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.
Flame-Retarding Woo' Textile Materials and the Evaluation of Thermally Stable Polymers for Commercial Airplanes Sally A. Hasselbrack. ABSTRACT Current methods of flame-retarding woo! fibers and the rationale for selecting the low smoke zirconium finish as the basis for decorative applications used within the airplane cabin are discussed. Upholstery and drapery applications for using flame-retardant washable woo! is outlined with recommendations for its usage. An overview of the Boeing methodology for evaluating the inherently flame-retardant and thermally stable polymers with their possible application is presented. INTRODUCTION AND HISTORICAL OVERVIEW The selection of ineeAor textile materials for commercial airplanes that are light in weight, possess the desirable in-service use characteristics, and incorporate all the aspects of fire safety provides a challenging opportunity to incorporate the latest technology into the mainstream of commercial production. The Boeing 757 and 767 models were the first to include design guidelines stipulating the incorporation of materials with limited smoke and toxic gas emission wherever possible. These 1976 guidelines are found in Table I. Prior technology incorporated the use of polyviny~chIoride (PVC) fiber in the warp and a phosphorous-based flame-retardant (FR) finish on the woo} filling yarns as the primary flame-retardane system for upholstery and drapery. Concurrently with the design guidelines, a publication from the International Woo} Secretariat QWS) in the United Kingdom described laboratory research utilizing zirconium and titanium compounds for flame-retarding woo! (Benisek and Phillips, 1977b, c). The attributes found only in the zirconium finish were (a) white or pale shades were not altered by the FR finish, and (b) the finish was adaptable to a low-smoke version, termed "Zirpro." With the cooperation of the Boris Kroll Company, Boeing entered into a small, ronrietarv pilot stubv incorporating the low-smoke Ziroro finish in a commercial production r - - r - - -- .~ ~ - - ~ , - - ~ ~ ~ ~ environment. All of the development work was done on the 757 and 767 flight fleck upholstery because it was the one upholstery remaining constant throughout both programs. The low-smoke zirconium finish was compared at two pH and add-on levels. The pilot program clearly demonstrated that the amount of nylon in the blend was more sensitive in a production environment than demonstrated in laboratory findings. The critical findings were (a) the finish was capable of withstanding 10 dry cleanings, although, a nylon content exceeding 10 percent *Boeing Commercial Airplane Group, Seattle, Washington. 165
166 I Improved Fire- arm Smoke-Resistant Materials TABLE 1 Design Guidelines for Boeing 757 and 767 Models, 1976 Smoke generation at 4 min D', < 50 large areasa <200 small areas Toxic gas emission at 4 min Draeger tubes CO HCN 3,500 150 Flame-spread indent I8 <25 Vertical burns HE HC1 SO2 NOx 50 500 100 100 12-s ignition, vertical orientation all textile materials 60-s ignition, vertical orientation textile wall applications bonded to panel NOTE: Smoke and toxic gas emissions evaluated in the NBS Smoke Chamber at 2.5 W/cm2, flaming mode. aLarge area is greater than 100 sq. in. Flame-spread index pertains to all panel constructions, passenger service units, cargo liners, floor panels, floor beams, and acoustic and thermal insulation. .. CFederal Aviation Regulation 25.853 Amendment 32. compromised the number of dry cleanings; and (b) the FR finishing bath was very sensitive to slight deviations in the quantity and type of acid used and to the pH level of the bath. Based on the results of this study, Boeing Materials Technology recommended that the upholstery/drapery suppliers utilize the low-smoke Zirpro treatment and limit the nylon content to 10 percent. Table 2 and Figure ~ illustrate the quantity and the sequence in which the chemicals are added to the finish bath and the recommended time/temperature profile for their application. The Federal Aviation Administration regulation pertaining to carpet and drapery is a 12-second vertical Bunsen burner test. Upholstery is required to pass the Bunsen burner and the of} burner tests (see below). TABLE 2 Low-Smoke Zirpro Treatments for 100 Percent Wool Upholstery and Carpets (in percent) Upholstery Carpets Formic acid (90%) Citric acid monahydrate Potassium hexafluorozirconate Zirconium acetate solution (22% ZrO2) 10.0 6.0 3.5 7.7 10.0 8.0 2.3 10.0
Sally A. Hasselbrack 70 C/ 160 F 1.5 ° C/min / or 1 onger ~ / ~ 25°C/80°F ~30 ~ IT A B C D FIGURE 1 Application procedure for low-smoke treatments of 100 percent wool. 167 it\ 1-2O C/min ~ ~ 40°C/105 F 30 1 1 5-30 1 ¢~ min E Step A: 0.1 g/1 non-ionic wetting agent, if required. Step B: to percept formic acid, x percept citric acid. Step C: y percent potassium hexafluorozirconate (dissolved). Step D: z percent zirconium acetate solution (diluted with cold water) (bath will become cloudy; it will clear as the temperature reaches 45-50 °C). Step E: rinse for 10 minutes in cold water (do not use overflow rinsing). EFFECT OF THE SEAT FIRE BLOCKING REGULATION The regulation credited with saving the most lives in post-crash fires has been the 1984 Seat Fire-Blocking Regulation, Federal Aviation Regulation (FAR) 25, Appendix F. Part IT, effective November 1987. This regulation was a quantum step forward in flammability testing, for the following reasons: · It used a more realistic ignition source to initiate combustion simulating a post-crash fire (oil burner- 2-minute ignition at 1800 OF, minimum). · It was the first regulation for airplane interiors to evaluate materials in their composite form. · It mandated retroactivity for the entire fleet. In the initial phase, almost all of the fire-blocking layers were woven or needled blends of polybenzimidazole (PBl) Nomex/Kevlar in varying ratios. Two other fiber blends also meet the regulation but do not hold up in service, a spun-laced aramid and a preoxidized carbon/aramid blend. At present, two types of FR foam permit most wool upholstery to pass the test without the need for a blocking layer. WOOL FLAME RETARDANT TECHNOLOGY FOR LOW HEAT RELEASE APPLICATIONS Textile products applied to vertical surfaces (i.e., nser/dado panel and wall tapestries) were included in FAR 25 Amendments 61 and 66, pertaining to low heat release. To meet this
168 Improved Fire- aru] Smoke-Resistant Materials requirement, a tetrabromophthalic acid (TBPA) treatment is used in combination with the Zirpro treatment. The success of the TBPA treatment is due to the synergism between the zirconium finish, which works in the solid phase, and the organobromine acid, which works in the vapor phase. Initially the IWS Onternational Woo} Secretariat) recommended two forms of the TBPA compound: Apex 160, an emulsified paste with a 40 percent bromine content, and FR 756, a liquid form of disodium salt of TBP anhydride with a 15 percent bromine content. Zirpro~?ame- retardant trea~nents references only the FR 756 from Great Lakes Chemical. Table 3 and Figure 2 identify the FR formulation, the sequencing of the flame-retarding chemicals, and the suggested time/temperature model. As of this writing, the Boeing company is nearing the end of a test program using various chemical ratios in the TBPA finish in combination with different backcoating systems to reintroduce woo} as the riser material. Collaborative work with a vendor and an FR finisher for over ~ year has sought to identify a formulation that is repeatable and reduces both the heat release and the smoke numbers to a level that makes the fabric suitable for all pane! constructions where the riser fabric can be placed. Two woo! constructions have been identified: TABLE 3 Treatment for Wool-Rich Materials Required to Meet the Ohio State University Rate of Heat Release Test (in percent) Formic acid (90%) Citric acid monobydrate Potassium hexafluorozirconate Zirconium acetate solution (22% ZrO) FR 756 8.0 6.0 3.5 7.7 10.0 60 C/140°F 1.5 ° C/min or longer' 25 ° C/80 ° F ~ _ __ _, 5 10 5 5 A B C D 25 1 30 1 1 0-20 1 40 C/105 F ¢~ man E F FIGURE 2 Application procedure for low-smoke Zirpro + FR 756 treatments of wool-rich blends to meet the Ohio State University heat release test. Step A: 0.1 g/1 non-ionic wetting agent, if required. Step B: x percent hydrochloric acid, 6 percent citric acid. Step C: 3.5 percent potassium hexafluorozirconate (dissolved). Step D: 7.7 percent zirconium acetate (diluted with cold water) (bath will become cloudy; it will clear as the temperature reaches 45-50 °C). Step E: y percent FR 756 (dissolved in cold water). Step F: rinse for 10 mmutes in cold water (do not use overflow rinsing).
Sally A. Nasselbrack 169 a repp weave with 100 percent woo} wrapped around a Nomex core with a Nomex filling yarn; and a low-level-Ioop grospoint construction of 100 percent woo! face pile with a Kermel/Panox backing yarn system. For wall mural/tapestry applications, the only TBPA-treated woo} fabric identified to date is the grospoint. Grospoint material does not have the design capability (Ion" fibers in multiple yarn counts and complex patterns) the Boeing design staff prefers. Test results in Table 4 show that ache ratio of cut to loop cannot exceed! 40 percent without exceeding the pass/fai! criterion of 65/65; and jacquard patterning is not possible (Hasselbrack, 19901. The application of a thin aluminum foil between the fabric and the panel, or some other "work-around," might somewhat expand the patterning capability, but such remedies would not meet the needs of the design staff. TABLE 4 Heat Release and Smoke Data for Textile Wall Applications Heat Release Smoke Construction 2-min, kW · min\m2 Peak, kW/m2 D8 at 4 min 100% P-84 grospoiIlt, low-level loop 13 43 12 100% wool-face ~rosnoint Kermel/ 31 37 85 Panox backing yarns, all loop pile 100% wool-face grospoint Kermel/ 36 33 122 Panox backing yarns, 50% cut-50% loop 100% wool-face grospoint Kermel/ 52 48 161 _ Panox backing Yarns. all cut pile 100% wool-face grospoint Kermel/ 52 60 192 Panox backing yarns, jacquard pattern, all loop Replin-100% wool wrap Nomex core, 43 53 141 Nomex filling NOTE: Application is to 0.437-inches-thick graphite phenolic sandwich panel. WASHABLE FLAME RETARDANT WOOL The utilization of TBPA is also an essential part of the FR washable woo} program, the latest area of interest within the Boeing company for new flame-retardant developments associate with wool. When the IWS, located in Ilkley, West Yorkshire, United Kingdom (officially renamed Wools of New Zealand in July 1994), first announced a washable FR woo} about 10 years ago, Boeing Materials Technology (BMT) tested their submitted yardage. At that time, shrinkage after the first laundering approximated 5-7 percent, and there was minimal progressive shrinkage. With the strong correlation between dimensional stability and abrasion or "wear," Boeing did not recommend the finish without a yardage wet-out prior to the cutting and sewing process. However, as more Boeing 737s were sold to emerging countnes, Boeing received requests for an FR washable woo} finish. Boeing's Payloads Furnishings contacted several TWS branches around the world requesting an improved finish that would meet the
170 Improved Fire- and Smoke-Resistant Materials shrinkage and appearance requirements of the airline industry. The U.K. branch, working in connection with all the traditional Boeing suppliers, initiated a new study to significantly reduce the shrinkage problem while maintaining a Boat of 15 launderings and 2 percent change in dimensional stability. O O ~~ ~o ~ -~ ~- .r ~- ~ ~~~ ~ ~ ~~-~o ~ Wools of New Zealand (WNZ) states that any type of shrink-resist process must be compatible with the Repro process because it is the only flame-retardar~t treatment for woo! durable to washing. Their suggested options for the shrink-resist treatment of woo} include (a) chionne/Hercosett, an established process applied to the fiber in the worsted tops form; and (b) the combination of two polymers, Synthappret BAP/Neoprene 400, applied in equal amounts to He woven, dyed fabric by a padding process. With either method, the flame-retardant finish is applied after the shr~nk-resistant finish. As of this writing, some of the WNZ technical staff state a preference for the BAP process due to better appearance retention and abrasion resistance as measured by Martingale. It is expected that the BAP process would be slightly more expensive because more processing steps are required. The technical literature states that the standard zirconium finish is less durable in combination with the shnnk-resistant treatment than is the low-smoke version, and recommends the FR 756 to enhance washing fastness (WNZ, 19941. Test data showing burn length and extinguishing time as related to the number of launderings are available. A brochure supplied by WNZ specifies in some detail the finish application and the precise laundering instructions. With this information, interested carriers can obtain samples of fabric with the FR washable finish and conduct internal tests to ensure their woo! material is conducive to the procures to which the fabric will be subjected. The WNZ recommends that the weight of the fabric not fall below I l oz/yd2, as they have found that the number of washing cycles is related to the fabric weight. Additional guidelines regarding the types of material that are suitable for the additional processing include: at ~ oz/yd2, only 100 percent wool should be considered; and both the shrinkage and appearance retention are improved with a tightly woven fabric. At 14 oz/yd2 WNZ finds that the materials can withstand up to 19-20 launderings; the number of launderings can be extended if 100 percent wool is selected; and · 10 percent nylon can be added to the blend (Winterburn, 19941. Several vendors have convinced Boeing customers that a 10 percent level of nylon will enhance the service life of the upholstery. There are probably no good data to prove the fact one way or the other; nylon does, however, enable vendors to operate looms at a faster rate, which affects price, as does the presence of the nylon. FLAME RETARDANT AND THERMALLY STABLE FIBERS Boeing Materials Technology (BMT) considered six inherently flame-retardant or thermally stable fibers as candidates for airline application: CS Trevira (polyester), Nomex
Sally A. Hasselbrack 171 (aramid), P-84 (polyimide), PB! (polybenzimidazole3, Kerme! (polyamide-imide), and FR Viscose. The rayon and polyester fibers contain FR additives that make them acceptable for airline usage, and both are readily available in a large palette of ultraviolet stable colors. Experience indicates that uses of FR Viscose within the airline industry is principally as a drapery material. While it was initially tried as an upholstery material, BMT found that the fiber cost was comparable to wool, the smoke numbers were equal or higher than wool, and the fiber did not have the abrasion resistar~ce of the woo! (Hasselbrack, 1980~. The fiber producer Died altering the drawing of the molten polymer to increase the crystallinity or a scouring of the surface of the fiber to impart greater intra-yarn cohesiveness. A blend of woo} and FR Viscose was tried, but there was no market niche for such a blend. At present, the greatest volume of FR Viscose is in the protective clothing market in blends with thermally stable polymers. CS Trevira (FR polyester) The Hoechst (Germany) CS Trevira fiber has penetrated the drapery and upholstery airline market in countries where dry cleaning facilities are nonexistent or very expensive. With the import tax into the Unit States, the fiber approximates the price of wool. The following details which must be considered when using Trevira upholstery: (a) only the texturized CS Trevira can be used because of the pilling propensity; (b) the finished fabric must be subjected to a carefully controlled singeing treatment to minimize pilling; (c) the fabric must either be run through the tenter frame in a relaxed state to minimize relaxation shrinkage, or be given a post- loom scouring; and (~) CS Trevira has difficulty passing the Park Oil Burner test unless the fire blocker contains a high ratio of PBl. With PB! approximating a cost of $70 per pound, the fire blocker becomes costly and overshadows the savings realized by eliminating dry cleaning. However, two types of FR foams enable the CS Trevira to pass. While the airline industry is currently focused on CS Trevira, the U.S. division of Hoechst Celanese is beginning to pursue the airline market with their low-pill Trevira FR type 370, a staple yarn. The Trevira FR differs from the CS Trevira in two ways: (~) special additives in the monomer are designed to minimize the pilling problem; and (2) there is less of the FR additive in the U.S. polymer (O'Connor, 19941. As of this writing, the U.S. branch is just initiating work with the airline upholstery and drapery suppliers. Nomex (araniid) In the early days of the Boeing 747 (1968-1969), Nomex was used for carpeting and upholstery. These materials showed very poor color fastness in service and had to be replaced. However, two or three carriers continued to use Nomex carpet in three or four colors for about 10 years. Out of curiosity, a Nomex carpet was tested in the NBS smoke chamber for comparison with a BMS Hoeing Material Specification) qualified woo} carpet. Observers were amazed at that time that the FR woo} pile carpet with the appropriate backcoating produced less smoke than the Nomex pile carpet of comparable weight. With today's experience, the
172 Improved Fire- aru] Smoke-Resistar~t Materials composition of the backcoating would be among the first questions asked. There is, however, tilde incentive to ask manufacturers to exert much effort to improve the smoke emission of a carpet: multiple investigations of crash incidents demonstrate that the carpet is almost never involved in the fire. Since 198S, Nomex has been used as the standard rise/dado pane! fiber, replacing wool, which initially could not pass the Low Heat Release Rule. At Boeing, our experience demonstrates that Nomex is a fiber plagued by serious dye-Iot inconsistencies, a limited color palette, and ultraviolet instability. In addition to the greater cost per pound (about three times greater than wool), the CGF Nomex fiber requires solution dyeing, hence, a large minimum run. We find it difficult to obtain subsequent production lots with a color match close enough to enable placement of yardage from different production lots adjacent to each other. The color matching is done in the fiber state, so traditional instrumented color matching equipment is not entirely usable even though a common h~ber-combing technique for color evaluation was worked out with DuPont. At present, color approvals of the dyed fiber are conducted under standard lighting conditions by the human eye. At times, the woven yardage mace from the approved staple fiber Hfluff' appears substantially off-shade, and a new fiber production run is required. Furthermore, our customers complain about service "wear" issues and have filed warranties against Boeing. Two common complaints are (a) fiber splaying and fabric separation with minimum rubbing in the area where the fabric is folded over the top of the riser panel; and (b) a tendency toward some fabric raveling where it is rubbed by briefcases, shoe scuffing, or other abradants. DuPont is currency marketing an alternate Nomex fiber, Thermacolor. While there are no data to suggest that Thermacolor incorporates improved abrasion resistance, DuPont asserts that the fiber is feasible for risers and wall tapestries because (a) the fiber can be dyed at a commercial dye facility, thereby eliminating the cost of minimum runs; and (b) Boeing color approvals would be possible in the woven, tufted, or finished state. While there is no alternate fiber that can produce the white and light paste! colors desired for wall murals, the P-84 polyimide fiber, discussed next, has an ultraviolet stable color palette of over 50 colors. P-84 (polyimide) At the present time, the P-84 fiber is recommended for drapery and riser applications. It is available in multiple yarn sizes so it can be used in a wide variety of end uses, especially if blended with wool or cotton. Two drawbacks presently limit its usage: (~) it is a solution-dyed fiber with a minimum run of 250 pounds; and (2) it has poor abrasion resistance unless it is used in a blend. Except for research evaluations, the manufacturer freezing) has not been asked to produce fiber for the airplane industry; but the fiber is widely used in the protective clothing industry in blends with Kevlar, PBI, and FR Viscose. PB! (polybenziniidazole) Because PB! can withstand the highest temperatures of all the thermally stable fibers before decomposing, it is an excellent fire blocker, especially as a needled felt in combination
Sally A. Hasselbrack 173 with difficult materials (e.g., leather upholstery). The PBI begins to break down when exposed to the light; the fiber is difficult to stabilize and shrinkage becomes a problem in garments; it does not have the strength of P-84; and the palette is limited to two or three harsh colors. There is a substantial cost differential between the two fibers: PB! costs approximately $70 a pound; P-84 costs about $22 a pound. Kerme! (poly~iide-~iide) Kermel has not penetrated the airline textile market. An effort to stimulate interest among the fiber producers in Lyon in airplane applications led to the conclusion that the fiber does not have a unique application within the airplane. The fiber is costly to produce in small airplane quantities; and, to date, the color palette is limited to 14 solution-dyed colors. Kermel blended with FR Viscose is widely used in the protective clothing field because of the comfort due to . . . . .. . ~ ~ . ~ . . . . . good moisture absorption, and softness, and the fire protection provided even in a lightweight construction (7.5 oz/yd21. A promotional brochure ("Kermel 7.5 oz. Argumentation: A Rhone- Poulenc-Amoco Fibers loins Venture") states that if Kermel is used in blends, it must be used with fibers that do not require dyes with "carriers" for the carriers have resulted in an unsafe garment condition due to fiber wicking. SU10IARY AND CONCLUSIONS T have tried to provide a comprehensive status overview of all the wool flame-retardant finishing technology used within the airline industry today, and the growing interest by the airline customer toward a washable product, which includes washable FR wool. A review of the advances that have improved fire safety within the passenger cabin over the past 15-20 years is an occasion for feeling a sense of pride and accomplishment. Boeing will continue our quest for materials that will enhance airplane fire safety. It is our belief that wool will remain the primary fiber for decorative applications because it possesses (~) an outstanding acceptance of dyestuffs; (2) inherent resistance to burning; (3) ease of flame retardance; and (4) a great capacity to tolerate the "wear and tear" of service life. It does not appear feasible that thermally stable fibers will become a significant entity within the passenger cabin because of their large minimum quantities, low abrasion, ultraviolet instability, color limitations, and high cost. It remains to briefly address the controversial subject of nylon carpet. Each year more airlines elect to install nylon carpet because of the "throwaway" nature of airline carpeting. The entryway and aisles have to be replaced within 3 or 4 weeks, in spite of attempts at good maintenance. Nylon's low cost, coupled with a longer wear life in a lightweight construction, cannot be matched in wool. Because nylon fiber itself is not manufactured as an inherently flame-retardant polymer, a PVC compound must be placed in the backcoating to enable the carpet to pass the 12-second vertical Bunsen burner test. Insofar as accident investigations demonstrate that the carpet is rarely involved in the fire, there is no need for great concern about the quantities of smoke produced in He smoke chamber when comparing nylon to woo} carpets. It would be more realistic to change the flammability test for carpet to a horizontal position, for
174 Improved Fire- arm Smoke-Resistant Materials that orientation represents its actual use; also, the high smoke producing backcoating could be replaced with the types of backcoating used on woo} carpets- viny} acetate/ethylene copolymer or styrene buladiene with alumina Dehydrate as the filler for both compounds (Benisek and Phillips, 1977a). REFERENCES Benisek, L., and W.A. Phillips. 1977a. The effect of backing fibre and latex type on the burning behaviour and smoke emission of woo} carpets for aircraft interiors. Journal of Fire and Flammability 8:516-531. Benisek, L., and W.A. Phillips. 1977b. The effect of flame resistant treatment on smoke emission of woo} upholstery fabrics. Journal of Fire and Flammability 8:458-477. Benisek, L., and W.A. Phillips. 1977c. The effect of pH on the smoke emission of wool. Journal of Fire and Flammability 8:247-254. Hasselbrack, S. 1990. Intenor Cabin Materials for the 1990's: Resins to Decorate Textiles. Paper presented at the 1990 Conference Recent Advances in Flame Retardancy of Polymeric Materials, Stamford, Connecticut, May. Hasselbrack, S. 1980. Textile materials for commercial transportation vehicles. Pp. 816-828 in 12th National SAMPE Technical Conference. Covina, California: Society for the Advancement of Materials and Process Engineering. O'Connor, G. 1994. Personal communication to S. Hasselbrack, August 27. WNZ. 1994. Options for the Manufacture of Washable and Flame Retardant Aircraft Upholstery. Technical Information Letter, Section 2.2. Ilkley, West Yorkshire, U.K.: Wools of New Zealand. Winterburn, S. 1994. Personal communication to S. Hasselbrack, August 20.
