disaster response can composites prevent this? composites prevent this? can composites ... courtesy...

Construction, Corrosion, & Infrastructure Conference • April 25-27www.acmanet.org

Disaster Response

February 2007

Qualify for a free subscription at www.cmmagazine.org

Can Composites Prevent This?

Can CompositesPrevent This?

Disaster Response

Construction, Corrosion, & Infrastructure Conference • April 25-27 • www.acmanet.org

20 | Composites Manufacturing | www.cmmagazine.org | February 2007

By John Rowen

ultrusion is a continuous composite manufacturing process that is efficient and

economical for the production of high-strength structural components that

compete directly with steel and aluminum parts. For many markets, composite

parts have two major advantages over steel and aluminum—high strength-to-

weight ratio and corrosion resistance. For example, the U.S. Navy is accelerating the incorpo-

ration of composite materials into the fleet in order to both reduce weight and decrease mainte-

nance. Lower weight translates to improved ship stability, speed, range, payload capability and

fuel efficiency. However, one serious concern is the flammability of composite materials.

Whether a fire is accidental or the result of hostile attack, those on board can’t simply

withdraw as if they were exiting a building. To address this concern, composite manufacturers

have turned to bromine compounds as additives to their resins, with the result that flammabil-

ity is reduced, but smoke production is increased. Other flame retardants are better at reducing

Low Smoke PolyesterPultrusions at Low Cost

Surface Mat

P

FLAMERETARDANT

Official Magazine of the American Composites Manufacturers Association | 21

smoke, but they are less efficient and must beadded in such great quantity that the strengthof the part is compromised.

There is a novel method of fire hardeningfiberglass composites using a fire-retardant-coated veil as the outer layer of a laminationschedule. Since only the surface is coated,the strength characteristics of the part remainthe same, while fire is blocked from penetrat-ing the interior of the part. Upon exposure toopen flame or high radiant heat, the resultantmanufactured product has surface flamma-

bility reduction comparable to a brominatedresin manufacture, but with much less smokeand toxicity. For example, several Navydevelopment programs that have benefitedfrom the incorporation of this fire retarding(FR) and smoke-suppressing veil are theElectromagnetic Aircraft Launch System(EMALS)(Fig. 1) for aircraft carriers, theLight Weight Composite Stanchion (Fig. 2)for cargo ships, and the Very Light WeightHatch (Fig. 3).

The Problem DefinedPultruded polyester (PE) components arewidely used in a variety of building applica-tions such as I-beams and C-channels, tubesand angles used for walkway and handrailsystems, decking, ballistic panels, and cabletrays. Pultrusions are often specified for appli-cations requiring high strength and lightweight, but in particular for use in corrosiveenvironments.

However, upon exposure to open flame orhigh radiant heat these materials can exhibithigh surface flammability, smoke generationand toxicity (FST) characteristics. This is aparticular problem when the end use is withina partially or entirely enclosed environmentrequiring egress. This fire hazard is subject toregulated life safety performance criteria set bylocal, state and Federal code governingauthorities.

Current Solutions: Bromineand ATHTo address surface flammability, PE resinscan be adjusted to have a significant degreeof fire resistance (FR) by combining themwith FR additives. Bromines, e.g., decabro-modiphenyl ether oxide (DBDPE), are themost effective FR compounds in general use,especially when combined with the synergistantimony trioxide (ATO). However, under afire load these brominated compoundsproduce large volumes of acrid particulatesmoke and soot. Additionally, brominatedFRs liberate decomposition byproducts suchas hydrobromic acid (HBr) and othernoxious species that can render theimmediate environment biologically toxic,causing acute eye irritation and eventualasphyxia. Less toxic additives such asaluminum trihydrate (ATH) are preferable

Fig. 1 The Electro-Magnetic Air Launch System (EMALS) design is a robust,highly reliable aircraft carrier launch system that will meet or exceed all Navy

performance goals.

Courtesy of General Atomics

from a toxicity standpoint, however therequired high loading levels significantlydegrade the desirable physical characteristicsof the pultrusion.

The Problem of SmokeSuppression: Acrylics andPhenolicsClearly, smoke suppressant additives for PEresins have not evolved to match the effective-ness and sophistication of the fire retardingadditives. The difficulty of chemicallysuppressing smoke created by the combustionof a cured resin is exacerbated by the complexinteraction between the resin and the flameretardant agent. The total complexity hasstymied the development of any universalsmoke suppressing additive options. Somesuccess has been achieved with thermosettingmodified acrylic and phenolic resins, however,these resins have their own set of issues. Suchas very high hydrated mineral, e.g., ATH,loading for acrylic, and voids due to liquidvapor byproducts in phenolic. Compared toPE, composites comprised of these resins havea reputation of exhibiting brittle physicalcharacteristics.

What About IntumescentCoatings?Painting PE laminates with intumescent FRcoatings can dramatically reduce the ability ofa laminate to combust or generate smoke.Intumescents are a family of fire protectivecoatings containing constituents that reactunder high heat to form a protective charredlayer. Once the under-laminate is shut offfrom the atmosphere by a sturdy, carbonificbarrier, the combustion process is greatly

Fig. 2 Plunger actuated composite stanchion for the three hold types:Freeze/Chill (F/C), Specialty Cargo, and General Stowage.


Fig. 3 The Very Light Weight Hatch(30” x 60”) can be operated by one

individual as opposed to three for thesteel hatch, thus eliminating the

requirement for a secondary scuttlehatch. Tested to the UL 1709

protocol.

Courtesy of General Dynamics / NASSCO Courtesy of KaZak Composites

Figure 4. Intumescent Coating ASTM E-662 Standard TestMethod for Specific Optical Density of Smoke Generated bya Solid Material OPTICAL DENSITY TEST RESULTS SUMMARY

NON-FLAMING FLAMING

Ds 1.5 min. average: 0.6 3.9

Ds 4.0 min. average: 0.4 30.1

Dm(corr)(20.0 min.) average: 4.7 178.3

Figure – 4 contains the results of an ASTM E-662 test on a non-fireretardantPE resin / balsa / glass sandwich panel that had an Avtecintumescent coating applied to it. Generally, if the non-flaming and flamingaverage values at 20 minutes are added together and have a result below200, the specimen will likely have an ASTM E-84 Class A Smoke of <450.

Figure 5. 50 kW/m2 Cone Calorimeter Test Results Comp-aring Three FR Mat Specimens to a Brominated Specimen

Sample /Test # Placard Mass Correlated Correlated Duration of Peak Heat Initial Correlated Total Average Test (s) (kW/m2)Mass(g) Conversion Smoke Smoke Release

Factor Release(g) Production(g/m2/s)

Sample 1 “glass out” 80.7 1.00 20.01 .017 1198 91(glass against die)

Sample 2 “FR out” 81.6 .99 18.58 .019 1058 123(FR against die)

Sample 3 FR mat sandwich 82.2 .98 14.95 .011 1349 110(two FR mats)

Pultruded Bromine PanelPolyester Veil 40.6 1.99 52.26 .140 465 185

Note: the Brominated panel at the bottom of the chart released a 261% greater quantity of smoke (corre-lated 52.26gm) compared to the FR Mat – Sample 1 (20.01gm). Samples 2 and 3 illustrate contrastingresults when the FR Mat is positioned in alternative orientations. Sample 2 had the FR side positioned asthe surface i.e., glass rich side down. Sample 3 contained two FR Mats with the FR sides back to backforming a sandwich construction.

Cut Costs & Turn Timesfor Lay-Up Tools

Patented, CMPT™ lay-up mold technology for large,contoured composites slashes tooling, handling and

energy costs. Custom made from any alloy at our fourfacilities. Drastically reduces mold-making time and

costs. Ideal for complex-shaped reinforced composites.

A General Magnaplate Division

1-800-441-6173 • [email protected]

slowed. This layer both reduces surfaceignitability, and reduces smoke by entrap-ment of the resultant smoke particles. Fig. 4shows the results of an ASTM E-662 (“TheStandard Test Method for Specific OpticalDensity of Smoke Generated by a SolidMaterial”) on a PE balsa cored sandwichpanel coated with a fire-retardant paint. Theresults indicate low smoke productionbecause the longer the exposure time, themore smoke and soot is packed into thecoating, thus further shielding the laminatefrom ambient oxygen. This shielding resultsin a slow charring of the substrate rather thana more flammable, vigorous oxidation.

However attractive from a life safetyviewpoint, a secondary over coating onpultruded parts has numerous drawbacks. Itcan require code regulated paint applicationand drying areas. Spray application necessi-tates over-spray, often resulting in a loss ofmore than 30 percent of coating material.Coatings can be scored, scratched or chippedat any point during the coating’s life cycle.For all of these reasons, the coating option hasnot received much acceptance.

New Solution: Deliver Intumes-cents on a Surfacing VeilUsing a veil or mat as a transport medium todeliver intumescent FR and smoke suppressingconstituents to the surface of a pultrusionduring a manufacturing process is a noveloption. Surfacing veils, both glass and organic,have traditionally been employed to provide aresin-rich surface that enhances and maximizesthe corrosion resistance and life cycle of apultrusion. Augmenting that surfacing veilwith FR agents on the underside provides aninter-laminate FR envelopment comparable toan intumescent secondary coating. Since theFR material is “wet out” by the PE resin, itbecomes integral to the finished part, andcannot peel or chip. However, there remainsthe question of whether the PE resin on thesurface will provide too much fuel to allow thefinished part to pass a fire test. This and otherissues were investigated by actual burn testingin the fire lab at Worcester PolytechnicInstitute (WPI).

Cone Testing the Performanceof Intumescent-Coated MatsThe Cone Calorimeter was used as thescreening test apparatus because of its conve-nience and repeatability. Although the Coneinstrument results don’t predict specific E-84results, better results on the Cone tests arehighly correlated with better results on E-84tests. ASTM E-1354 Cone Calorimetryenables fire safety engineers to measure thetotal quantity of particulate smoke released by

weighing each specimen during exposure tohigh radiant heat. In addition, the apparatuscalculates smoke release over time, andmeasures peak heat. Each specimen wassubjected to 50 kW/m2 of radiant heat.

The test run described in Fig. 5 comparesfour samples; three are various configurationsof FR mat-protected vacuum-bagged samples,and one is a bromine-protected pultrusion.The FR mat has both a glass rich surface andan intumescent surface, so it was decided totest with 1) Glass side out, 2) Intumescentside out, and 3) Two veil layers with the

intumescent sides touching. For reasons ofpultrusion die wear, it is desirable to run withthe glass side against the die, so the three testswere chosen to determine whether one wouldsacrifice fire performance by choosing thatorientation. The mass of Sample 1, glass sideout, was designated to be the baseline for masscorrelation. As can be seen in Fig. 5, Sample1 did indeed perform well, with a very lowsmoke and peak heat as compared to thebrominated panel, and comparable to theother FR mat samples. Even though thevacuum bagged specimens had lower glass


content than the brominated pultruded placard, e.g., 41 percent verses52 percent, what the table clearly illustrates is a dramatic reduction inthe propensity of the part to generate smoke for all FR mat specimens.Based upon these and many other Cone tests, the results were deemedsufficient to justify the fabrication of pultruded test panels andformally test them to the E-84-01 protocol.

E-84 Comparison PE Pultrusions Protected byFR Mat Versus Bromine The FR mat test specimen was an assembly of four 1-foot by 12-footpultruded panels butted together to form a 2-foot by 24-footspecimen. The bromine-protected specimen comprised two 12-footpultrusions 2 feet wide, butted end-to-end. The tests were run atSouthwest Research Institute. Figs. 6a and 6b show the E-84-01results for those two specimens. While both specimens had nearlyidentical flame spread indices (FSI), the bromine-protected panel hada smoke developed index (SDI) more than 250 percent higher thanthe FR mat panel. In fact, the accompanying graph shows the bromi-nated panel producing nearly total visual obscuration throughout thetest. The FR panel actually under performed its potential, since at 3.5minutes into the test the panel strips buckled slightly and the burnerflame penetrated the center seam of the assembly, combusting theback of the panel which was not protected by the FR Mat. Smoke wasobserved entering the tunnel chamber through the center seam, origi-nating from the unprotected, organic veiled surface. At 8 minutes thephenomenon and obscuration subsided. Despite the sizable penaltythe panel seams created, the specimen easily passed the Class A smokerequirement. Initial concern about the burn-off of the PE resin on the


Figure 6a. Brominated Panel ASTM E-84-01Smoke Developed Index Result

Figure – 6a contains the results of an ASTM E-84-01 test resultpreformed upon a commodity, PE pultruded flat sheet containingDBDPE.

surface of the FR mat were confirmed by the early “spike” in the Fig.6a graph, but that burning was only momentary. The net comparisonof graphs in Fig. 6a and 6b clearly illustrates that the FR Mat dramat-ically reduces the capability of PE pultrusions to generate smoke whenexposed to open flame.

In summary, the fire and smoke testing presented in this articlevalidate that a low cost, general purpose pultruded product can meetnot only the flame spread requirement of E-84 Class 1, but addition-ally satisquare-feety the Class A low smoke requirement. Thiscombined Class 1/A certification is a prescribed requirement by regula-tory code governing authorities for enclosed environments requiringlife safety egress. Additionally, the testing results suggest that new andpromising opportunities in markets dominated by aluminum andsteel, such as infrastructure, e.g., bridges, tunnels, electrical transmis-sion, marine terminals; energy, e.g., mining, gas and oil; and, trans-portation, e.g., rail, truck, ships, are now open to low-smoke pultrudedproducts.

Subsequent application tests have verified that the FR mat willprovide similar results when processed by filament winding, vacuuminfusion and reinforced thermoplastic consolidation. For more detailsof these and other test results, you can log on to www.avtecindustries.com.

John B. Rowen is executive vice president & technical director at AvtecIndustries in Hudson, Massachusetts. A graduate of the University ofMassachusetts with a B.S. in Chemistry, Rowen worked for AKZO Nobel andEnduro Composite Systems. 978.562.2300; [email protected].

CM


Figure – 6b contains the smoke obscuration result, SDI, of an ASTME-84-01 derived from a low cost, PE pultruded flat sheet containingATH as well as the resultant FR Mat. The % Light Obscuration at notime exceeds 50%. The FR Mat provided the necessary smokesuppression to achieve an SDI Class A rating.

Figure 6b. Non-Brominated Panel ASTM E-84-01 Smoke Developed Index with FR Mat

disaster response can composites prevent this? composites prevent this? can composites ... courtesy...

Documents