By Gail Porter,2014-05-10 20:43
10 views 0

Ignition, Combustion, Toxicity, and Fire Retardancy of

    Polyurethane Foams: A Comprehensive Review

     12Harpal Singh,A. K. Jain

    1Central Building Research Institute, Roorkee 247667, India

    2Indian Institute of Technology, Roorkee 247667, India

Received 21 March 2008; accepted 21 August 2008

    DOI 10.1002/app.29131 Published online 17 October 2008 in Wiley InterScience (

    ABSTRACT: This review provides insight into the igni- incorporation of phosphorus-containing compounds, hal- tion, combustion, smoke, toxicity, and re-retardant per- ogen-containing compounds, nitrogen-containing addi- formance of flexible and rigid polyurethane foams. tives, silicone-containing products, and miscellaneous This review also covers various additive and reactive fire- organic and inorganic additives. Some heat-resistant retardant approaches adopted to render polyurethane groups such as carbodiimide-, isocyanurate-, and nitrogen- foams fire-retardant. Literature sources are mostly techni- containing heterocycles formed with polyurethane foams cal publications, patents, and books published since 1961. also render urethane foams re-retardant. Fire-retardant It has been found by different workers that polyurethane additives reduce the flammability, smoke level, and toxic- foams are easily ignitable and highly flammable, support ity of polyurethane foams with some degradation in other combustion, and burn quite rapidly. They are therefore characteristics. It can be concluded that despite many required to be fire-retardant for different applications. significant attempts, no commercial solution to the fire Polyurethane foams during combustion produce a large retardancy of polyurethane foams without some loss of quantity of vision-obscuring smoke. The toxicity of the V 2008 physical and mechanical properties is available. combustion products is much higher than that of many Wiley Periodicals, Inc. J Appl Polym Sci 111: 11151143, 2009 other manmade polymers because of the high concentra- Key words: additives; halogenated; thermogravimetric tions of hydrogen cyanide and carbon monoxide. Polyur-

    ethane foams have been rendered re-retardant by the analysis (TGA)

    INTRODUCTION greater usage in mattresses and furniture cushion-

    ing. On the other hand, rigid polyurethane foam is The commercial development of polyurethane foams produced in a lower volume (28%) and nds appli- was first studied in 1937 when Otto Bayer found cations in transportation, carpet underlay, refrigera- that the reaction product of diisocyanate and polyol tion technology and appliances, building and 1has properties that make it a polymer of interest. construction industries, the automotive industry,

    packaging, and sporting goods.Polyurethanes are extremely large and complex 2 polymers produced by the reaction of isocyanate Polyurethane foams, being highly cellular poly- (RAN? C? O) with compounds containing at least mers, are easily ignitable and highly flammable. The two active hydrogen atoms (RAOHA). A typical poly- ammability of polyurethane foams has long been a urethane foam structure may contain, in addition factor that limits their greater uses. The fire retard- to the urethane linkages, aliphatic and aromatic ancy of polyurethane foams is mostly required in hydrocarbon, ester, amide, disubstituted urea, biuret, mattresses, furniture cushioning, packaging, and

    building and construction industries with typical allophanate, isocyanurate, uretidione, and carbodii-

    applications in insulation boards, light-weight con- mide groups. Polyurethane foams are the most im- crete blocks, wall blocks with integrated insulation, portant thermoset polymers and are manufactured curtain wall construction, preformed rigid panels, in large quantities in the form of flexible and rigid spray-applied wall construction, and many other foams. The worldwide demand for polyurethane industrial applications.foams has been estimated to be about 5% of the total 3 world consumption of plastics. Flexible polyurethane There are a few approaches for enhancing the re foam is produced in a large volume (48%) and finds retardancy of polyurethane foams: (1) the incorpora- tion of fire-retardant additives into the foam compo- nents by simple mechanical mixing at the Correspondence to: H. Singh ( compounding stage; (2) the addition of fire-retardant

    compounds containing functional groups, particu- Journal of Applied Polymer Science, Vol. 111, 11151143 (2009) larly hydroxyls, which become chemically bound in V 2008 Wiley Periodicals, Inc.


    TABLE I the polymer chain; and (3) coatings on the top sur- Dissociation of Polyurethane Foam Linkages face of the flammable foam by means of fire-retard- at Different Temperatures ant materials. In the first approach, the additives

    Dissociation usually adversely affect the physical properties of Linkage temperature ( C) the foam. A decrease in the closed-cell content and

    strength properties and an increase in water absorp- Allophanate 100120 tion often occur. A signicant reduction in strength Biuret 115125 Urea 160200 properties and dimensional stability generally occurs Urethane 180200 under humid conditions, particularly if the additives Disubstituted urea 235250 are used in excess of 15 wt %. The second approach Carbodiimide 250280 seems to be superior to the first because fire-retard- Isocyanurate 270300 ant additives take part in the foaming reaction and

    become part of the polymer. The third approach is

    useful only for spray-applied foams for outdoor

    applications, for which low water vapor permeabil- 2IGNITION OF POLYURETHANE FOAMS ity and good weather protection are desired.

    The ignition of polyurethane foams includes all gas- Most known re-retardant additives, particularly

    phase processes that occur between the fuel produc- aliphatic phosphates, cause scorching (discoloration) tion step and the occurrence of a visible hot flame. of polyurethane foams because of the sensitivity of The ignition of polyurethane foams occurs by the the foam components to even low concentrations of interdiffusion of the flammable gases with air.12acids released on decomposition. Aliphatic phos- The

    basic physical and chemical aspects of gas-phase phates are more hydrolyzable than aromatic phos- ignition reactions have been studied by several phates. Thus, aliphatic phosphates tend to be more 13researchers.prone to aggravating scorch than aromatic phos- The ignition of polyurethane foams 4has been extensively studied with heat irradiation phates.Another problem encountered with some sources, thermogravimetric analysis (TGA), small re-retardant additives in polyurethane foams is the pilot flames, and heat apparatus.migration of the additives during processing or 14,15 By using a se-

    ries of selected oven temperatures and measuring long-term use of the polymer, which might lead to a 5times to ignition, one can establish the minimum loss of fire retardancy.Because of the migration and heating rate required for ignition and the initial relatively high absorption of moisture, some fire- decomposition temperature at that rate. To obtain retardant additives can undergo hydrolysis, which such relationships, the analytical tools of TGA, dif- leads to a decrease in the mechanical and physical ferential thermal analysis (DTA), and differential characteristics of polyurethane foams. Thus, the sta- scanning calorimetry (DSC) have been used. These bility, compatibility, migration of additives, effect on provide excellent insight into the reactions when the physical and mechanical properties, smoke and conventional and fire-retardant polyurethane foam toxicity, cost effectiveness, and color stability are samples are heated at a standard rate in air and

    nitrogen atmospheres by qualitatively and quantita- some of the key factors in the selection of re-retard- 6tively measuring the heat absorbed or liberated by ant additives for polyurethane foams.Therefore, the the sample because of a phase change or a chemical choice and selection of suitable fire-retardant addi- change and by indicating the rate and extent of tives for polyurethane foams are rather limited. weight loss.The fire retardancy of polyurethane foams has

    been previously studied and surveyed by many 1618 The thermal degradation process 7,8and its relation to foam ignition have been studied workers.A review of the fire retardancy of polyur- with a variety of analytical and existing re-test ethanes, with an emphasis on commercial flame methods.retardants in use, was published by Weil and 1922 The thermal stability and ignition of a 9conventional polyurethane foam mainly depend on Levchik.Another review of the thermal decomposi- 23the composition.tion, combustion, and fire retardancy of polyur- When a polyurethane foam is

    subjected to heat, various polyurethane linkages are ethanes was presented; however, it was limited to 10broken at different temperatures. The dissociation of publications of 1995 and later.Recently, a brief polyurethane foam linkages at different tempera- review of fire retardants for polymeric foams, cover- tures is shown in Table I. The ignition temperature ing only physical and chemical aspects of intumes- of a polyurethane foam at a heating rate of 5 C/min 11cent fire retardants, was published.This review is is 150 C, whereas at a heating rate of 10 C/min, it is different from the earlier published reviews because 260 C. The minimum heating rate required to ignite it covers the ignition, combustion, smoke, toxicity, the polyurethane foam at the initial weight loss is and fire-retardant additive/reactive approaches of 500 C/min, and the minimum decomposition tem-

    polyurethane foams on the basis of patents and pub- perature at this heating rate is 400 C.24lications since 1961. Using similar Journal of Applied Polymer Science DOI 10.1002/app


     25data for different cellular polymers, Miller et al. ing foam increases the evolution of smoke and toxic

    gases, particularly carbon monoxide and hydrogen suggested a relative ignition hazard scale. According cyanide. to the ignition hazard, polyurethane foams fall

     between polyoxymethylene and cotton but are less 12hazardous than polyacetal and polyoxymethylene. COMBUSTION OF POLYURETHANE FOAMS Measurements of the ignition and extinction limits The combustion of polyurethane foams occurs only were carried out, and it was found that polyur- in the presence of a sufcient amount of oxygen. On ethane foams ignite at a 20% weight loss and are combustion, polyurethane foams produce four types extinguished at an 80% weight loss. The ignition of of products: combustible gases, noncombustible cellular polymers in the glow wire test was charac- gases, entrained solid particles, and carbonaceous 26terized with thermography.It was found that poly- char. These combustion products vary with the foam urethane foams ignite faster than cotton and composition, temperature level, rate of temperature cellulose acetate but more slowly than polyoxy- rise, endotherms, exotherms, and rate of volatile 27methylene. Suzuki et al.studied polyurethane evolution. The heat of combustion raises the temper-

    ature of combustible and noncombustible gases, foam smoldering with a siliconite heater as a heat

    resulting in increasing heat transfer by radiation. source. It was found that smoldering spreads faster The heat transferred from the combustion zone to in the upward direction than the downward direc- the adjacent material produces further decomposi- tion under natural draft conditions. Upward smol- tion and ignition resulting in ame propagation. dering of foams in natural convection can be Polyurethane foams exhibit a highly viscous melt on controlled by the initial smolder process being combustion. Thermal analysis has shown that the changed to an endothermic pyrolysis reaction that glass-transition temperature increases with the precedes the smolder reaction. A polyurethane foam decrease in foam density, but the thermal stability ignited in a special apparatus consisting of a decreases with the decrease in foam density.nichrome wire sandwiched between porous ceramic 35Mor- honeycomb plates shows that downward smoldering

    phological changes that occur during combustion is controlled primarily by the supply of the oxidizer have been extensively studied with scanning elec- to the reaction zone. The oxygen supply and heat 36tron microscopy.In other publications, the com- loss are the main factors that inuence the foam 28bustion of polyurethane foams has been reported: ignition and smoldering.The forced forward smol- experimentation was conducted with a ventilated dering propagation velocity increases with air ow tunnel and conrmed that polyurethane foams are and then decreases with the air ow rate in a foam highly combustible materials.3729The combustibility of material placed horizontally.A polyurethane foam

    polyurethane foams has also been measured with covered with cellulose-based fabric poses a serious 38some standard test methods.TGA, differential smoldering hazard if exposed to a burning cigarette

    thermogravimetry, and DTA studies have indicated because of the low temperature (400 C), which pro- that rigid polyurethane foams decompose in nitro- duces a substantial amount of carbon monoxide.28,30 gen and degrade in air through two and three In other experiments, the ignitability, heat release weight-loss stages, respectively. Foam pyrolysis in rate, effective heat of combustion, and mass loss nitrogen and combustion in air lead to 15 and 0% were obtained by the exposure of polyurethane char residue, respectively. The results indicate that foams under cone calorimetry test conditions. It has the thermal stability of rigid polyurethane foam is been found by different workers that the uniform better in nitrogen than in an air atmosphere.39When heat ux and peak rate of heat release depend to a

    urethane foams decompose in different atmospheres, large extent on the melting behavior and thickness the decomposition rates are almost identical in vacuo, of the foam, which should be limited to 25 mm. in nitrogen, and in air at 250 C; however, at higher Density was found to be the key variable in control- temperatures, the rate of decomposition is highest 31ling ignition resistance.The ignition behavior of in vacuo and lowest in air. Complete weight loss polyurethane foam and fabric mock-up combinations takes place at about 750 C in air but at 950 C in has also been studied under cone calorimetry test nitrogen. The evolution of hydrogen cyanide starts conditions. Covering a foam with a fabric results in at 550 C, and its quantity is almost equal to the a delay in the ignition and peak rate of heat nitrogen content of the foam.1234 release.32,33 Checchin et al.studied the postignition

     behavior of polyurethane foams with a cone calorim-

    eter. The cone calorimetry apparatus allows us to SMOKE AND TOXICITY measure the evolution of carbon monoxide and Requirements regarding smoke density and toxicity hydrogen cyanide, which are considered the major are becoming increasingly stringent because of the causes of fatal causalities during re. The presence increasing interest in re and consumer safety. of bromine and phosphorous compounds in a burn-

    Journal of Applied Polymer Science DOI 10.1002/app



    Decomposition Products of Polyurethane Foams by Mass Spectrometry

    Peak Peak Decomposition Decomposition nomenclature nomenclature product product

    a r Nitrogen Pyridine

    b s Carbon Toluene dioxide c t Methyl pyridine Ethylene d u Methyl pyridine Ethane e v Cyclooctatetrene Water f w Vinyl pyridine Propane g x Benzonitrile Hydrogen cyanide h y Not identied Not identied i z Indene Butyne or butadiene j A Methyl Acetonitrile cyanobenzene Methyl k B Acrylonitrile cyanobenzene Not identied l C Propionitrile Not identied m D Methyl

    acrylonitrile n Benzene E Naphthalene o Vinyl acetonitrile F Isoquinoline p G Not identied Not identied q Pyrrole

Many techniques are being used to control smoke resulting in an increase in the smoke density. Poly-

    urethane foams produce more smoke (1.07.4 mg of and toxicity problems. It is generally accepted that deposited smoke) than rigid polystyrene (1.7), wood, polyurethane foams produce large quantities of wood wool, and phenolic foam but less than poly vision-obscuring smoke during combustion; how- (vinyl chloride) (28.9), acrylics (40.6), and nitrocellu- ever, smoke is mostly generated in the beginning of lose crystals.4042combustion. Herringtonobserved a parallel trend Polyurethane foams produce smoke

    that is double in volume with respect to wood com- for smoke production, mass loss, and heat release by ponents. In the aming mode, a exible urethane putting polyurethane foams in a heat release rate ap- foam produces less smoke than a rigid foam, paratus at Ohio State University. The foam samples whereas in the nonaming mode, the smoke differ- 3(100 150 25 mm) were placed in a horizontal ence is quite low. Hurd243position and exposed to a 1.0 W/cmbackground found that 1 kg of foam

    generates smoke that is equivalent to 12 kg of bitu- heat ux and a 0.18-kW-intensity single-point gas men. The dependence of smoke formation on the ame ignition source. The ignition source was posi- temperature at which the polyurethane foam is tioned to impinge perpendicularly at the center of exposed to pyrolysis and combustion was studied in the foam surface. The production of smoke from a ceramic boat tube furnace at 200500 C in nitrogen polyurethane foams was also estimated with the and air.44placement of the sample in the vertical position. The It was found that the maximum evolution

    of smoke occurs above 650 C and that it contains sample was ignited on a wire gauge with a ame virtually all the nitrogen of the original foams. At from a propane burner placed beneath the gauge, lower temperatures, decomposition proceeds rather and the smoke produced in ame and nonaming slowly to generate a signicant amount of smoke. modes was examined by a light/photocell arrange- Woolley et al.4145ment for optical density.The quantity of smoke studied the combustion of polyur-

    ethane foams in a small ceramic boat inside a silica production remains almost constant up to 10 min in furnace tube for 15 min at 200800 C in air. At 200 the beginning. However, it depends on the density 300 C, yellow smoke is generated, which appears to of the foam. High-density foams produce more be a polymerized or condensed and somewhat free smoke than low-density foams. The results also form of toluene diisocyanate (TDI). At 500 C, the show that re-retardant foams release roughly 5 nitrogen contents start evaporating at about 35% times more smoke than untreated foam. In particu- weight loss. The yellow smoke remains stable up to lar, phosphorous re-retardant compounds reduce 750 C, decomposes at a temperature of 800900 C, the thermal decomposition temperature of foams,

    Journal of Applied Polymer Science DOI 10.1002/app


    and produces cyano compounds together with other HCN formation. At low temperatures at which the 46oxidation of HCN is still negligible, the rate of HCN organic nitriles.In another experiment, the foam generation increases linearly with an increasing per- samples were decomposed in a furnace system centage of oxygen in the gas mixture. At higher tem- under conditions likely to be encountered in re. peratures, the oxidation of HCN becomes The volatile products released from the thermal and appreciable, and its formation rate rises to the maxi- thermal oxidative decomposition of polyurethane mum with increasing oxygen content of the gas mix- foams were collected in a refrigerated trap and were tures. According to Morikawa and Woolley,45identied quantitatively and qualitatively by gas any

    nitrogen-containing material, except nitro com- chromatography/mass spectrometry, ultramicroanal- 47pounds, gives off HCN when heated above 700 C, ysis, and mass spectrometry.Peaks of the decom- and the evolution of HCN is almost equal to the position products of yellow smoke obtained from nitrogen content of the materials. The evolution of the foam samples that decomposed at 850 C were HCN from polyurethane foams has been determined 48interpreted with the data of Cornu and Massot qualitatively and quantitatively at different tempera- and are shown in Table II. The corrosivity of polyur- tures (7001000 C), and its value has been compared ethane foam smoke has not been studied in great with those of other polymeric materials. Woolley et detail, although the effects of some gases are well al.

    known. 43,45 studied the combustion of polyurethane foams

    in a silica tube heated inside a furnace with the max- The toxicity of the thermal decomposition and imum temperature up to 1000 C in air. They combustion products of polyurethane foams has detected benzonitrile, benzene, pyridine, acryloni- been intensively documented. This topic was trile, acetonitrile, toluene, CO, HCN, methyl pyri- 49reviewed in detail by Woolley and Field,who dine, butadiene, propane, and water at relatively found that typical pyrolysis and combustion prod- high concentrations, and COucts from exible and rigid polyurethane foams do , ethylene, ethane, pro- 2pionitrile, methyl acrylonitrile, pyrrole, vinyl pyri- not appear to differ greatly. Apart from relatively dine, indene, methyl cyanobenzene, naphthalene, heavy polyurethane chain fragments, N, CO, CO, 22quinoline, and isoquinoline were found to be minor HO, CHCH, and HCN have been detected and 2653products. The toxicity coefcient calculated for all reported by many authors.40,4345,49,50 Twogases showed that combustion gases from polyur- approaches to the estimation of the toxicity of degra- ethane foam are more toxic than those from wool dation, pyrolysis, and combustion products of poly- and nylons. A series of gas chromatography/mass urethane foams have been reported: (1) analysis of spectrometry analyses was carried out with polyur- volatile products and calculation of their toxicity ethane foams in combustion chambers in the aming and (2) toxicity tests with various animals. The com- combustion mode at 700, 800, 900, and 1000 C and

    in the nonaming combustion mode at 600 C.position of gases that evolve during the thermal

    decomposition of polyurethane foams in air and 45,49 nitrogen was studied by Woolley and coworkers.4345 The combustion products were primarily nitrogen-

    containing compounds and not oxygen-containing The evolution rate of each gas initially increases

    oxidation products other than CO, COslowly with temperature, but at a critical tempera- , and HO. 22The condensation of high-boiling products was also ture, the rate begins to increase rapidly. At 300 C, observed on the inside walls of the furnace. In con- there is a rapid and complete loss of the TDI unit of trast to other nitrogen-containing polymers, polyur- foams as volatile gases leaving a polyol residue. At ethane foams yield only one product, that is, HCN at 800 C, low-molecular-weight nitrogen-containing 1000 C. Thermal degradation and evolving gaseous products are isolated. When the temperature reaches products from the pyrolysis of rigid polyurethane the range of 900950 C, benzonitrile and hydrogen foams have also been studied with thermal analysis/ cyanide are virtually predominant. At 1000 C, mass spectrometry and thermal analysis/Fourier approximately 70% of the available nitrogen of the transform infrared spectroscopy.53polyurethane foams is converted into hydrogen cya- The degradation

    of urethane foams, studied with a cone calorimeter, nide. At 1000 C, polyurethane foams generate almost and evolved gaseous compounds, quantied by Fou- equal quantities of HCN (3.8%) in air and nitrogen, rier transform infrared, shows high concentrations of which are less than those produced by polyacryloni- isocyanates, amino-isocyanates, and amines.54trile and urea formaldehyde resin and more than 45those produced by nylons. The similarity of the acti- Woolley et al.detected mostly HCN by heating a

    vation energies for HCN evolution in air and nitro- polyurethane foam under air or nitrogen at 700

    1000 C. A higher evolution rate was observed at gen suggests that the mechanism of gas evolution is 51higher temperatures, so the amount of HCN that was not affected by oxidation reactions.In contrast, Jel- 52produced increased with increasing temperature. A linek and Dunklesuggested that urethane groups higher concentration of HCN was generated when the (ANHCOOA) oxidize by atmospheric oxygen during polyurethane foam was decomposed in a cup furnace the decomposition of polyurethane foam, resulting in

    Journal of Applied Polymer Science DOI 10.1002/app


    via a two-phase procedure (nonaming phase fol- cause 50% lethality (LC50) during a planned 30-min ex-

    posure and 10-min recovery period were determined lowed by ramped heating) than when the foam was along with the lethal blood HCN and carboxyhemoglo- decomposed under only nonaming or aming condi- 55bin (COHb) concentrations. LC50 means that the con- tions.When a copper compound such as cuprous centration produced under the stated conditions leads oxide (CuO) was added to the polyurethane foam, 2to the death of 50% of the exposed animals within 14 the formation of HCN during thermal decomposition days following exposure. The concentrations of toxic in a quartz beaker set in a cup furnace was reduced re gases determined for the LC50 data were related to substantially. When a cuprous oxide containing poly- the mass of the test specimens used. From a series of urethane foam was decomposed in a small-scale test, experiments with polyurethane foam, the LC50 was it showed an 87% reduction in the concentration of calculated to be 6.6 g, and the time needed to cause

    50% lethality was 9.5 min.HCN, whereas during a large-scale test, this reduction 59was 70%. A melamine-treated polyurethane foam gen- The blood and ambient

    concentrations of gases from the combustion of the erated 10 times more smoke than a conventional foam polyurethane foam indicated CO and HCN as the prin- when both decomposed in the two-phase cup furnace cipal toxicants. During the LC50 determination of the smoke toxicity test. Under similar conditions, a mela- polyurethane foam, it was found that the blood cya- mine-containing foam generated 90% less HCN when nide value was very high; this indicated that HCN was 56it was treated with CuO.Cuprous oxide acts as an 2the primary toxicant because the COHb levels were oxidative catalyst that decomposes gaseous HCN into very low. It appears that HCN was absorbed very N, CO, HO, and a small amount of nitrogen oxides, 222quickly into the blood, resulting in a low oxygen con- resulting in a reduction of the HCN concentration. At centration that caused a rapid toxic effect, probably a higher temperature, the addition of inorganic com- preventing the normal process of tissue oxidation and

    paralyzing the respiratory center of the brain, thus pounds has little effect on the formation of HCN.

    resulting in death. Similarly, the toxicity of CO is During combustion processes in which HCN is mainly due to its afnity to hemoglobin (Hb; the main formed, air pollution by cyanoarenes and aza-arenes structural protein of blood). Hb has 200300 times (part per million concentrations) may occur. Dennis et more afnity toward CO than O33al.studied combustion products of a composite ma-

    terial based on a polyurethane foam and wool fabric. . CO, when breathed 2It was shown that the evolution of the main toxic in along with air, is absorbed by the blood, reducing gases CO and HCN depends on the air ow in the the O-carrying capacity of blood. Hence, CO readily 2combustion area. The evolution of HCN takes place reacts with Hb to form COHb, a stable compound toward the end of combustion at a low rate of aera- resulting in Odeciency in the body tissues, which 2tion, whereas CO releases in the rst minute of com- causes headache, mental dullness, and tightness in the bustion at a high rate of aeration. The total content of 60chest, which leads to death.The air and blood data both toxic gases increases when a low concentration gathered during polyurethane foam combustion sug- of oxygen is passed through the zone of combustion. gest that death can be attributed to HCN and CO in The addition of ammonium polyphosphate (APP) in 61the Ourethane foams sharply decreases the emission of CO -decient environment. Sakai and Okukubo 2and HCN, smoke density, and formation of soot. Ex- derived similar conclusions from their toxicological pandable graphite, when added to polyurethane, ena- experiments with animals. bles a decrease in toxic gases in a lower proportion 56Levin et 344 male Fischer rats to than APP.thermal decomposition products of a polyurethane 57foam and a polyurethanepolyester combination. The toxicological effect depends on the The decomposition products of the polyurethane concentrations of both gases together with their ratio. foam and the foam in combination with polyester The combustion gases with higher ratios of HCN to produced no animal deaths during exposure and CO are more toxic. The combustion of polyurethane caused postexposure deaths only in the nonaming foams was studied in a National Bureau of Standards modes. Babrauskas et al.(NBS) chamber, and it was found that the gases which 62exposed rats to thermal evolve from the nonaming combustion of polyur- decomposition products of a re-retardant polyur- ethane foams are more toxic than those from aming ethane foam using a poly(methyl methacrylate) rec- combustion.41tangular apparatus. Only the head of each animal was exposed for 30 min to avoid heating of the The acute toxicity by inhalation and lethality of ther- whole body. The toxicity of the combustion products mal decomposition products of polyurethane foams from the re-retardant foam was attributed to the were investigated with a cup furnace smoke toxicity formation of a bicyclic phosphate ester in the smoke, apparatus, a poly(methyl methacrylate) rectangular which resulted in the immediate death of the ani- box, and a quartz tube inside an electrically heated an- mals. An analysis of the combustion products nular furnace according to DIN 53436.58The concentra- revealed that HCN, CO, and vinyl pyridine are

    tions of pertinent re gases individually and in various probably responsible for the toxic action. Various

    combinations and the amount of material needed to

    Journal of Applied Polymer Science DOI 10.1002/app


    polymeric materials including polyurethane foams the addition of reactive re retardant can be greatly

    reduced in the presence of a synergist, without any have been evaluated at different temperatures, heat- 66reduction of the re-retardant rates, and air ow rates for thermophysical and A ammable 63foam can also be rendered re-retardant by the pro- toxicological responses.The weight of the pyro- tection of its surface with re-retardant coating lyzed material, which corresponds to the lethal compositions. effect, is the weight of the material, which effectively

     causes death. Because the toxicity of the gases increases with increasing char yield for polymers containing nitrogen, it is believed to be indicative of

    the presence of toxicants other than CO and HCN. Phosphorus-containing additives

    Many workers have come to similar conclusions in Phosphorus their toxicological experiments with various cellular Inorganic phosphorous compounds are used for re plastics and polymers.64,65 Small-scale and large- retardants either by blending with polyurethane scale experiments on the toxicity of polyurethane components or by reacting into the main polymer 56foams were conducted by Levin et al.In the small- chain. Piechota67was the rst to investigate polyur- scale experiments, mortality depended on the

    ethane foams and found red phosphorus to be very amount of material burned; thus, the amount of effective as a re retardant. Although red phospho- material required to produce 50% mortality (LC50) rus is used in polyurethane foam formulations, was measured. LC50 for the polyurethane foam was phosphorous compounds in the form of phosphates, 6.6 g, lower than that for nylon (7 g), acrylic (8 g), phosphites, phosphonates, phosphonitrides, phos- cotton (10 g), or wood (11 g). phoric acid, phosphonic acid, and halogen-contain- ing phosphorous compounds are more effective. As reported in the literature,FIRE RETARDANCY OF POLYURETHANE 68,69 the general mechanism FOAMS of the re-retardant action of phosphorus in polyur-

    ethane foams is similar to that in other polymers. Fire retardancy requires the disruption of the burn- Phosphorus-containing re retardants mainly inu- ing process at one or more stages so that the process ence the reactions taking place in the condensed is terminated within an acceptable period of time. In phase. Thus, phosphorus appears to retard the com- general, three methods have been employed to bustion mechanism occurring primarily in the con- render polyurethane foams re-retardant. The reac- densed phase in three steps.70tive re retardants participate in the foaming reac- First, phosphorus

    may form anhydrides of phosphoric and related tions and build chemically into the polymer acids by thermal decomposition, and they may act molecule together with the other starting foam com- as dehydrating agents, extract water from the pyro- ponents. This prevents them from bleeding out of lyzing polymer, and promote char formation. The the polymer, and their re retardancy is thus presence of char will result in lower heat transfer retained. They have no plasticizing effect and do not from the ame to the condensed phase, and this will affect the thermal stability of the foam structure. Ba- interfere with the heating and decomposition pro- sically, these compounds are based on phosphorus cess. Second, phosphoric and related acids may act and halogen. Phosphorus is present in the form of as a heat sink because they retard the oxidation of

    phosphate, phosphite, phosphinate, phosphonate, carbon and oxygen to carbon dioxide; this will

    decrease the heating process. Third, the acids may phosphonitride, and organophosphorous polyols.

    form a thin glassy or liquid protective coating on the Halogens are effective in brominated or chlorinated condensed phase, thus lowering oxygen diffusion forms or in a combination of both derivatives. The and heat and mass transfer between the gas and the nonreactive re retardants are not believed to partic- condensed phases. This barrier disturbs the oxida- ipate in the foaming reaction, and they provide a tion process of carbon at the carbon monoxide stage, degree of re retardancy on a weight basis. If they thus decreasing the exothermic heat of combustion. are compatible with the polymer, they act as plasti- According to Granzow,cizers; otherwise, they are considered llers. They 71are often volatile or tend to bleed, so their re the phosphorus-containing

    re retardants can also be effective in the gas phase. retardancy may be gradually lost during the aging Phosphorous compounds break down into small process. A wide variety of nonreactive additives fragments that can be detected in the gas phase. based on phosphorus, halogens, nitrogen, sulfur, bo- These fragments catalyze the recombination of ron, aluminum, antimony, carbon, and silicones are hydrogen atoms into hydrogen molecules and thus being used. A combination of reactive and additive reduce the energy of the ame. Weil72re retardants produces a synergistic effect. Syner- reviewed the

    re-retardant mechanism of phosphorus-containing gists have achieved great importance because they compounds. A recent review of the phosphorus- are less expensive than actual re retardants, and

    Journal of Applied Polymer Science DOI 10.1002/app


    based re retardants was written and published by cies are derived from the thermal decomposition of 73Levchik and phosphorus, which reacts in the presence of O 2

    with HCl to yield PClThe addition of 1.5 wt % phosphorus increases the , a well-known ame inhibi- 3tor. The synergistic effect has also been observed char formation of a polyurethane foam from weak to 115with some metal oxides such as MgO.MgO cata- strong. Weight loss also is reduced from 100 to 26%

    lyzes the reaction of red phosphorus with Obecause of more char formation. The incorporation in the 2presence of moisture to yield phosphoric acid and of phosphorus is effective when its concentration is its derivatives. Thus, MgO induces a higher rate of in the range of 1.01.5 wt % in the total formulation, phosphoric acid formation, which increases the char- and a further increase seems to produce no further ring rate on the burning polymer surface. Vanadium 74benet.However, scorch generation is the main oxide80has been found to be an efcient synergist problem in phosphorus-modied re-retardant ure- 50with red phosphorus. Vanadium oxide facilitates the thane foams.7577 Ravey and Pearceincorporated oxidation of phosphorus, leading to the formation of phosphorus into a polyurethane foam formulation vanadium phosphate, which in turn catalyzes char- using HPO4 (85%) in acetone, with the concentra- 3ring of the polymer. Levchik et al.81found that apart tion ranging from 0.2 to 5.6%. TGA showed that the

    from vanadium pentoxide, molybdenum trioxide presence of phosphorus reduces the thermal stability and tungsten trioxide are mildly benecial coaddi- of the polyurethane foam. A vertical ame test con- tives to phosphorus. Depending on the concentra- rmed that the burning length is also reduced from tion, these additives improve the LOI rating. The 100 to 18 mm after 60 s of exposure. The phospho- general assumption is that phosphorus mostly rous compounds hinder the ow of the molten poly- shows re-retardant properties only in the presence mer and thereby prevent aming drips; this of an oxygen-containing polymeric substrate.78improves performance according to ASTM D 1692 How-

    ever, the researchers, noting the limited efciency of 59T, BS 4735, and DIN 4102-B3 re tests. The main phosphorus in nonoxygenated polymers, suggested characteristic of a red-phosphorus-containing poly- another mode of action. On heating, phosphorus is urethane foam is that the foam not does melt during depolymerized almost quantitatively into volatile re exposure but forms a protective crust. The addi- white phosphorus, which diffuses from the polymer tion of red phosphorus does not change the mechan- to the burning surface, at which it is oxidized into ical properties of such foams. The main Hdisadvantages of red phosphorus as a re retardant POderivatives. At the burning polymer surface, 34

    the formed Hfor polyurethane foams are its color and generation POacts as a char-forming agent, thus 34physically limiting oxygen access and fuel volatiliza- of highly toxic phosphine through a reaction with 68tion. Although the mechanism of interaction of red water.A stabilizer such as a metal oxide can be phosphorus and polyurethane foam is not very clear, used successfully to minimize trace amounts of by nding parallel trends for the LOI and combus- phosphine. It has been found that copper oxide, cad- tion index measured in a milder oxidizer (Nmium oxide, or zinc oxide can efciently transform O), 268they experimentally showed that red phosphorus phosphine into phosphoric acid.The efciency of provides a condensed- and gas-phase re-retardant red phosphorus can be increased if it is dispersed in 79action.78Furthermore, the amount of char produced d-caprolactam before polymerization.The presence

    from the polyurethane foam increases with an of red phosphorus reduces the polymer molecular increasing content of phosphorus. weight, enhancing the re-retardant efciency of

     phosphorus to the maximum. The limited oxygen

    index (LOI) values of polyurethane foams containing Phosphorus-containing organic products 0.3, 0.5 and 0.7 wt % phosphorus were 20.5, 21.8, A high-level effort is shown in the patent literature and 22.8, respectively, whereas without phosphorus 79(Table III) with respect to organophosphorous re the value was 16.5.Similarly, LOI and TGA values retardants for polyurethane foams. Structurally were also determined for a polymethylene poly- bonded organophosphorous-based polyols are more phenyl isocyanate (PAPI) based phosphorus-contain- effective than the nonreactive additives.7Foams ing polyurethane foam, which showed better re

    formed from such polyols are more effective in performance than foams based on 4,40-diphenylme- retaining their re retardancy after aging. Trivalent thane diisocyanate (MDI) and TDI. A polyurethane phosphorus (phosphines, phosphinites, phosphonites foam containing 1.542.0 wt % phosphorus was self- or phosphites, and phosphonates) usually exhibits extinguishing in air. low thermooxidative stability; therefore, it is a per- Phosphorus shows synergistic action with halo- fectly suitable re-retardant additive for polyur- gen-containing compounds and thus increases the ethane foams. Curtat et al.110re retardancy of polyurethane foams. It has been studied the re-

    retardant effect of phosphine oxides and phospho- suggested that a phosphorus/halogen molar ratio of nates in rigid polyurethane foams. The experimental 1 : 1 is optimal. It has been speculated that Pspe- 4

    Journal of Applied Polymer Science DOI 10.1002/app


results have shown less weight loss than for other (11.1%), diethyl-N,N-bis(2-hydroxyethyl)-aminomethyl

    phosphonate (12.2%), and tris(dipropylene glycol) re retardants by the formation of polyphosphate phosphite (7.17.3% P). With a 3% phosphorus con- layers in char, which offer greater resistance to the tent in the urethane foam, the LOI values of phos- diffusion of fuel volatiles and raise the mechanical phate-, phosphite-, and phosphonate-containing stability of char. Phosphate polyols such as chlori- urethane foams are 20.9, 21.0, and 21.2, respectively; nated aliphatic phosphites [tris(2-chloroethyl phos- however, at a level of more than 4% phosphorus, 111phate)s] have been used successfully.The there is a considerable increase in the LOI value of a main disadvantage of phosphate polyols is that the phosphate-containing urethane foam (23.1), whereas presence of even a small amount of water causes hy- LOI values of phosphite- and phosphonate-containing

    drolysis. Hydroxyethyl phosphate and dimethyl- urethane foams remain unchanged. A TGA study of

    phosphate-, phosphite-, and phosphonate-containing phosphite can be made to react with pentaerythritol

    foams showed that they decompose at 180, 140, and or trimethylol propane to obtain phosphate polyols 112200 C, respectively, and at 450 C, a phosphite-based with as much as 20% phosphorus.Phosphine foam shows maximum weight loss (82%), whereas a oxides, in contrast to phosphites, are among the phosphate-based foam shows minimum weight loss most thermally and oxidatively resistant organo- (63%).phosphorous products. However, their relatively 119high cost probably prevents their use as re retard- TGA and oxygen index studies of polyur-

    ethane foams containing poly[bis(carboxylatophenoxy) ants for polyurethane foams. phosphazene], diethanolaminomethyl phosphate, and Most phosphites used in the polyurethane foams trisodium pyrophosphate have shown higher char are ultimately converted into phosphonates. Amino- yields and increased values of the oxygen index.methyl phosphonate and dimethylmethyl phospho- 120,121 nate (DMMP) have been used as reactive re- The phosphorus contents of re-retardant urethane

    foams can be reduced greatly in the presence of halo- retardant additives to polyurethane foams and have 113gens, which exhibit synergistic action with phospho- been found to be very effective.DMMP contains rus. Usually, a 1.5% concentration of phosphorus is 25% phosphorus, and only about 8 wt % is required required to produce re-retardant polyurethane foams in rigid foam formulations. Diethylethyl phospho- in the absence of halogens. The content of phosphorus nates or triethyl phosphates are also used for the can be reduced to 1.0% with the addition of 1015% same purpose. Bayer recently introduced dimethyl- chlorine, and 47% bromine may lower the need for propyl phosphonate and diethylpropane phospho- phosphorus to about 0.5%. Thus, phosphorusbro- nate as halogen-free re retardants for urethane mine synergistic systems are more efcient than phos- 114foams.Triphenyl phosphate, isopropylphenyldi- phoruschlorine systems, although the quantities of

    chlorine and bromine are reduced considerably when phenyl phosphate, tricresyl phosphate, and trixylene

    they are used in combination with Sbphosphate have found some use in rigid foam for- 115122mulations.Approximately 15% phosphonate re O.LOI val- 23ues of polyurethane foams containing 5% phosphorus retardant was required to introduce 1.52.0% phos- in the form of phosphate, phosphonate, and phosphite phorus into the nished foam. The burning length of in combination with 2% bromine increase from 19.5 to the foam decreased to 8 mm, and the self-extinguish- 23.2, 22.7, and 21.4, respectively. These LOI values ing time was less than 15 s according to ASTM D indicate that bromine is more effective in combination 1692-59T. Another phosphonate that contains phos- with phosphates and phosphonates than with phos- phorus connected to hydroxyl groups has been phites. Maximum char yields have been reached with claimed to produce urethane foams with good re 5% phosphorus from both phosphate and phospho- retardancy. Phosphorussulfur additives such as nate with 2% bromine, at which maximum re retard-

    ancy has also been observed. Foams containing PS, PS, and PShave also been suggested for pol- 4104743116phosphite produce about 30% char and are insensitive yurethane foams.It has been reported that 2% to the addition of bromine.phosphorus, 6% antimony, 10% bromine, and 13% 119chlorine alone are necessary for the nonburning of These foams exhibit re

    retardancy just sufcient to pass the ASTM D 1692- urethane foams. Diisocyanates and triisocyanates such 67T ame test. A exible polyurethane foam contain- as phosphoryl triisocyanate have also been used to ing Phosgard 2XC20 (Monsanto Chemical Co.) with introduce phosphorus into the polyurethane foam 10.6% phosphorus and 35.2% chlorine passes the am- 117118structure.Bakhitov et al.described the prepara- mability test (DOC FF-2-70 tablet test), and the diame- tion of phosphorus-containing urethanes from equi- ter of the hole burned out has been reported to be 3.4 molar amounts of tris(hydroxylmethyl) phosphine, in.123124tris(hydroxymethyl) phosphine oxide, tetrakis(hydrox- Larsen and Eckerstudied the thermal stability

    and decomposition temperatures of polyurethane ymethyl) phosphonium chloride, hydroxymethyl foams containing haloalkyl phosphates such as phosphonic acid, and diisocyanates. Flexible urethane tris(2,3-dibromopropyl) phosphate, tris(tribromoneo- foams have also been prepared with phosphorus con- pentyl) phosphate, and pentabromodiphenyl oxide. tents of 0.31.0% with a phosphate-containing polyol

    Journal of Applied Polymer Science DOI 10.1002/app



    Phosphorus Reactive Products and Their Synergistic Combinations Disclosed in the Patent Literature

    Number Fire retardant Best example Reference 1 Phosphorus-containing polyol and No example has been 82 oxide-modied sucrose reported. Flame-resistant polyurethane foam with 90% closed cells, good dimensional stability, and a K factor less than 0.13 Reaction product of methyl 2 83 Foam containing 2.6 wt % 3-(dimethylphosphono) propio- phosphorus and classied nate and polyoxypropylene as nonburning according sucrose containing 5.2% to ASTM D 1692-59T phosphorus

    Hydroxyl phosphonate, a mixture Flame-resistant polyur- 3 84

    of trimethylolpropane(trimethy- ethane rigid foam having lolpropane butane phospho- a burn length of 33 mm nate according to UL test

    Rigid polyurethane foam 2-Chloro-1-hydroxyl ethane-1, 4 85

    1-diphosphonic acid or di- which exhibits a reduced phenyl 1,2-dihydroxyethane-1,1- burning rate and can be classied as nonburning diphosphonate 5 86 Terephthalic acidpolyoxypropy- Self-extinguishing rigid lene alkoxy diphosphates polyurethane foam (phos- phorus 0.36 wt %) with a burning extent of 38 mm according to ASTM D 1692-59T 6 Phosphorus (15%) and chlorine 87 Resultant polyol can be used to form polyur- (12%) containing polyol ethane foams having obtained by the reaction reduced ammability between Union Carbide Niax polyol, propylene oxideethyl- ene oxide adduct, bis(b-chlor- oethyl) vinyl phosphonate, and vinyl chloride Resultant foam is nonburn- 7 88 Phosphated starch polyether con- taining 4.2% phosphorus with a ing according to ASTM D hydroxyl number of 303 1692-59T and has a 22- mm extent of burning

Journal of Applied Polymer Science DOI 10.1002/app

Report this document

For any questions or suggestions please email