Improving Flame Retardant Properties of Aliphatic Polyketone (POK)-Based Composites

The effect of zinc borate (ZB) and high-molecular-weight siloxane (SIL) on flame retardancy, mechanical, and thermal properties of aliphatic polyketone (POK)-containing aluminum diethyl phosphinate (OF) was investigated in this study. Ten wt % OF is sufficient to obtain V0 rating according to the UL94 test. As the weight fraction of OF was increased, the flame retardancy properties and LOI values improved, while the tensile and impact properties decreased. To avoid the degradation in mechanical and impact properties as much as possible and obtain the same and better flame retardancy properties, synergists such as SIL and ZB were used. Flame retardancy of POK-based composites was determined by the limiting oxygen index (LOI) test, UL94 measurement, and cone calorimeter test. The additions of 1 wt % SIL and ZB have not led to a considerable decrease in the tensile strength and impact properties of POK-10OF. While ZB and SIL are very efficient in decreasing the smoke density, ZB is more efficient than SIL in increasing the LOI value of the composite. The addition of 1, 2, and 4 wt % ZB and SIL synergists did not lower their UL94 ratings. Moreover, it can be added that ZB is more efficient than SIL in decreasing the fire growth rate (FIGRA) and maximum average rate of heat emission (MARHE) values. Using OF (10 wt %) and ZB (4 wt %), LOI values higher than 32% and smoke density values lower than 150 were obtained.


INTRODUCTION
Thermoplastic materials are used in all areas of our lives, such as white goods, automobile, furniture, electrical devices, electronics, etc. 1 Thermoplastic materials add different properties according to their usage areas. These features cover concepts such as safety, convenience, ergonomics, cost, and high mechanics. 2 Thermoplastics are materials with a high tendency to burn due to their natural properties. Today, it has become possible to add flame retardant properties to thermoplastic materials with the technology that has developed over the years. 3 Safety problems in the vehicle industry are tried to be solved by thermoplastic materials with flame retardant properties.
Safety in rail transportation systems is important for both passengers and valuable cargo. Fire safety is of great importance for rail systems. The safety risk can be reduced by choosing materials suitable for fire safety for trains. In addition, accidents can be prevented before they happen. 4 According to EN 45545:2013, for the hazard level HL3, limiting oxygen index values must be min 32% for interiors (R22) and exteriors (R23) of railway vehicles. In addition, smoke density values must be a maximum of 150 and 300 for R22 and R23, respectively.
Polyamide is used in various applications in the automotive industry such as door handles, body/chassis/structure, seat components, safety restraints, engine esthetic under-the-hood covers and pulley covers, air/fuel/oil (air inlet manifolds), front-end modules, and cables and cable conduits in rail, and railway tracks. 5 However, polyamide-based composite materials developed have certain disadvantages. High dehumidification capacity, deterioration in mechanical values over time, and changes in dimensional stability can be listed as the most important of these disadvantages. 6 Studies on new materials that are more advantageous and alternative to the widely used thermoplastic materials are continuing. 7 It is known that aliphatic polyketone (POK) is a polymer produced from olefin monomers and carbon monoxide (CO). Fundamental patents on catalysts and composition emerged in the early 1970s. However, these early resins could no longer be processed due to the catalyst. In 1996, Shell commercialized a terpolymer of carbon monoxide, ethylene, and small amounts of propylene under the trade name CARILON. 8 POK exhibits very high impact strength, acceptable abrasion resistance, chemical resistance, and low gas permeability. In addition, the consumption of CO during POK production contributes to the world′s air pollution removal. 9 Moreover, POK has the potential to be identified as a new environmentally friendly and high-performance engineering plastic. In addition, due to its chemical properties, POK is superior to polyamides in terms of flame retardant properties. 10 To satisfy the European fire protection norm EN 45545:2013 for railway vehicles, flame retardant POK composites were prepared. In this study, the restrictions about LOI and smoke density values for the interiors (R22) and exteriors (R23) of railway vehicles have been tried to be met. Toward this aim, halogen-free, flame retardant, POKbased composite materials were prepared using aluminum diethyl phosphinate as a base FR additive. To improve the flame retardancy performance, synergists such as zinc borate and high-molecular-weight siloxane were used. The effect of synergists on UL94, the fire growth rate (FIGRA), the maximum average rate of heat emission (MARHE), and peak heat release rate (PHRR) were also investigated. 2.5. SEM Analysis. SEM observations of the tensile fracture surface of the specimens of POK and its composites were conducted using SEM, COXEM, and EM-30 Plus operated at 10 kV. Before SEM analysis, gold was deposited on the surface of the specimens via a plasma sputtering apparatus.

MATERIALS AND
2.6. Differential Scanning Calorimetry (DSC) Analysis. Crystallization and melting behaviors of POK and its composites were investigated with a differential scanning calorimeter (TA Instruments Q20 DSC). The samples were heated from 10 to 300°C at a rate of 10°C/min under a N 2 gas atmosphere and cooled to −80°C at the same rate after 3 min of isothermal holding at this temperature. In the last stage, it was heated from −80 to 300°C with a heating rate of 10°C /min. 2.7. Thermogravimetric Analysis (TGA). The thermal stability of POK and its composites was determined by thermogravimetric analysis. TGA was carried out using a TG analyzer (TA Instruments Q50 TGA) by heating from room temperature to 800°C with a heating rate of 10°C/min under a N 2 gas atmosphere to prevent oxidation effects.
2.8. Thermomechanical Analysis (TMA). Thermomechanical analyses of POK and its composite materials were conducted using a thermomechanical analyzer (TA Instruments TMA Q400), and thermal expansion coefficients were determined. TMA was performed in the expansion mode. Samples with a size of 10 mm × 8 mm × 4 mm were heated from −30 to 120°C with a heating rate of 5°C/min under a load of 0.02 N.
2.9. Flammability Tests. The vertical burning tests for the specimen dimensions of 125 × 13 × 3 mm 33 were conducted according to UL94 standard using an ATLAS horizontal and vertical burning tester. Limiting oxygen index (LOI) values of samples were carried out according to the ISO 4589 standard with an LOI instrument (Fire Testing Technology). The flammability of the samples was determined according to ISO 5660 using an external heat flux of 50 kWm −2 with a dimension of 100 × 100 × 3 mm 3 . The examined flammability parameters are FIGRA, effective heat combustion (EHC), MARHE, time of peak heat release rate (tPHRR), PHRR, smoke density, and total heat release (THR) values. Smoke density measurements were carried out according to the ISO 5659 standard using a smoke chamber.

Tensile Properties.
The variations of the tensile strength of POK and its composites are presented in Figure 1.
As can be seen from Figure 1a−e, OF addition into POK resulted in decreases in the tensile strength values of samples. Adding more OF into POK led to more decrease in tensile strength values. It is known that at a higher filler content, the interaction between the filler materials and polymer matrix was impeded, which leads to lower strength values. 11 The effect of synergist addition together with OF on the tensile strength can has not led to a considerable decrease in the tensile strength of POK-10OF. However, 2 wt % of SIL caused a more significant decrease in the tensile strength than that of ZB. When the synergists were added by 4 wt %, SIL and ZB decreased the tensile strength by 10%. It is known that the tensile strength of a particle-filled polymer composite is strongly related to the interfacial adhesion developed between the filler and polymer matrix. If the filler−polymer adhesion is weak, the bond may be broken when the load is applied. 12 The variations of the tensile modulus of POK and its composites are given in Figure 2. It is seen that 10, 15, 20, and 25 wt % OF additions into POK increased the tensile modulus of POK by 17, 50, 60, and 72%, respectively. The effect of synergists at a weight fraction of 4 wt % on the tensile modulus of POK-10OF can be seen in Figure 1b,f,g,h. The addition of ZB increased the tensile modulus values of POK-10OF. It can be noted that the greatest tensile modulus value was obtained by adding 4 wt % ZB. Incorporation of harder materials, compared to the polymer matrix, such as ZB into the polymer matrix led to higher tensile modulus values. 13 3.2. Impact Strength. In general, when powder additives are added to polymer composites, it affects the mechanical values negatively. This is due to agglomeration of the flame retardant additive, poor interfacial interactions, imperfections, and voids. 14 Izod notched impact strengths of POK and POKbased composites containing flame retardant additives are given in Figure 3. When 10%, 15%, 20%, and 25 wt % flame retardant additives were added to the POK composites, it did not exhibit a noticeable effect on the Izod notched impact strength. The impact strength of the POK composite containing 10 wt % OF was obtained relatively higher than the other ratios. For this reason, 1%, 2%, and 4 wt % SIL and zinc borate as synergist materials were added to the POK composite group containing 10 wt % OF, and their effect on mechanical values was investigated. Impact values were positively affected by adding 2 wt % SIL to POK composites containing flame retardant additives. The addition of zinc borate did not have a positive effect on the impact values of POK composites containing 10 wt % OF. Figure 4, and the data obtained from Figure 4 are summarized in Table 1. The melting temperature (T m ), crystallization temperature (T c ), and melting enthalpy (ΔH m ) of the samples were observed with DSC analysis, and the degree of crystallinity (X c ) was calculated for each sample. 15 The degree of crystallinity (X c ) of the composites was calculated from the enthalpy of melting (ΔH) according to eq 1

DSC Analysis. DSC curves of samples are shown in
where ϕ is the weight fraction of POK in the composite, and ΔH m 0 is the melting enthalpy of the 100% crystalline POK polymer, reported as 227 J/g. 16 According to the literature data, the melting temperature for aliphatic polyketone polymers was found to be around 200°C . 17 The melting point of the polyketone polymer (POK) used in the study was measured as 193°C, the crystallization temperature was 145°C, and the ΔH m was measured as 26.2 J/g.
Melting temperatures of POK-10OF, POK-15OF, POK-20OF, and POK-25OF were measured as 171°C. Polymer composites with POK-10OF and SILMA/ZB additives at different weight fractions were studied, and the effect of additives on the melting temperature was evaluated. SILMA and ZB additives at all added proportions tended to increase the melting temperature of POK-10OF.
3.4. Thermogravimetric Analysis (TGA). TGA curves of samples are presented in Figure 5, and the data obtained from Figure 5 are summarized in Table 2. Considering the degradation temperatures of POK-10OF, POK-15OF, POK-20OF, and POK-25OF, the highest degradation temperature was achieved with POK-10OF. Polymer composites with the POK-10OF base and SIL and ZB added at different weight fractions were studied and their decomposition temperatures were evaluated. SIL did not significantly change the decomposition temperature of POK-10OF; however, the ZB addition caused a decrease in the decomposition temperature of POK-10OF by about 20−30°C. This is because the heat absorption capacity for ZB is lower than that for POK. When the weight ratio of ZB is increased, the threshold energy to initiate the degradation process is reached at lower temperatures, since ZB absorbs less heat in the composite. 18 When the mass losses occurring in the period up to 800°C are evaluated, the mass losses for POK-10OF, POK-15OF, POK-20OF, and POK-25OF are 62, 60, 59, and 61%, respectively. The mass loss was determined as 79% for POK. The addition of SIL and ZB additives significantly reduced the  considerably the temperature at 5% mass loss. It can be said that the thermal stability of the polymer decreased by the addition of FR materials according to the temperatures measured at mass losses and 5% mass loss. 19 3.5. Thermomechanical Analysis (TMA). The dimension change versus temperature was determined by TMA using the expansion mode. Thermal expansion coefficient (CTE) values of composites are given in Table 3. It was observed that the CTE value of POK decreased by adding FR materials. The decrease in the CTE value can be interpreted as an improvement in the dimensional stability of the polymer. 20 Likewise, it has been observed that SIL and ZB additives in all ratios are used to improve the dimensional stability of the polymer.
3.6. Flame Retardancy Properties of POK-Based Composites. The parameters obtained from the cone calorimeter are collected in Table 4. It is known that PHRR   23 It is known that the smoke density (SD) is considered an important parameter to evaluate the performance of smoke suppression, and a low SD value indicates a high performance of smoke suppression. 24 Smoke density values of POK, POK-    24 In other words, ZB increases the cross-linking network in the char layer. 25 It can be said that the effective char formation results in excellent smoke suppression 24 and ZB addition led to lower smoke density values.
Other indices for fire hazards of the POK-based composites are FIGRA (equal to the value of PHRR/time to PHRR) and MARHE. 26 From Table 4, 10 wt % OF addition into POK decreased MARHE and FIGRA values by 32 and 64%, respectively. When POK-10OF-4SIL and POK-10OF-4ZB are considered, it is seen that ZB leads to lower FIGRA and MARHE values. One can note that ZB is more efficient than SIL to decrease FIGRA and MARHE values. The THR is the integral of the HRR during the cone test and corresponds to the radiant flux levels. 27 10 wt % OF addition into POK decreased THR values by 10% and adding more OF led to lower THR values. With the addition of SIL and ZB, lower THR values were obtained. Moreover, ZB is more efficient than SIL in terms of the THR value. The EHC is the heat released from the combustion of the volatile portion of the material and is obtained by dividing the heat release rate by the mass loss rate. The higher smoke production and lower EHC value indicated that noncombustible gases exist in the gas phase; 22 10% OF addition into POK decreased the EHC value by about 12% and adding more OF led to lower EHC values. Table 5 shows the LOI and UL94 flame ratings of samples for a thickness of 3 mm. The neat POK has a HB burning class according to the UL94 standard. When a 10 wt % OP-based flame retardant additive was added to the neat POK, the flame rating reached a V0 rating class. Formation of specific phosphorus species by means of the degradation reaction of phosphinates makes the char layer covered on the residual sample surface to inhibit further fusion and combustion. 14 OF acted mainly through flame inhibition and phosphorus was released to the gas phase and induced the char formation in the residue. 28 While the FR ratio was increased up to 25 wt %, the V0 flame rating of POK compounds did not change for the studied thickness. The LOI value of POK was obtained to be 21.0. When 10, 15, 20, and 25 wt % OF were added to POK, the LOI values increased to 29.8, 36.7, 40.0, and 39.6%, respectively. Considering LOI and UL94 ratings, the FR ratio was selected as 10 wt %. To see the synergistic effect of ZB and SIL, 1, 2, and 4 wt % synergists together with 10 wt % FR composites have been produced. ZB and SIL addition did not lead to variations in the UL94 ratings for the studied thickness. However, LOI values increased considerably with the addition of ZB and SIL. LOI values of POK-10OF increased from 29.8 to 31.8 and 31.1% with the addition of 1 wt % of SIL and ZB, respectively. As can be seen from Table 5, adding a larger amount of SIL did not lead to an increase in the LOI values. However, the addition of 2 and 4 wt % of ZB increased the LOI values to 32.0 and 34.8%, respectively. One can note that ZB is more efficient than SIL to improve the LOI values. This result is correlated with the fact that ZB is more efficient than SIL to decrease the smoke density values. Since ZB is used as a flame retardant and smoke suppressant synergist, the combination of ZB with some flame retardant systems could enhance the char formation, which results in the improvement of flame retardancy. 25 Moreover, it can be added that ZB is used as a high-efficiency smoke inhibition agent and reduces the average CO production 29 3.7. SEM Analysis. SEM images of POK and its composites are presented in Figure 6. As can be seen from Figure 6a, a continuous matrix structure can be seen. It could be clearly seen from Figure 6b that the particle size of OF was from about 1 to 10 μm. Pull-out particles that may indicate poor adhesion 30 were observed. Since adhesion is weak, the particles pull away as the sample breaks. In Figure 6c, particles, which are in the range of about 800 nm to 3 μm, and holes from 1 to 3 μm, can be seen. Figure 6d shows that large  particles are available within the structure. This may be attributed to the agglomeration of fine ZB particles.

CONCLUSIONS
In this study, the restrictions about LOI and smoke density values for the interiors (R22) and exteriors (R23) of railway vehicles according to the European fire protection norm EN 45545:2013 were met. Using OF (10 wt %) and synergists ZB (4 wt %), an LOI value higher than 32% and a smoke density value lower than 150 for the hazard level HL3 were obtained. The addition of 10 wt % OF into POK is sufficient to obtain a V0 rating according to the UL94 standard. However, to obtain higher LOI values, more OF or synergists such as ZB and SIL must be added to POK. Additionally, for lower smoke density, FIGRA, MARHE, and EHC values, instead of adding more OF, synergists such as ZB and SIL should be used. It can also be reported that ZB is more efficient than SIL to improve the LOI values and to decrease the smoke density, FIGRA, MARHE, PHRR, and THR values. On the basis of POK−OF composites, it can also be added that ZB is more efficient in smoke suppression than SIL. However, ZB led to lower Izod notched impact strength values than SIL.