Processing and Properties of Fire Resistant EPDM Rubber-Based Ceramifiable Composites

Materials

Ethylene-propylene-diene rubber (EPDM) “Keltan 21” (Mooney viscosity in 125 °C – 25, ethylene content – 60 %) was purchased from Lanxess AG (Germany). The reinforcing filler – precipitated silica “Arsil” (bulk density – 150 g/cm³) was obtained form Rudniki S. A. (Poland). Reference mineral filler – wollastonite “Termin 939–304” (L/D – 8/1, hardness – 4.5 Mohs, density – 2.85 g/cm³) was obtained from Quarzwerke GmbH (Germany). The curing agent – dicumyl peroxide (DCP) (purity – 98 %) was purchased from Sigma-Aldrich (USA). The glassy frit used as a fluxing agent – “FR-2030” (composed of 13.7 wt % Na₂O, 2.0 wt % BaO, 23.5 wt % ZnO, 2.0 wt % Al₂O₃, 43.1 wt % SiO₂, 15.7 wt % B₂O₃) was obtained from Reimbold & Strick GmbH (Germany).

Methods

EPDM rubber mixes were prepared by two roll (diameter – 150 mm, length – 200 mm) mill (Bridge, UK) which operates with friction of 1.1. For the composition of mixes, see Table 1. The crucial parameter which distinguishes the mixes is the content of the fluxing agent. Each mix was filled with 300 weight parts of mineral fillers per 100 weight parts of rubber, but the content of the fluxing agent amounts to 30, 40 and 50 % wt of summary fillers weight, in EPDM_30, EPDM_40 and EPDM_50 respectively. Kinetics of vulcanization of prepared mixes was measured by means of Metalchem MW-05 vulcameter (Poland) in 150 °C. Samples of vulcanized rubber were prepared in a heated mold press accordingly to the kinetic of their vulcanization under pressure of 10 MPa. Samples for tests were shaped using a mechanical punch press. Micromorphology of vulcanized composites was pictured by means of Molecular Imaging Metrology MI 2000 atomic force microscope (AFM) (USA), working in an oscillating mode with frequency of ca. 170 kHz. Mechanical tests were performed by means of Zwick 1435 device (Germany) accordingly to the ISO 37:1998 standard. Thermal stability of the samples was studied by means of Netzsch 449 F3 TG-DTA instrument (Germany), operating with heating speed of 10 °C/min, under air atmosphere from room temperature to 1,000 °C. Combustibility of the samples was tested by means of a cone calorimeter produced by Fire Testing Technology Ltd. (UK). Square shaped vulcanizates ((100×100) ± 1 mm) of 2.0±0.5 mm thick were heated by 35 kW/m² radiant flux. The ceramification ability of samples was investigated by heat treatment in a laboratory furnace. Samples were heated in two different conditions – slow heating from room temperature to 800 °C or 1,000 °C with heating speed of ca. 8 °C/min and fast heat treatment in which samples were placed in a preheated furnace chamber (800 °C or 1,000 °C). Resistance against fragmentation of ceramic char created after heat treatment of the samples was measured by means of Zwick/Roell Z 2.5 tester (Germany).

Kinetics of vulcanization

Scorch time of the mixes vary from 2.7 to 3.4 min and its change is not directly connected to the amount of the fluxing agent, whereas vulcanization time slightly increases from 36.1 to 38.2 min, with an increase of fluxing agent content, what shows that its presence affects the curing process (Table 2). Generally, all torque values decline significantly with a decrease of silica content what was expected as the precipitated silica is an effective reinforcing filler. However, the torque growth decreases visibly with the increase of silica content, what means that the curing process is less efficient. It could be an effect of a relatively high amount of moisture adsorbed on the surface of silica particles. This may create a favorable environment for acidic impurities leading to the disruption or even prevention of peroxide curing [15, 16, 17]. The vulcanization time of EPDM rubber shortens with the addition of each mineral composition.

Mechanical properties

The values of all mechanical parameters increase with the amount of the fluxing agent (Table 3). This seems to be not due to the effect of good reinforcing properties of the fluxing agent rather than of reinforcing silica overload and higher effectiveness of curing of mixes containing lower amounts of silica. On the whole, the mechanical properties of all samples are sufficient for technological applications increasing the tensile strength and tear resistance of EPDM rubber with a slight decrease in its elasticity.

Micromorphology

AFM picture of EPDM_40 sample surface shows accurately the micro-morphological character of the composites studied (Figure 1). The high amount of mineral fillers is clearly visible. Wollastonite and the fluxing agent occur in form of relatively big primary particles, whereas silica creates aggregates and agglomerates in an EPDM matrix due to great difference between polar character of silica surface and non-polar aliphatic polymer matrix. Wollastonite is present in the form of needle-shape particles (see the down right corner of photo 1), whereas the fluxing agent, as an amorphous milled matter, occurs in the form of irregular-shaped particles.

Figure 1:

AFM photographs of EPDM_40 surface micromorphology, taken in two different magnifications (1, 2) from topography channel (A), amplitude channel (B) and phase channel (C).

Thermal stability

EPDM_40 sample presents the best thermal stability (Table 4). The onset temperature of rapid decomposition of this sample is higher than for EPDM_50, however it equals EPDM_30. It is very likely that the higher amount of reinforcing silica added to EPDM rubber in EPDM_30 and EDPM_40 samples caused the creation of higher amount of bound rubber adsorbed on silica surface, which has been proved to be thermally more stable than unbounded rubber [18]. The amount of ceramic char residue after TG test is the highest for EPDM_40 and the kinetics of its decomposition is the slowest (Figure 2). All samples show a characteristic, wide, exothermic DTA signal with a double maximum peak preceded by endothermic signal of moisture evaporation (Figure 3). EPDM_40 shows the lowest value of peaks and the lowest temperature of first peak, whereas EPDM_50 presents exothermic signal of the smallest width. The samples after laboratory furnace heat test do not keep the same shape, however they maintain a consistent porous structure exhibiting protective properties (Table 5). The force required to crush the ceramic residue is not directly comparable for the samples because their shape was slightly different. However the mechanical resistance of samples after treatment in 800 °C is much higher than the standard char obtained after heating of common polymer composites. After 1,000 °C treatment only the EPDM_40 sample was able to create a strong ceramic residue, in fact much stronger that after heating up to 800 °C, regardless of heating rate. The strongest ceramic structure was created after slow heating of EPDM_40 up to 1,000 °C. It is very probable that in such temperature very low viscosity of the fluxing agent enhances the stickiness of mineral particles leading to a significant increase of mechanical endurance of the sample. A very high amount of the fluxing agent in EPDM_50 composite causes melting and deformation of the samples after heat treatment in 1,000 °C, whereas a low content of the fluxing agent in EPDM_30 causes blowing of the sample due to rapid decomposition of EPDM rubber matrix or leads to the creation of a very weak ceramic structure. This data correspond precisely to the TG analysis which shows that EPDM_40 exhibits the slowest decomposition rate and produces the highest amount of ceramic residue. The reference sample (EPDM_prist) also decomposes in a three-step process. However, the onset temperatures of each step are shifted into lower values. This indicates that the used filler systems increase the thermal stability of EPDM rubber.

Figure 2:

TG curves of samples studied.

Figure 3:

DTA curves of samples studied.

Figure 4:

Heat release rate (HRR) curves of samples studied.

Figure 5:

Total heat release (THR) curves of samples studied.

Figure 6:

Averaged heat release rate (ARHE) curves of samples studied.

Figure 7:

Mass loss curves of samples studied.

Figure 8:

Photographs of samples studied residue after cone calorimetry test.

Combustibility

The most important parameters describing resistance to combustion of a composite material are time to ignition (t_i), peak of heat release rate (HRR_p) and total heat release (THR) (Table 6, Figures 4–7). Taking into account the first two of them the EPDM_30 composite exhibits outstanding properties. Its time to ignition is almost 50 s higher than that of all other samples and heat release rate peak is the lowest. It is highly likely that the large amount of silica provides high thermal stability as the TG/DTA analysis showed but it also contains the highest amount of water which evaporates suppressing the temperature of the composite surface during the cone calorimetry test.

The EPDM_40 composite, which showed the best ceramifiable and thermal properties exhibits also low combustibility having the lowest total heat release parameter. This is probably the result of an effective ceramification on the surface of the composite leading to the production of dense and barrier ceramic char and preventing combustible volatiles from escaping and burning in the fire zone (Figure 8).