Fire hazards are primarily the result of a combination of different factors including ignitability, ease of extinction, flammability of the volatiles generated, amount of heat released on burning, rate of heat release, flame spread, smoke obscuration and smoke toxicity. The most important fire hazards are heat, smoke and toxic gases [1].
A high rate of heat release causes a fast ignition and flame spread. It also controls the fire intensity and is therefore much more important than ignitibility, smoke toxicity or flame spread. The time available for fire victims to escape is also controlled by the heat release rate. Smoke production is a further important fire hazard. People get disoriented in dark smoke and therefore are unable to escape unless they can see. In addition, fire fighters have severe problems rescuing people in dark surroundings.
The acute toxicity of fire gases is mainly controlled by the carbon monoxide content. CO is responsible for over 90 percent of people killed by fires [2]. Each year about 5,000 people are killed by fire in Europe and more than 4,000 people in the USA. Direct property losses account for around 0.2 percent of the gross domestic product and the total costs of fires are around one percent of the gross domestic product [3]. Therefore it is important to develop well designed flame retardant materials to decrease these fire hazards. Polymers are used in an increasing number of applications, where specific mechanical, thermal and electrical properties are required. One further important property is the flame retardant behaviour of the polymers, which can be traditionally achieved by a number of routes. These include use of intrinsically flame retardant polymers like PVC or fluoropolymers. Alternatively, flame retardants can be used such as alumina trihydrate, magnesium hydroxide, organic brominated compounds or intumescent systems to prevent the burning of polymers like polyethylene, polypropylene, polyamide and others.
However, in some cases these flame retardant systems can have significant disadvantages. These include that applications of alumina trihydrate (ATH) and also magnesium hydroxide require a very high portion of the filler within the polymer matrix; filling levels of more than 60 wt. percent are necessary to achieve a suitable flame retardancy, for example in wires and cables. Clear disadvantages of these filling levels are the high density and the lack of flexibility of the end products, poor mechanical properties and problematic compounding and extrusion steps. In addition, in Europe there are reservations about a general use of brominated compounds as flame retardants. Finally, intumescent systems are expensive and electrical requirements can restrict the use of these products.
The use of a new class of materials, called nanocomposites, can avoid the disadvantages of traditional flame retardant systems. Generally the term ‘nanocomposite’ describes a two-phase material with a suitable nanofiller (usually a modified layered silicate) dispersed in the polymer matrix at a nanometre (10−9 m) scale.
Nanocomposites properties
Compared with virgin polymers, corresponding nanocomposites can show particular improvements; the content of the modified layered silicates is often ranged between just two and ten weight percent. Some of the most important improvements include mechanical properties such as tension, compression, bending and fracture. Barrier properties like permeability and solvent resistance can also be improved, as well as optical properties. Ionic conductivity can also be improved. A review [4] discusses these improvements.
Other highly interesting properties of polymer-layered silicate nanocomposites concern their increased thermal stability and ability to promote flame retardancy at very low filling levels. The formation of a thermal insulating char, which also has low permeability to volatile combustion products caused by a fire is responsible for these improved properties [5-8]. The low filler content in nanocomposites and the drastic improvement in thermal stability is highly attractive for the industry because the end-products can be made cheaper and easier to process.
Experimental
Materials
A commercially available layered silicate based on montmorillonite modified by dimethyl-distearylammonium cations was used as a nanofiller. Ethylene-vinyl acetate (EVA) copolymers (Exxon's Escorene types) with different weight percent amounts of vinyl acetate were used in this study. Such copolymers have demonstrated their ability to promote nanocomposite formation by melt blending with nanofillers [9-11]. A precipitated alumina trihydrate was used.
Properties of EVA nanocomposites
Depending on the nature of the filler distribution within the matrix, the morphology of the nanocomposites can evolve from the so-called intercalated structure with regular alternate layered silicates and polymer monolayers to the exfoliated (delaminated) structure with the layered silicates randomly and homogeneously distributed within the polymer matrix. The easiest and technically most attractive way to produce these types of materials is kneading the polymer in the molten state with a modified layered silicate, such as montmorillonite. The native Na+ interlayer cation within the silicate has been exchanged by a quaternary alkylammonium cation. The modified filler is called a nanofiller and is much more compatible with the polymer matrix.
Mixing was done on several compounding machines. Rolling mill and internal mixer as discontinuous compounding machines were used; a BUSS co-kneader (with a rotating and simultaneously oscillating screw, 11 L/D, 46 mm screw diameter) was used as a continuous compounding machine. A processing temperature of 160°C was used for all the different compounding machines.
Information on the nanocomposite morphology (see Figure 1) has been obtained by transmission electron microscopy (TEM) and X-ray diffraction (XRD) observation. Exfoliated silicate sheets are observed together with small stacks of intercalated montmorillonite. This structure may be described as a semiintercalated, semi-exfoliated structure that does not primarily change with the vinyl acetate content of the EVA matrix; indeed a greater number of stacks are observed for EVA with lower vinyl acetate content [10]. There are also no great differences with the morphology of the nanocomposites related to the different compounding routes.
Table 1: Maximal temperature of the main degradation peak (DTG) in air, 20°C/min, for EVA and EVA nanocomposites. EVA: Escorene UL 00328 with 28 weight percent VA | Nanofiller content (weight percent) | Maximal temperature peak at the main degradation peak (°C) |
| 0 | 452.0 |
1 | 453.4 |
| 2.5 | 489 |
| 5 | 493.4 |
15 | 454.0 |
Results
Thermal stability
Thermogravimetric analysis (TGA) is widely used to characterize the thermal stability of a polymer. The mass loss of the polymer due to volatilization of products generated by the thermal decomposition is monitored as a function of a temperature ramp. Non-oxidative decomposition occurs when the heating of the material is done under an inert gas flow like helium or nitrogen, while the use of air or oxygen allows analysis of oxidative decomposition reactions.
The experimental conditions of the degradation highly influence the reaction mechanism of the degradation. The thermal stability of EVA-based nanocomposites was investigated [10] as partially intercalated and partially exfoliated structures independent of the EVAs used. TGA under helium (non-oxidative decomposition) and under air (oxidative decomposition) were investigated. EVA is known to decompose in two consecutive steps. The first one is identical to both oxidative and non-oxidative conditions. It occurs between 350°C and 400°C and is linked to the loss of acetic acid. The second step involves the thermal decomposition of the unsaturated backbone that has been obtained either by further radical scissions (non-oxidative decomposition) or by thermal combustion (oxidative decomposition). In helium, the EVA-nanocomposite has a negligible reduction in thermal stability compared to the virgin EVA or the EVA filled with Na-montmorillonite (microcomposite). In contrast, when decomposed in air, the same nanocomposite exhibits a rather large increase in thermal stability because the maximum of the second degradation peak is shifted 40°C to higher temperatures while the maximum of the first decomposition peak remains unchanged (see Table 1). In this case the explanation for the improved thermal stability is the char formation occurring under oxidative conditions. The char acts
as a physical barrier between the polymer and the superficial zone where the combustion of the polymer is occurring. The results in Table 1 on the maximal temperatures at the main degradation peak for EVA nanocomposites demonstrate that the optimum for thermal stabilization is achieved at a layered silicate level of 2.5-5.0 weight percent.
Flammability properties
From an engineering point of view it is important to know what hazards within a fire must be prevented and only then can strategies for measurements and improvements be developed. Extensive research at NIST (National Institute for Standards and Technology, USA) led to the important conclusion that allows significant simplification of the problem for hazards in fires: The heat release rate, in particular the peak heat release rate, is the single most important parameter in a fire and can be viewed as the ‘driving force’ of the fire [12]. Therefore, the universal choice of an engineering test for flame retardant polymers is currently the cone calorimeter. The measuring principle is the oxygen depletion with a relationship between the
mass of oxygen consumed from the air and the amount of heat released.
The cone calorimeter test is standardized as ASTM E 1354 and ISO 5660. In a typical cone calorimeter experiment the polymer sample (as a plate of 100 × 100 × 5 mm) in aluminium dishes is exposed to a defined heat flux (mostly 35 kW/m2 or 50 kW/m2). Simultaneously the properties of heat release rate, peak of heat release, time to ignition, total heat released, mass loss rate, mean CO yield, and mean specific extinction area, for example, can all be measured.
The flame retardant properties of the EVA nanocomposites have been determined using cone calorimetry under a heat flux of 35 kW/m2 (see Figure 2). Under such conditions, simulating a small fire, the effect of the nanofiller is initially observed for 3 weight percent. A decrease by 47 percent of the peak of heat release as well as a shift towards longer times are detected for a nanocomposite containing 5 weight percent of the nanofiller when compared to the virgin matrix EVA. Increasing the filler content to 10 weight percent does not improve the reduction of the peak of heat release further. As a decrease in the peak of heat release indicates a reduction of the burnable volatiles generated by the degradation of the polymer matrix, such a drop clearly indicates the
flame retardant effect due to the presence of the nanofiller and its molecular’ distribution throughout the matrix. The flame retardant properties are further improved by the fact that the peak of heat release is spread over a much longer period of time. The flame retardant properties are due to the formation of a char layer during the nanocomposite combustion. This char acts as an insulating and non-burning material that reduces the emission of volatile products (fuel) into the flame area. The silicate layers of the nanofiller play an active role in the formation of this char but also strengthen it and make it more resistant to removal.
Cone calorimeter experiments with a heat flux of 35 kW/m2 also show that virgin EVA is completely burned without any residue. In contrast to the previous result an early strong char formation is found for the EVA nanocomposite in a similar cone calorimeter experiment; but now this char is stable and does not disappear by combustion.
Finally, compared to the virgin EVA matrix, the nanocomposite burns without producing burning droplets (UL 94 vertical procedure) [14], a characteristic feature that further limits the propagation of a fire. This is an important characteristic for products to be classified within the new Euroclasses that regulate flame retardancy classification in Europe.
NMR-investigation and FR-mechanism
The degradation of EVA and EVA nanocomposites was investigated by solid phase CP − MAS −13 C − NMR spectroscopy. The principles and the measurement method are described in detail by Le Bras et al [16]. EVA (Escorene UL 00112 with 12 weight percent vinyl acetate content) and also a nanocomposite based on EVA (Escorene UL 00112 with 12 weight percent vinyl acetate content) with 5 weight percent of the nanofiller were irradiated within a cone calorimeter by a heat flux of 50 kW/m2. Samples were taken from the heat flux after 50, 100, 150, 200 and 300 seconds and the presence of EVA and the char formation was measured.
The following results were obtained [17]:
Before irradiation of EVA and EV
nanocomposite:
• 33 ppm => − CH2 – by polymer backbone
• 75 ppm => − CH3 by acetate group
• 172 ppm => −C=O by acetate group (small signal)
After irradiation of EVA:
• 50 seconds: New signals at 130 ppm(char: aromatics and polyaromatics)
and 180 ppm (− C=0 with beginning of oxidation), EVA signals present
• 150 seconds: no signals => no organic material present
After irradiation of EVA nanocomposite:
• 50 seconds: New signals at 130 ppm (char: aromatics and polyaromatics)
and 180 ppm (− C=0 with beginning of oxidation), EVA signals present
• 100 seconds: char-formation and EVA signals present
• 200 seconds: char-formation and EVA signals present
• > 300 seconds: no signals => no organic material present
The results show that the formation of nanocomposites clearly promotes char formation and delays the degradation of EVA.
Intercalation versus exfoliation
It is often reported that exfoliation is the most effective structure for maximal enhancement of properties improved by nanocomposites. Therefore it was of interest to shift the ratio of the mixed intercalated/exfoliated structure that is observed within EVA-nanocomposites [10] to the exfoliated structure. This was done by meltcompounding EVA (Escorene UL 00328) with 5 phr of the nanofiller in a twin-screw extruder. Two screw designs were used: one screw for maximal mixing using mixing elements and the second screw for maximal dispersion using kneading blocs. The screws were used from 300 to 1200 rpm.
TEM and XRD demonstrate that for the highest shear rate (1200 rpm) and highest friction (second screw) the mixed structures are shifted to the exfoliated one. However, cone calorimeter data show that there are no changes in the peak heat release rates for all the melt-compounded nanocomposites. Obviously the mixed intercalated/exfoliated structures within the EVA nanocomposites have already achieved the maximal reduction
in peak heat release rates.
Combination of the traditional filler ATH with a nanofiller
In order to achieve typical flame retardancy for cables required by the most important international cable fire test (IEC 60332-3-24) [15] a ratio of 65 weight percent of ATH and 35 weight percent of a suitable polymer matrix like EVA must often be used for cable outer-sheaths [13]. Therefore, the performances of two compounds were compared. Both compounds were prepared on a BUSS co-kneader (46 mm screw diameter, 11 L/D). One compound contained 65 weight percent ATH and 35 weight percent EVA Escorene UL 00328 and a second compound contained 60 weight percent ATH, 5 weight percent of the nanofiller and 35 weight percent EVA Escorene UL 00328. Both compounds were investigated by TGA in air and by cone calorimeter at 50 kW/m2. TGA in air clearly shows the delay in the degradation by the small amount of nanofiller.
The char of the EVA/ATH/nanofiller compound created by the cone calorimeter is rigid with only a few small cracks; but the char of the EVA/ATH compound is much less rigid (reduced mechanical strength) and with many big cracks. This is also the explanation why the peak heat release rate in the case of the nanocomposite is reduced to 100 kW/m2 compared to 200 kW/m2 for the EVA/ATH compound. In order to obtain the same decrease of peak heat release rate by the flame retardant filler ATH alone, the content of ATH must be increased to 78 weight percent in the EVA/ATH compound. The significant improvements in flame retardancy by the nanofiller also allow the possibility of decreasing the level of ATH within the EVA polymer matrix. In order to maintain 200 kW/m2 as an adequate peak heat release level, the content of ATH can be decreased from 65 weight percent to 45 weight percent by the presence of only 5 weight percent nanofiller within the EVA polymer matrix. The reduction in the total amount of these fillers also results in improved mechanical and rheological properties of typical EVA-based cable compounds.
Coaxial cable passing UL 1666 with a nanocomposite-based outer sheath
There are many applications for indoor cables passing the riser test defined in UL 1666, which has a 145 kW burner in a two-storey facility. This very severe fire test establishes two important measurement points of the flame propagation characteristics of rising cables – 12 feet maximum temperature of 850°F (454°C)
For passing this fire test halogenated cabl compounds are often used. However, increasingly non-halogenated flame retardant (FRNH) cables are requested by the market to pass the riser test. Cables based on nanocomposite compounds are demonstrating promising performances for this fire test. An example of a FRNH cable that passes UL 1666 is shown in Figure 3. The outer sheath is based on a FRNH nanocomposite with an EVA/ATH/nanofiller composition. A similar FRNH coaxial cable was also tested with an outer sheath based on just EVA/ATH. In both compounds thepolymer/filler relationship was the same and Table 2 shows the results.
The improved flame retardant properties are due to the formation of a char layer during the nanocomposite combustion. This insulating and non-burning char reduces the emission of volatile products from the polymer degradation into the flame area, which reduces the maximal temperature and height of the flames.
Conclusion
The thermal properties of EVA are improved by very low loadings levels of a suitable nanofiller within the polymer matrix. For these EVA nanocomposites TGA in air shows a delay of the degradation, and the
peak of heat release measured by a cone calorimeter is dramatically reduced. Nanocomposites improve char formation, which results in better flame retardancy. The results are also valid for EVA nanocomposites in
combination with metal hydroxides, such as alumina trihydrate, and open the possibility for new flame retardant compounds for cables with reduced total filler contents. A coaxial cable with an outer sheath based on non-halogenated flame retardant nanocomposites passes the UL 1666 riser test.
For references, contact m.holmes@elsevier.com.