse analysis. A sensitivity analysis is performed on the fully parameterized models to examine the effect of prominent parameters on the model's ability to predict mass loss rate from CAPA II tests. It is determined that the two materials are sufficiently similar to be represented with a generic model.

A new experimental setup was developed to study turbulent, buoyancy-driven flame spread on a corner wall. Two 146 × 50 × 0.58 cm panels of cast black poly(methyl methacrylate) (PMMA), whose pyrolysis properties were fully characterized in a separate study, were ignited using a triangular propane burner. The corner wall assembly was placed into a well-ventilated enclosure and heat release rate (HRR) and flame heat flux to the corner wall were measured in 7 repeated experiments. The HRR measurement was designed to achieve a fast response time (13 s) and was correlated with the flame heat flux, which was obtained using water-cooled heat flux gauges positioned in 28 locations distributed over the PMMA surface. Simultaneously, a monochromatic, 900 nm, video of the spreading flame was recorded and analyzed to determine the evolution of soot radiation intensity and flame geometry. Using collected data, an empirical model relating spatially resolved heat feedback from the flame to the solid fuel surface and HRR was developed and integrated with a detailed pyrolysis model. Simulations were conducted in an uncoupled and coupled mode. The uncoupled simulations, where the flame heat feedback was effectively prescribed to match the experimental measurements, revealed the importance of the knowledge of the split between the convective and radiative portions of the flame heat flux. The coupled simulations, where the flame heat feedback was computed from the simulated HRR, revealed that the coupling amplifies even relatively small uncertainties in the model parameters to produce large errors in the HRR predictions.

A novel experimental technique, Copper Slug Battery Calorimetry (CSBC), was employed for the measurement of the energetics and dynamics of the thermally-induced failure of 18650 form factor lithium ion batteries (LIBs) containing three different cathodes: lithium cobalt oxide (LCO), lithium nickel manganese cobalt oxide (NMC) and lithium iron phosphate (LFP). The heat capacity of these LIBs was evaluated to be 1.1±0.1 J g−1 K−1 for all three types. It was shown that the total heat generated inside the batteries increases with increasing amount of electrical energy stored. The maximum total internal heat generated by fully-charged LIBs was found to be 37.3±3.3, 34.0±1.8 and 13.7±0.4 kJ/cell for LCO, NMC and LFP LIBs, respectively. Detailed modeling of heat transfer in the CSBC experiments was carried out to evaluate thermal conductivities of the LIBs and demonstrate that the assumptions associated with the CSBC experiment analysis are valid. Additionally, experiments were carried out in which the CSBC technique was combined with cone calorimetry to measure the heat produced in flaming non-premixed combustion of vented battery materials. The released combustion heat varied between 35 and 63 kJ/cell for LCO LIBs, 27 and 81 kJ/cell for NMC LIBs, and 36 and 50 kJ/cell for LFP LIBs.

A methodology for parameterization of kinetics and thermodynamics of decomposition of polymeric materials has been extended to blends of a polymer with condensed-phase active flame retardants. This methodology is based on Thermogravimetric Analysis, Differential Scanning Calorimetry, Microscale Combustion Calorimetry and inverse numerical modeling of these experiments. Material systems consisting of poly(lactic acid), melamine and ammonium polyphosphate were used to demonstrate this parameterization process. The resulting model consists of a set of first and second order (two component) reactions that define the rate of gaseous pyrolyzate production, heats of these reactions, heat capacities of the condensed-phase reactants and products and heats of combustion of the components of the gaseous pyrolyzate. This model is shown to reproduce results of all aforementioned experiments with a high degree of detail and predict relation between the outcome of these experiments and material composition. It is expected that a combination of this model with thermal transport parameters, which determination will be a subject of separate study, will yield a complete pyrolysis model capable of predicting the dynamics of burning and flame spread on these materials and dependence of this dynamics on the flame retardant content.

Brominated flame retardants (BFRs) are widely used in polymers due to their high effectiveness at relatively low cost. Recent studies suggest that certain BFRs may present health and environmental hazards, yet obtaining adequate replacements is an ongoing challenge. To develop new additives, it is of paramount importance to better understand the mechanisms governing the action of flame retardants in polymers, particularly their action in the gas phase, and to develop effective screening techniques for potential candidates. To address this challenge, a novel apparatus, the Milligram-scale Flame Calorimeter (MFC), is proposed. In MFC, the pyrolysis and gas-phase combustion processes are uncoupled. Samples of 35 ± 5 mg are pyrolyzed in an anaerobic atmosphere and their pyrolysis products are burned in a laminar, near axisymmetric diffusion flame under controlled and customizable conditions in a fully enclosed system. Heat release information is obtained through the oxygen consumption calorimetry. The masses of post-combustion pyrolysis residue and solid particulate combustion products (e.g., soot) are measured. In this manuscript, a description of the design, parametric optimization of test conditions, and overall testing methodology is given. Heat release measurements are presented for polystyrene with increasing amounts of brominated polystyrene as well as poly(butylene terephthalate) with increasing concentrations of aluminum diethylphosphinate and results are compared to the Cone Calorimetry (Cone) and Microscale Combustion Calorimetry (MCC) measurements conducted on the same materials. Using combustion efficiency as a metric, the sensitivity to both bromine and phosphorus gas-phase activity is determined. The impact of these flame retardants is pronounced in the Cone experiments and largely undetected by the MCC. The MFC shows trends with respect to the flame retardant content comparable to the Cone. However, the absolute values of MFC combustion efficiencies tend to be notably higher that those observed in the Cone experiments. Overall, the MFC results demonstrate that this novel apparatus can be used to detect gas-phase activity of flame retardants using milligram-sized samples.

A methodology to parameterize (or calibrate) comprehensive pyrolysis models is demonstrated on plasticized and rigid poly(vinyl chloride) specimens. Thermogravimetric analysis, differential scanning calorimetry and microscale combustion calorimetry were employed to characterize the kinetics and thermodynamics of thermal decomposition and heats of complete combustion of gaseous decomposition products. Controlled Atmosphere Pyrolysis Apparatus II gasification experiments were utilized to determine the thermal transport parameters within the undecomposed material and developing char. A numerical framework, ThermaKin2Ds, was employed to analyze and simulate all gasification experimental results. The two-dimensional (axisymmetric) models of plasticized and rigid poly(vinyl chloride) were shown to reproduce the gasification mass loss rates with mean errors of less than 19%. Idealistic, one-dimensional simulations of large burning panels in the presence of external radiant heat flux were subsequently conducted using the fully parameterized pyrolysis models. A quantitative comparison between the simulated plasticized and rigid poly(vinyl chloride) large-scale burning indicated that these materials have similar mass loss rate histories. However, significant differences become apparent when the mass fluxes of gaseous decomposition products are converted to heat release rate; the heat release rate of plasticized poly(vinyl chloride) was found to be approximately a factor of 2 greater than that of the rigid poly(vinyl chloride).

To investigate thermal runaway propagation in lithium ion cell arrays (referred to as cascading failure), a set of experiments was conducted under well-controlled conditions in a wind tunnel where 18650 form factor cells were arranged into a densely packed array. The cell arrays were constructed from 12 lithium ion cells arranged in a rectangular configuration (3 columns by 4 rows) with no inter-cell spacing (each cell was in a contact with the adjacent cells). Commercial cells of different cathode chemistry, lithium cobalt oxide (LCO), lithium nickel manganese cobalt oxide (NMC), and lithium iron phosphate (LFP), were employed as test samples in this study. All studied cells were fully charged. The experiments were conducted in both anaerobic (N2) and air environments to demonstrate the impact of flaming combustion. In each experiment, thermal runaway was triggered into the middle cell of row 1 using an electric heater and was observed to propagate through the array via thermocouples attached to the cell bottom surfaces. The dynamics of cascading failure were quantified together with thermal energy release for each cathode chemistry. Furthermore, the yields of the ejected O2, total hydrocarbons (THC), CO, CO2, and H2 were determined; corresponding fire hazards were assessed. For LCO and NMC cell arrays tested in an N2 or air environment, it was found that thermal runaway of the intentionally failed cell was sufficient to initiate cascading failure that fully propagated through the array. LFP cell arrays, however, underwent incomplete/partial cascading failure (maximum of six and eight cells out of twelve failed in N2 and air, respectively). When LCO, NMC, LFP cell arrays were tested in air rather than N2, the row-to-row thermal runaway propagation speed increased by factors of 9, 2, and 7, respectively, which was attributed to the influence flaming combustion of ejected materials. On average, individual LCO, NMC, and LFP cells lost 38.4%, 54.7%, and 11.7% of their initial mass, respectively, when tested in N2, and this percentage increased by 0.3-3.2% when tested in air. Measurements showed that failed LCO and NMC cells produced minor mass yields of O2 and H2 in addition to relatively large amounts of THC, CO, and CO2. The yields of gases ejected from LFP cells were significantly less than the yields obtained for LCO and NMC cells. The lower flammability limit of the flammable portion (THC, CO, and H2) of the ejected gases was found to be 5.79 ± 0.12, 5.92 ± 0.15, and 4.28 ± 0.15 vol.% in air for LCO, NMC, and LFP cells, respectively. The maximum volume of an enclosure where the gas ejection from a single cell creates a flammable mixture in an air environment was estimated to be 0.087 ± 0.017, 0.076 ± 0.0146, and 0.029 ± 0.019 m3 for LCO, NMC, and LFP cells, respectively. The sum of the chemical heat (associated with chemical reactions between battery materials inside and outside the cell casing during failure) and flaming combustion energy (associated with burning of ejected cell materials) was found to be 3.2, 1.9, or 2.5 times greater than the electrical energy stored for LCO, NMC, or LFP cell, respectively. Overall, the obtained results showed that LFP-based cell arrays are the least hazardous arrays. Therefore, they are recommended as a safe option, even though the nominal capacity of the LFP cells is relatively low.

© 2018 Elsevier Ltd This paper provides a report of the discussions held at the first workshop on Measurement and Computation of Fire Phenomena (MaCFP) on June 10–11 2017. The first MaCFP workshop was both a technical meeting for the gas phase subgroup and a planning meeting for the condensed phase subgroup. The gas phase subgroup reported on a first suite of experimental-computational comparisons corresponding to an initial list of target experiments. The initial list of target experiments identifies a series of benchmark configurations with databases deemed suitable for validation of fire models based on a Computational Fluid Dynamics approach. The simulations presented at the first MaCFP workshop feature fine grid resolution at the millimeter- or centimeter-scale: these simulations allow an evaluation of the performance of fire models under high-resolution conditions in which the impact of numerical errors is reduced and many of the discrepancies between experimental data and computational results may be attributed to modeling errors. The experimental-computational comparisons are archived on the MaCFP repository [1]. Furthermore, the condensed phase subgroup presented a review of the main issues associated with measurements and modeling of pyrolysis phenomena. Overall, the first workshop provided an illustration of the potential of MaCFP in providing a response to the general need for greater levels of integration and coordination in fire research, and specifically to the particular needs of model validation. ispartof: Fire Safety Journal vol:101 pages:1-17 ispartof: location:Switzerland status: published

This work demonstrates an approach to building a material burning model capable of predicting the material's behavior as a function of concentration of a condensed-phase-active flame retardant. This approach relies on a newly developed gasification apparatus to measure back surface temperature, mass loss rate, and shape profile evolution for 0.07-m-diameter disk-shaped samples exposed to well-defined radiant heating in an anaerobic environment. Inverse analysis of these gasification measurements, using a predetermined thermal decomposition reaction mechanism, yields properties that define heat and mass transport in the pyrolyzing solid. In the current study, this approach is applied to a set of materials comprised of glass fiber reinforced polyamide 66 blended with red phosphorus. A single pyrolysis model is developed that relates the concentration of the additive to the material burning rate. During the model construction, it was revealed that, to successfully model pyrolysis of this glass-fiber-filled thermoplastic, it is important to take into account so-called wick effect, through which the molten polymer was transported toward the heated surface. Incorporation of this effect into the model enabled a more accurate prediction of the burning rate. Idealized cone calorimetry simulations were conducted to demonstrate that red phosphorus has a significant impact on the heat release rate. Any red phosphorus added above 2 wt% reduces both the first and second heat release rate peaks by approximately 38% and this reduction can be attributed to the condensed-phase thermal barrier effect and, to a lesser degree, dilution of the gaseous decomposition products with less combustible gases.

A new model, which represents a unified description of material thermal degradation and burning under thermally thin (non-dimensional), surface radiant heating (one-dimensional) and upward flame spreading (two-dimensional) conditions, has been developed by coupling the numerical pyrolysis solver ThermaKin2D, whose function is to compute the transient rate of gaseous fuel production of a material in response to external heat transfer, with an empirical flame model that predicts a wall flame's heat feedback profile as a function of material mass loss rate. A previously developed pyrolysis model of poly(methyl methacrylate), which was parameterized using a combination of milligram-scale simultaneous thermal analysis experiments and gram-scale gasification tests, was validated in this study using gasification experiments distinct from those utilized in the parameterization process. A previously developed wall flame model was reformulated to include results of new heat flux measurements from 3 to 20 cm above the base of the flame. Inclusion of these results improved the model's predictive capabilities at ignition and across a larger range of flame sizes. The new unified model was employed to predict vertical burning and upward flame spread on 4 and 17.5 cm tall samples of poly(methyl methacrylate). The model predictions – including time to ignition and initial, peak, and rate of rise of sample mass loss rate – were found to closely match experimental results. The impacts of melt flow effects and uncertainties in the flame model formulation on the unified model predictions were examined and found to be moderate.