Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
SESSION H: Fly PE0O~CE PETER SESSION OBJECTIVES Determine the fire performance parameters that need to be considered for specifying matenals and on what they can be based. PARTICIPANTS Chair: Richard Gann, National Institute of Starboards arm Technology Committee: Frederick Dryer, Princeton University Howard Emmons, Harvard University Patricia Tatem, Naval Research r~aborato~y Participants: Toni Arnold, Federal Aviation Administration Steven Beare, DuPont Robert Buch, Dow Corning Dougal Drysdale, University of Edinburgh Sally Hasselbrack, Boeing Commercial Airplane Group Vahid Motevalli, Worcester Polytechnic Institute Thomas OhIemiller, National Institute of Standards and Technology lames Quintiere, University of Maryland Gus Sarkos Federal Aviation Administration Usman Sorathia, Naval Surface Warfare Center SESSION REPORT Many of the session participants believed that the measures of fire performance of aircraft interior materials should be consistent with an overall fire-hazard analysis. Fire Performance Parameters For current aircraft and materials, the Federal Aviation Administration (FAA) Technical Center has demonstrated that the environment within a passenger cabin becomes untenable as the Ore changes from localized to full-cabin in extent. Delaying this transition results in longer time for evacuation. Should this determination hold for future aircraft, then the Federal Aviation Administration goal of a 50 percent reduction in fire deaths could be achieved by using materials 228
Part 11- Workshop Summary 229 that further delay or even prevent ~flashover."' It must also be established that a lethal or incapacitating atmosphere does not exist prior to flashover. There are conditions that can lead to passengers being unable to evacuate: high temperature, excessive thermal radiation (causing death by burns), and inhalation of toxic gases. The length of exposure to these conditions could be increased by the formation of sufficient amounts of smoke particles and aerosols so that vision is hindered and evacuation is impeded. The potential for the production of corrosive smoke was not addressed. Corrosivity does not affect life safely but may compromise the re-use of the aircraft. The importance of each threat depends on the time frame for its manifestation relative to the timing of other catastrophic events and the time needed for evacuation. Thus, the type of fire determines which types of materials performance measures are valid. To illustrate this point, the participants considered two classes of fire scenarios (post-crash, fuel fires and in-flight fires) and their variations. Plarze on the ground; e~cterr~al Velures; one or more holes ir' the fuselage; only flame radiation enters. Interior materials are subjected to piloted radiation-induced ignition.2 Should a seat or interior pane! begin burning, then the next important event is the subsequent ignition and burning of adjacent entities. A mode! of the ignition and fire growth (especially of a seat/assembly) and the transition to more extensive burning of these fire-hardened materials would relate the discrete materials performance measurements to the system behavior. Some of the important properties can be identified from existing knowledge. The mode! development should be guided by and valida~ using real-scale tests. The rate of heat release is the principal driver of fire growth. This can be measured using a device such as the Cone Calorimeter. For low heat release rates, the oxygen depletion may be too small, and an alternative measure based on, for example, CO/CO2 yield may be necessary. Some participants noted concerns that any bench-scale device may produce artificial phenomena that do not occur at real scale or may miss phenomena that do (e.g., pane! buckling and delamination). These concerns will require some testing using larger samples and eventual real-scale testing to resolve. It is important that typical values of the incident radiant flux be obtained from real-scale testing. According to participants, no current testing device has been shown to give a measure of visual obscuration indicative of the real-scale fire. While the smoke-yield data can be related to visual obscuration via a simple model, it is not yet understood how to use this data in a meaningful way for fire scenarios expected for aircraft. The radiant smoke toxicity apparatus (ASTM E-1678) measures yields of toxic products that have been related to real-scale fires. In that apparatus, the samples are exposed as in this fire scenario. Other toxic potency methods have not been related to real-scale fires. Analysis of safety. ~ A more precise definition of the flashover phenomenon is needed to characterize properly the threat to life 2 While it is presumed that the aircraft interior materials will be fire-hardened, passengers' clothes and carry-on items will be readily ignited by the high flux.
230 Improved Fire- aru] Smoke-Resistant Materials compartment fires has shown that for combustibles of normal toxicity, heat is the initial threat to life safety. Thus, the premium may well be on determining only if the smoke is of extreme toxic potency. An additional hazard is that exposure to the fire's hot upper gas layer may cause the ceiling materials to melt, drip, or fall down. This could cause further ignition of interior materials. There is no current measurement method to characterize the physical stability of the installed panels. Plane on the ground; external fuel fire; one or more holes in the~selage; flames arm smoke eraer. In this case, the upper layer of the cabin is quickly vitiated, and thus the decomposition of the exposed panels is different. This could result in a longer flame extension from any burning materials, as well as smoke of higher toxicity, which results from less complete combustion. In the case of vital upper layer, measurements from the above scenario should be supplemented with the yield and flammability limits of the pyrolyzate. There is no procedure for measuring pyrolyzate yield and flammability limits today. The external flames would drive spread on the ceiling, while fire growth over the seats would likely be less important. Plan on the ground; e~cternai fuel Are; no holes in fuselage. The heating of the airplane skin win eventually heat the back side of the interior wall panels, resulting in some degree of aerobic pyrolysis. At the same time, there is a threat that the flames will burn through the fuselage, resulting in a fire like the two types described above. Representing the former process will require data on the pyrolysis rates under varying thermal stresses, as well as measurement of the toxicity of the pyrolyzate. Since the fuel fires following a survivable crash have (to date) heated only part of the fuselage, participants thought it likely that evacuating through one or more doors would be practical. Thus, effective hull protection that stays in place could provide ample egress time, and a comparison of the time to appreciable heating and degradation of the wall panels with the time for evacuation is important. Measurement of the insulation system quality is thus an important materials evaluation. Plane in~?ight;fire starts within the cabin or lavatory. Participants regarded this as a secondary problem. A small, accessible fire is quickly suppressed with hand-held extinguishers. Plane in flight;fire starts ir' an inaccessible area (cargo hold or behind cabin liningsJ. For such a fire, the tenability of the cabin would have to be maintained for up to 3 hours. Should a suppression system not be installed or should it not work, this would require palette containment materials and cabin isolation materials of extraordinary fire resistance. Fortunately, the air leakage into the cargo hold could be made quite small, limiting the burning rate of the combustibles. Research should be conducted to determine whether such materials are possible.
Part 11- Workshop Summary 231 The current test method for the cabin liners is very severe, involving resistance to an intense burner impingement. Due to increased use and complexity of electronic controls and systems, participants thought that future planes will likely have more electrical cabling running behind the wall panels and under the floors. There is always the potential for ignition in concealed spaces from an electrical fault or an overheated wire. The nearby materials would be exposed to a sustained, but small, hoe spot or flame. Participants thought the f~re-resistance of current materials is probably adequate to survive this threat. Moreover, there has been a lot of work on materials resistance to small ignition sources, so this may not be a research but an implementation issue. Needed Development In Materials Evaluation and Characterization Methods The above sections describe the need for characterization of the finished products in order to enable evaluation of their appropriateness for use on board aircraft. A second series of characterizations are those needed for guiding the development of new materials. Particinants saw this as especially important in the early stages of exploration where only small tgram' samples of the polymer may exist. Of particular importance is understanding how to promote the formation of char during burning. Some present research includes experiments with the use of techniques such as solid-state NMR for analyzing the partially combusted sample and its char. Participants knew of no existing procedure for screening fire properties through tests on sub- gram samples. However, research on thermogravimetry and differential scanning calorimetry coupled with mass spectroscopy is investigating the relationship between rate of heat release and ignition behavior. Participants foresaw a critical advance in the characterization of a material's fire contribution when such an appraisal can be based on the molecular chemistry and thermal embodiment of the product. Such relationships, based on fundamental understanding, will enable efficient screening of new materials design. ~. . ~ At present, all materials properties are measured using "new" samples. Aircraft interior products stay in service for years, undergoing wear and tear, frequent cleaning and maintenance, and general aging. It is important that methods be developed for accelerated aging of new materials and structures, so that tests may be conducted to ensure they will retain their desirable Ore performance throughout their service life. As noted above, participants said that several devices are needed to obtain the full complement of data from burning materials for use in a hazard model. It would reduce the burden of testing and increase the reliability of results if a single device were developed to measure parameters such as heat release and smoke obscuration.
232 Imp roved Fire- and Smoke-Resistant Materials Long-Term Research Participants in the workshop session suggested the following areas for research on fire performance of materials. Develop methods for predicting fire performance of materials from chemical structure. Develop mesons for accelerating materials aging for predicting long-term flammability. Develop a verified computer mode} of ignition of and upward flame spread over low-flammability seals and wall panels. Develop accurate bench-scale methods to generate proper materials data. Develop new, very small sample guidance methods. Develop a validated two-dimensional or three-dimensional mode} of the evolution of habitability of the cabin environment to identify key materials parameters. Develop an understanding of flame spread in a vitiated upper layer and its impact on ignition of fire-hardened seats and wall panels, as well as of clothing and carry on items.
