Fire conditions for smoke toxicity measurement

This paper identifies those fire conditions most often present when smoke toxicity is the cause of death. It begins with a review of the evidence that smoke-inhalation deaths are in the majority in fire fatalities in the United States. Next, there is an analysis of the evidence from the national fire experience showing the connection between post-flashover fires and smoke-inhalation deaths. Third is a presentation of real-scale fire test results demonstrating that post-flashover conditions are necessary to produce enough smoke to cause smoke-inhalation deaths in the cases where they actually occur. The fourth component is a sampling of results from computer simulations of fires, affirming and broadening the results from the fire tests. It is concluded that smoke-inhalation deaths occur predominantly after fires have progressed beyond flashover. This conclusion then provides a focus for smoke toxicity measurement in particular and fire hazard mitigation in general.     

Sandwich panel performance in full-scale and bench-scale fire tests

Fire hazard assessments must be primarily driven by life safety variables. Concern is often highly focused on toxicity issues, since fire deaths, in the majority of cases, are found (in whole or in part) to be due to toxic gas inhalation. Procedures have recently been published by ISO, wherein the toxicity assessment of fire products is focused primarily on bench-scale testing for toxic potency (the ‘per-gram toxicity’). Yet hazards of products with regards to fire toxicity may be determined much more by their differences in burning rates than by any differences in toxic potency. Burning rates are not assessed in the pertinent standards (ISO 13344 and ISO TR 9122). For most product categories, techniques for predicting full-scale burning rates from bench-scale data are not yet in hand. Thus, today the best means of comparing actual, full-scale toxic fire hazards is the full-scale fire test, equipped with additional gas measuring instrumentation. Such an approach is not among the recommended methods of the international standards, yet it is the only one with innate validity. In the present work, a series of sandwich panel products were tested in a full-scale room configuration. Bench-scale comparison was made to the ISO 5660 Cone Calorimeter and the DIN 53436 tube furnace. The toxic gases were quantified in all cases by chemical analysis. The product which showed the best performance in the full-scale tests (rock wool insulated sandwich panel) did not achieve a good fire toxicity performance due to minimization of toxic potency. Instead, the successful performance was attributed wholly to reduction of burning rate. Bench-scale measurements of toxic potency were shown to lack relevance to reality in such cases where even the full-scale toxic potency is not a determining factor. © 1997 John Wiley & Sons, Ltd.     

The effect of temperature and ventilation condition on the toxic product yields from burning polymers

A major cause of death or permanent injury in fires is inhalation of toxic gases. Moreover, every fire is unique, and the range of products, highly dependant on fire conditions, produces a wide variety of toxic and irritant species responsible for the most fire fatalities. Therefore, to fully understand each contribution to the toxicity it is necessary to quantify the decomposition products of the material under the test. Fires can be divided into a number of stages from smouldering combustion to early well‐ventilated flaming through to fully developed under‐ventilated flaming. These stages can be replicated by certain bench‐scale physical fire models using different fuel‐to‐oxygen ratios, controlled by the primary air flow, and expressed in terms of the equivalence ratio (the actual fuel/air ratio divided by the stoichiometric fuel/air ratio). This work presents combustion product yields generated using a small‐scale fire model. The Purser Furnace apparatus (BS7990 and ISO TS 19700) enables different fire stages to be created. Identification and quantification of combustion gases and particularly their toxic components from different fire scenarios were undertaken by continuous Fourier transform infrared spectroscopy. The relationship between type of the fire particularly the temperature and ventilation conditions and the toxic product yields for four bulk polymers, low‐density polyethylene, polystyrene (PS), Nylon 6.6 and polyvinyl chloride (PVC) is reported. For all the polymers tested, except PVC, there is a dramatic increase in the yield of products of incomplete combustion (CO and hydrocarbons) with increase in equivalence ratio, as might be expected. For PVC there is a consistently high level of products of incomplete combustion arising both from flame inhibition by HCl and oxygen depletion. There is a low sensitivity to furnace temperature over the range 650–850°C, except that at 650°C PS shows an unexpectedly high yield of CO under well‐ventilated conditions and PVC shows a slightly higher hydrocarbon yield. This demonstrates the dependence of toxic product yields on the equivalence ratio, and the lack of dependence on furnace temperature, within this range. Copyright © 2007 John Wiley & Sons, Ltd.     

Post-Flashover compartment fires

Fourteen mathematical models of post-flashover compartment fires are classified on the basis of fourteen principal modeling aspects. Expressions are presented for the potential of fire to spread by destruction and convection. The assessment of the fire resistance requirements for the compartment boundaries is discussed and measures to counter the potential of fires to spread by convection are outlined.     

Post-flashover compartment fires: Basis of a theoretical model

A theoretical model of compartment fires in the post-flashover stage is presented. The model incorporates the stirred reactor assumption and window flow approximation of Kawagoe, but treats fuel burning rates and burning regimes in a theoretical manner. The resulting formulation preserves the important features of the fire behavior but is simple enough to permit use for building fire safety design purposes.     

Determination of water application rates required for communities to suppress post-flashover informal settlement fires based on numerical modelling and experimental tests

Fires originating in informal settlements (ie, slums, ghettos, shantytowns, squatter camps) spread rapidly, due to the presence of densely packed, highly combustible dwellings, thereby making these communities inherently susceptible to large conflagrations. By the time, the fire brigades are notified and can get to the scene of the fire, the resulting conflagrations can be large. Thus, it is necessary to equip communities with the ability to combat smaller fires, although it is acknowledged that this is not ideal. Previous full-scale testing and firefighter experience have shown that water application through ‘bucket brigades’ can be very effective at suppressing fires. In this article, a model is developed for approximately quantifying the amount of water, and discharge rate, that is, required for communities to suppress fires of various sizes using bucket brigades. This is done to answer the question: based on the water supply infrastructure in an area could a community put out post-flashover fires of certain sizes? If this is not feasible, it would highlight the importance of communities having readily available pre-filled water buckets at homes. The model presented is developed in fire dynamics simulator and is calibrated based on full-scale experiments utilizing the bucket brigade technique. It is shown that standpipe discharge rates of 23 to 40 lpm are suitable for fire sizes of around 3.85 MW, based on a dwelling size of 2.4 x 3.6 x 2.4 m. This means that in communities with a single stand-pipe (water supply point) with flow rates less than 23 lpm, that fires greater than 3.85 MW (as produced by a home of 2.4 x 3.6 m with a timber fuel load of 25 kg/m²) cannot be suppressed in time without resulting in substantial fire spread to adjacent dwellings.     

A generalized relationship between the normalized yields of carbon monoxide and hydrogen cyanide

Experimental studies have demonstrated that there are close correlations between the normalized yields of carbon monoxide (CO) and hydrogen cyanide (HCN) from the combustion of materials containing nitrogen. In this paper, a generalized relationship using the stoichiometric oxygen to fuel mass ratio (SOFMR) is derived to represent these correlations. Using this generalized relationship, the predicted yields of HCN for nylon in tube furnace experiments and HCN concentrations in full‐scale cable fire tests are in good agreement with the corresponding measured data. The derived relationship is used to analyse the contributions of CO from different materials in a complex fire reconstruction. The generalized relationship is then used to predict HCN concentrations in two full‐scale nylon fires and the predicted concentrations are compared with both experimental data and predictions from a flamelet model. Finally, a method to incorporate the generalized relationship within CFD fire simulations to determine HCN (or CO) concentrations based on measurements of CO (or HCN) yields is presented. Copyright © 2010 John Wiley & Sons, Ltd.     

Furniture and furnishings in the home—some fire statistics

A statistical study of fires in the United Kingdom involving the ignition of furniture and furnishings is presented. This paper examines the data for one year (1970). The analysis shows that in fires starting in furniture and furnishings the chance of a fatality is over twice that in other domestic fires. The majority of furniture fires involve upholstery or bedding and over 90% were started by smokers' materials, electric appliances, space heating or as the result of the activities of children or suspected arsonists. Eighty-five percent of the fatalities were found in the room of origin of the fire. Eighty per cent were overcome by smoke or toxic gases. Sixty percent of the fatalities were either under 5 or over 65 years of age. Monetary values are assigned for damage, casualties and deaths in fire. These costs can be used to assess the value of fire precautions. With the values taken, the total losses in furniture fires in the home amounted to £19 million in 1970. Life loss accounted for the major part of this sum. The expected annual loss per dwelling as a result of the ignition of furniture is thus only about £1, and is only £3 for all dwelling fires. This low figure suggests an approach of either selective spending on those most at risk (the elderly and handicapped) or by government activity through publicity and education.     

Specimen heat fluxes for bench-scale heat release rate testing

When a specimen is testd for its heat release rate (HRR) behaviour using a bench-scale such as ISO 5660 or equivalent, one very important test condition is not pre-standaridized and must be set: the heat flux to be imposed on the specimen by the heater. The heat flux cannot be ligitimately standardized, since the value appropriately to be used will differ according to purpose or application. The present paper sets forth the considerations which should govern the correct choice of heat flux. A discussion is given of minimum ignitiability level; statistical variability at low heat fluxes; the ranges of heat fluxes associated with small actual ignition sources; the heat fluxes associated with fires away from the ignition sources, all the way up to fully-involved room fires; the application of the product; and the needs associated with mathematical modeling of room fires. Correlational approaches are also illustrated and contrasted to physics-based ones. Finally, the empirical nature of the present situation is emphasized. Judged from first principles, it would appear that successful prediction of room fire results from bench-scale test data would require both the testing at a large number of different heat fluxes and the use of algorithms to permit time-dependent interpolation. Such algorithms have been proposed; however, some very successful predictions are noted with much simpler techniques.     

Assessment of toxic hazards of smoke: Toxicological potency and intoxication rate thresholds

Smoke toxicity test methods which determine only lethal toxicological potency under ‘worst case’ conditions do not satisfactorily address the critical issue of relative hazard, including time-to-escape and tenability limits resulting from the fire performance of materials under comparable conditions. Since threats to escape from a fire are largely time-dependent, toxic insults produced by burning materials should also be considered as rate processes. Assessment of time thresholds exhibited by burning materials under test conditions to effect performance impairment (incapacitation) of an animal model would appear to be more relevant than lethal toxicological potencies in estimating probability of successful escape from fires. A model is advanced in which intoxication rate thresholds for materials are obtained using a rodent exposure test method. Concentration-time curves, obtained from experimentally derived concentration-time-response surfaces, are the basis for estimating rate thresholds which are distinctively different for each material and which vary as a function of test conditions. it is this performance impairment response surface which is potentially a key to the modeling of toxic hazards of smoke in perspective with other hazards presented by fire.     

Highway vehicle fire data based on the experiences of US fire departments

In 2003–2007, US fire departments responded to an average of 267 600 highway vehicle fires per year. These fires caused an average of 441 civilian deaths, 1326 civilian injuries, and $1.0bn (in US dollars) in direct property damage annually. Highway vehicles include cars, trucks, and other vehicles designed for highway use; highway vehicle fires can occur anywhere, not just on a highway. While these fires and associated losses have been falling in recent years, highway vehicles fires accounted for 17% of reported US fires, 12% of US fire deaths, 8% of US civilian fire injuries, and 9% of the direct property damage from reported fires. Data from the US Fire Administration's National Fire Incident Reporting System and the National Fire Protection Association's fire department survey were used to provide details about the circumstances of highway vehicle fires. Mechanical or electrical failures caused roughly three‐quarters of the highway vehicle fires but only 11% of the deaths. Collisions and overturns were factors contributing to the ignition in only 3% of the fires, but fires resulting from these incidents caused 58% of these vehicle fire deaths. The rate of bus fires per billion miles driven was 3.5 times that for highway vehicle fires overall. Copyright © 2012 John Wiley & Sons, Ltd.     

Analysis of work on smoke component yields from room‐scale fire tests

Smoke Component Yields from Room‐Scale Fire Tests (NIST Technical Note TN 1453) has recently been published. This was expected to be an important work in developing concentrations and yields of toxicants that could be used for evaluating the usefulness of small scale smoke toxicity apparatuses (or fire models) for use in the prediction of the toxicity of materials and products in real fires. However, the work has a number of uncertainties that limit its potential for use as a reference. There are three major problems with this work. First, the post‐flashover concentrations of CO are too low (as recognized by the authors who recommend that this part of the data not be used). Second, the post‐flashover concentrations of the main toxicants measured (HCN and HC1) were much higher than found in most studies. Third, the precision of the data was inadequate. The consequence of the first two issues is that the work seriously overestimates the toxicological importance of gases known to have only minor effects in post‐flashover fires, such as HCN and HCl. The very low concentrations of toxicants measured at pre‐flashover conditions might have a value not discussed by the authors: an indication that pre‐flashover fires of the type conducted here do not generate extremely toxic atmospheres. Accordingly, the report does not provide reliable characteristic room scale combustion gas data that can be used for validating small‐scale furnaces. Copyright © 2005 John Wiley & Sons, Ltd.     

An update to what's burning in home fires

This article is an update on the items that are first ignited in home fires and it examines the extent of their involvement in associated deaths, injuries and property loss. It will be limited to forms of material that are considered products. Trash, for example, is involved in a large portion of home fires, but is not a product in the home and therefore is not included. Also excluded are building materials such as structural components and exterior wall coverings. This exclusion is important, because many building items rank high in the overall list of materials first ignited, but are not considered consumer products. Further detail will be given to the top ten leading products associated with home fires and home fire deaths. Copyright © 2001 John Wiley & Sons, Ltd.     

Soft furnishing fires: They're still a problem

Home fires in which soft furnishings were first ignited or contributed to fire spread cause a disproportionate share of US home fire deaths. Although the death toll from these fires is much lower than in the 1980s, the rate of death per 1000 reported fires has doubled. Data from the National Fire Incident Reporting System and the National Fire Protection Association's annual Fire Experience Survey were used to create national estimates of these fires and casualties. Including both fires in which upholstered furniture was the item first ignited and contributed most to fire spread, in 2013‐2017, upholstered furniture was involved in an estimated average of 7120 (2%) home fires per year, resulting in an average of 570 (22%) deaths per year. Mattresses or bedding were involved in estimated average of 10 530 (3%) home fires per year, resulting in an average of 370 (14%) deaths annually. In most of these fires and deaths, the item was first ignited. Smoking materials were the leading cause of these fires (20%‐26%) and associated deaths (52%‐54%). Various types of operating equipment and small open flames also play a role. Understanding the causes of these fires is essential to develop new strategies to prevent them.     

Design and testing of a new smoke concentration meter

The design of a new smoke concentration meter based on light‐extinction measurements with a He‐Ne laser is described. The measurement allows the determination of the mass‐generation rate of smoke and smoke yield during a fire test with little more time or labour than is required for performing heat‐release‐rate and mass‐loss‐rate measurements. The new smoke concentration meter was motivated by the finding from several studies of a nearly universal value of the specific extinction coefficient of post‐flame smoke produced by over ventilated fires. Key design features include the use of a stabilized laser, purge flow to eliminate smoke deposition on the optics, U channel construction to minimize the effect of heating on the optical alignment and beam correction optics. The facility was fabricated almost entirely from commercially available components to allow this design to be easily reproduced by fire research and testing laboratories. The smoke concentration meter was able to measure a smoke yield as small as 0.005 for a propane fire to as large as 0.10 for a toluene pool fire. A detailed uncertainty assessment was made. The result for a 50cm diameter heptane pool fire agrees well with previous smoke yield measurements made for the same fuel and pool diameter based on filter collection and weighing. Copyright © 2000 John Wiley & Sons, Ltd.     

The fire performance of PVC

The role of PVC in fires is currently a controversial topic because of the many negative comments made about PVC on the occurrence of any major fire disaster. Critics also use many small-scale smoke and toxic gas tests to define the role of PVC in these fires. The purposes of this paper are (1) to summarize the current technical knowledge of real fire behavior, (2) use this understanding to interpret available data for PVC in large- and small-scale fire tests, and (3) help bring a sense of technical realism to the issues involved.     

Update on smoke toxicity of vinyl compounds

All organic materials burn and give off toxic products. These always include water, carbon dioxide, and the single gas causing the greatest hazard in fires—carbon monoxide (CO). The intrinsic toxicity of the smoke of all combustible materials, including PVC, is very similar in terms of lethality, with very few exceptions. Toxicity of vinyl compounds is due to two major gases: CO and hydrogen chloride (HCI). Since natural combustible materials are not chlorinated, speculation has arisen about the toxicity of HCl and of PVC smoke. Recent studies have shown that it takes similar doses of HCl and CO to kill rats. Furthermore, rats and baboons will tolerate the same levels of HCl. However, mice are much more sensitive than either rats or baboons towards HCl. Baboons are a very good model for humans; therefore, mice will be killed by exposure to much lower HCl levels than those required to kill humans. HCl concentrations in real fires are quite low: HCl decays rapidly by reacting with wall materials such as gypsum, cement, or ceiling tile. It does not, however, react rapidly with plastic or glass walls, which is where toxicity tests are carried out. Therefore PVC smoke is less hazardous in reality than it appears to be from toxicity test results. Since most products have similar intrinsic toxicities, as regards lethality, the real toxicity in a fire is a consequence of the rate of generation of gases. PVC is a difficult polymer to ignite and burns very slowly, so that it will give off less toxic products per unit time than many other common materials and cause lower fire hazard.     

Toxic gas tests in perspective

A study of the 1979 UK fire statistics and the Annual Report of the Office of Population Censuses and Surveys shows that only 1 death per 1000 is attributable to toxic gases from combustion. However, of the ‘deaths by violence’ it rises to 30 (3%). As 37% of all fires in occupied buildings were confined to the first item ignited and 53% to the area of origin it is postulated that such fires (and the consequent deaths) largely concerned substances which will burn on their own. From experience these are largely materials with limiting oxygen indices below 22 or 22–28 for restricted burning only. Those with limiting oxygen indices >28 cannot on their own sustain combustion. These considerations suggest the mode in which toxic gas tests should be conducted, i.e., horizontally or vertically and vertically only for the first two groups, respectively, and in the presence of a fire sustainer for the limiting oxygen index >28 group. Wood in the form of a wooden crib is suggested as suitable. To this end some experiments are described with PVC placed on top of a No. 5 crib (PSA). These latter show that, making the best assumption possible from the existing data, the wood contributes about 63% to the toxicity and the PVC 37%. On the large scale, other work has shown that wooden cribs alone will give lethal concentrations of carbon monoxide for significant periods. Thus difficult-to-burn substances need to evolve gases of exceptional toxicity for their contribution to be worth considering.     

Fire effluent component yields from room‐scale fire tests

Estimation of the time available for escape (ASET) in the event of a fire is a principal component in fire hazard or risk assessment. Valid data on the yields of toxic smoke components from bench‐scale apparatus is essential to accurate ASET calculations. This paper presents a methodology for obtaining pre‐flashover and post‐flashover toxicant yields from room‐scale fire tests. The data are to be used for comparison with bench‐scale data for the same combustibles: a sofa, bookcases, and electric power cable. Each was burned in a room with a long adjacent corridor. The yields of CO₂, CO, HCl, HCN, and soot were determined. Other toxicants (NO₂, formaldehyde, and acrolein), whose concentrations were below the detection limits, were of limited importance relative to the detected toxicants. The uncertainty values were comparable to those estimated for calculations used to determine ASET and were sufficiently small to determine whether a bench‐scale apparatus is producing results that are similar to the real‐scale results here. The use of Fourier transform infrared spectroscopy was useful for obtaining toxicant concentration data; however, its operation and interpretation are not routine. The losses of CO, HCN, and HCl along the corridor were dependent on the combustible. Copyright © 2009 John Wiley & Sons, Ltd.     

Reduced combustion efficiency of chlorinated compounds,resulting in higher yields of carbon monoxide

A series of experiments involving the combustion of inherently fire-retarded compounds has been performed using the DIN 53 436 furnace. Six chlorinated compounds comprising four pesticides, an organic solvent, and polyvinyl chloride (PVC) have been investigated with the focus on the combustion efficiency expressed as the CO₂/CO ratio in the fire effluent. It was found that increased chloride content in a compound resulted in decreased CO₂/CO ratio. The correlation was found to be $ \sqrt {{\rm CO}_{\rm 2} /{\rm CO}} = 0.61 \times ({\rm Cl/C)}^{- {\rm 1}} + 0.13 $. The results indicate higher concentrations of CO in the fire effluent from chlorinated compounds than from wood. Together with a release of HCI this could result in an increased toxic potency. The overall toxic hazard may or may not be increased, since (inherently) fire-retarded materials tend to have an improved fire performance.