Flame Retardancy and Thermal Stability of Materials 2018; 1: 1–13

Research Article

Hao Wu*, Rogelio Ortiz, Renan De Azevedo Correa, Mourad Krifa, Joseph H. Koo Self-Extinguishing and Non-Drip Flame Retardant 6 Nanocomposite: Mechanical, Thermal, and Combustion Behavior https://doi.org/10.1515/flret-2018-0001 Received June 7, 2017; accepted October 9, 2017 1 Introduction

Abstract: Incorporation of flame-retardant (FR) additives A major drawback of flame retarding thermoplastic and nanoclay fillers into thermoplastic polymers by incorporating particulate additives consists effectively suppresses materials flammability and melt of mechanical property losses [1]. In particular, dripping behavior. However, it largely affects other 6 composite systems treated with non-halogenated properties, such as toughness and ductility. In order to intumescent fillers achieved self-extinguishing non- drip recover the lost toughness and ductility of flame retardant behavior but suffered from significant losses in ductility and polyamide 6, various loadings of maleic anhydride elongation at break [2, 3]. In previous studies, our research modified SEBS elastomer were added and processed by addressed this issue through elastomer toughening of twin screw extrusion. TEM images showed exfoliated the composite systems. Thus, flame retardant polyamide nanoclay platelets and reveals that the clay platelets well 6 was toughened using a maleated triblock copolymer dispersed in the matrix. By balancing the ratio of containing styrene-hydrogenated butadiene-styrene flame retardants, nanoclay and elastomers, formulation (SEBS-g-MA). This approach achieved successful recovery with elongation at break as high as 76% was achieved. of ductility as evidenced by a significant improvement Combining conventional intumescent FR and nanoclay, in Izod impact strength and elongation at break, while UL-94 V-0 rating and the LOI value as high as 32.2 were preserving the flame-retardant performance [4]. In the achieved. In conclusion, effective self-extinguishing and current research, we further explore the combined effect non-drip polyamide 6 nanocomposite formulations with of rubber toughening and incorporation of nanoclay significant improvement in toughness and ductility were additives to compound a polyamide 6 nanocomposite achieved. system with self-extinguishing non-drip performance and recovered ductile property. Keywords: flame retardant, rubber toughening, polyamide The efficiency of nanoparticles in modifying material 6, nanocomposite properties has led to intensifying research interests on polymeric nanocomposites in recent decades. In the area of flame retardant applications, one of the most effective and widely used nano-fillers is nanoclay. Among different types of clay, the montmorillonite (MMT) nanoclay with surface treatment is commonly used to achieve good *Corresponding author: Hao Wu, Materials Science and interfacial properties and high levels of dispersion in the Engineering, Texas Materials Institute, The University of Texas at Austin, 204 E. Dean Keeton Street; Stop C2201 Austin, TX 78712, polymer matrix. For instance, cation exchange is often USA, E-mail: [email protected] carried out on the nanoclay before processing with the Joseph H. Koo, Materials Science and Engineering, Texas Materials polymer matrix [5, 6]. Numerous studies have demonstrated Institute, The University of Texas at Austin, 204 E. Dean Keeton that the incorporation of nanoclay even at very low Street; Stop C2201 Austin, TX 78712, USA loadings could greatly decrease the peak heat release Rogelio Ortiz, Renan De Azevedo Correa, Joseph H. Koo, Department of Mechanical Engineering, The University of Texas at Austin, 204 E. rate as characterized by cone calorimetry or microscale Dean Keeton Street, Stop C2200, Austin TX 78712, USA combustion calorimetry, by creating a carbonaceous- Mourad Krifa, Kent State University, 800 E. Summit St. Kent, OH silicate char layer [2, 7-9]. This clay-rich char layer could be 44240 caused by either pyrolysis of the polymer leaving the clay

Open Access. © 2017 Hao Wu et al., published by De Gruyter Open. This work is licensed under the Creative Commons Attribution- NonCommercial-NoDerivs 3.0 License. 2 H. Wu, et al. platelets on the surface or bursting bubbles of the polymer of loss of toughness and ductility, rubber toughening melt pushing the clay towards the surface [10, 11]. is used as an approach to tailor mechanical behavior. However, materials incorporating nano-fillers alone Details about structural, morphological, mechanical have been found to fall short in flammability tests such as and flammability characterization of the nanocomposite UL-94 [12, 13]. On the other hand, combinations of nano- material are discussed in this paper. additives and conventional flame retardants, such as intumescent fillers have shown potential synergistic effects and high performance in reducing total heat release, which 2 Methods equates to the highest UL-94 rating [6, 9, 14-16]. Another challenge for flame retardant incorporated polymer 2.1 Materials composites is the significant reduction in mechanical properties, such as ductility and elongation at break. Medium viscosity nylon 6 (density 1.13 g/cm3), was obtained The majority of the past researches in flame retardant from Honeywell, Inc. and dried at 80°C for 24 hours prior polymers only focused on reduction in flammability. to processing. Flame retardant additive Exolit® OP1312 However, a balance of properties, such as mechanical, consisting of a mixture of aluminium diethyl phosphinate flammability, and cost are often required in commercial (AlPi), melamine polyphosphate (MPP), zinc and boron systems. Only a few publications have addressed this oxide (ZnB) [18] was provided by Clariant International issue. Shi et al. modified MMT nanoclay with a cationic Ltd. The SEBS elastomer Kraton FG1901 G was provided by vinyl acetate copolymer to make a masterbatch and Kraton Polymers, Inc. Montmorillonite (MMT) organoclay then mix with Ethylene vinyl acetate (EVA). They found Cloisite 30B was provided by Southern Clay Products. desirable flammability and mechanical properties in these Cloisite 30B is produced by cation exchange and treated nanocomposites [17]. Thermoplastic elastomers can be with 90 mequiv./100 g clay of Ethoquad T12 with methyl used as modifiers for engineering thermoplastics including bis-2- hydroxyethyl tallow quaternary ammonium chloride to tailor the mechanical properties. In our previous [19]. Flame retardant, elastomer, and nanoclay were used work, a maleic anhydride modified SEBS elastomer was as received. incorporated into FR nylon 6 the system by twin-screw melt extrusion. The addition of the elastomer succeeded in significantly enhancing the Izod impact strength and 2.2 Compounding partially recovering the elongation at break without compromising the flame-retardant performance [4]. The formulations compounded are shown in Table 1. The objective of this research is to prepare a Previous studies on rubber toughened flame retardant nano-filled nylon 6 (PA6) with balanced properties in polyamide 6 have shown that 15wt% or lower elastomer flammability and mechanical properties. To optimize its loading would generate consolidated char layer and flame retardant properties a combination of nanoclay retain the UL-94 V-0 rating [4]. Considering the potential and non-halogenated conventional intumescent flame synergistic effect between OP1312 and nanoclay, the retardant additives were used. To overcome the problem FR loading was lowered to 15wt% in all formulations

Table 1. Formulation table Sample # Designation Flame Retardant Kraton Elastomer Nanoclay Nylon 6 (wt%) (wt%) (wt%) (wt%)

1 Neat - - - 100 2 20FR 20 0 0 80 3 15FR_5KR_2.5NC 15 5 2.5 77.5 4 15FR_5KR_5NC 15 5 5 75 5 15FR_10KR_2.5NC 15 10 2.5 72.5 6 15FR_10KR_5NC 15 10 5 70 7 15FR_15KR_2.5NC 15 15 2.5 67.5 8 15FR_15KR_5NC 15 15 5 65 Self-Extinguishing and Non-Drip Flame Retardant Polyamide 6 Nanocomposite 3 containing nanoclay. Two nanoclay loadings were chosen 3.3 Thermal Properties (2.5wt% and 5wt%) in order to provide enough post burning protective char layer. Thermogravimetric analyses (TGA) of all samples were All formulations were processed using a Thermo- performed using a TGA-50 from Shimadzu Scientific Scientific Process 11 parallel co-rotating twin screw Instruments. The samples were heated in air environment extruder (L/D=40). The co-rotating twin screw extruder from room temperature to 1000°C at a heating rate of provides high shear mixing which ensures proper 10°C/min. The air flow rate was 20ml/min. dispersion of the nano-fillers [20]. Materials were melted DSC was carried out using a Q20 from TA Instruments. and mixed at 240oC, with screw speed at 150 rpm. To For the DSC tests, injection molded samples were cut into ensure a homogenous dispersion, formulations with 7-8 mg square pieces and were directly put into the crucible varying concentrations were pre-mixed using a Thinky® for testing. To erase the thermal history, all samples were planetary centrifugal mixer prior to melt compounding. first heated from room temperature to 260°C at 10°C/min; Extruded composites were water cooled, pelletized, and after holding at 240°C for 2 min, the samples were cooled dried before injection-molding into ASTM D638 standard down to room temperature at -10°C/min. A second heating tensile bars and UL-94 specimens. A Mini-Jector injection cycle then started at the same rate to the maximum of molding system with barrel temperature at 270oC and 260°C. The flow rate was held constant at 50 ml/ mold temperature at 60oC was used for injection molding min at all time. To calculate the percent of crystallinity, test specimens. the heat of fusion ΔHf0 for pure crystalline of 240 J/g was

used. The ΔHf0 values for each sample are based on the 3 Characterization amount of pure PA6 within the formulation.

3.1 X-ray Diffraction (XRD) 3.4 Flammability Characterization

Wide-angle X-ray diffraction (WAXD) was carried out Thermal combustion properties were measured using a on both injection molded samples and melt-spun fibers Micro-scale Combustion Calorimeter (MCC2, Govmark, using a Rigaku R-Axis Spider with curved image plate (Cu Inc.) according to ASTM D7309-2007. MCC directly Kα radiation). A small section (~1 mm) of the injection measures heat release parameters of small amounts of molded samples was cut from the surface and directly materials in the size of 2-5 milligrams. The heat release placed on the sample holder with mineral oil. All samples rate during combustion and the total heat release are were set to rotate at 10°/sec during scanning with 10min. calculated by oxygen consumption rate and integration of the heat release rate versus time. The combustor temperature was held constant at 900°C, and the heating 3.2 Microstructure Characterization by TEM rate of the pyrolysis zone was 1°C/sec. Each sample was tested in 3 repetitions to calculate error bars. The elastomer toughened FR system and nanoclay UL-94 is a commonly used industry standard to test dispersion were characterized by TEM. Ultrathin sections the flammability of materials or their tendency of the extruded composite samples were obtained by to either self-extinguish or spread the flame once the cryogenic microtoming using a Leica Ultramicrotome with specimen has been ignited. In the vertical UL-94 test, a a diamond knife at -110°C. The sections were then placed 1.27 cm × 12.7 cm (1/2” × 5”) specimen is held in the vertical on 400 mesh copper grids. The grids were transferred position at one end as a burner flame is applied at its free into an FEI Tecnai Transmission Electron Microscope end for two 10-second intervals, separated by the time (TEM) setting accelerating voltage at 80 kV for phase it takes for flaming combustion to cease after the first contrast imaging. TEM images of samples under two application. The three ratings, V-2, V-1 and V-0 indicate different conditions were taken. To observe the flame that the material was tested in a vertical position and self- retardant particles, sample sections were directly put extinguished within a specified time after the ignition into TEM for observation whereas in order to resolve the source was removed. These ratings also indicate whether elastomer/nylon interface, samples were stained with the test specimen dripped flaming particles that ignited

2% phosphotungstic acid (PTA) for 30min to let the WO3 a cotton indicator located below the sample [21]. A UL-94 penetrate into the amorphous phase of nylon. Stained V-0 rating is indicative of a self-extinguishing non-drip samples were rinsed with water and dried before imaging. performance. 4 H. Wu, et al.

Limiting Oxygen Index (LOI) measures the minimum Neat nanoclay powder exhibits distinct basal reflections of concentration of oxygen that will support combustion of the stacked clay structure with 2θ peaks at 4.8°, 19.7°, and a test material. LOI was conducted in accordance with 35.2° corresponding to interlayer d-spacing of 18.7 Å, 4.6 ASTM D2863. All samples were conditioned at 25°C and Å, and 2.6 Å, respectively. Since the intensity of the lower 50% relative humidity for at least 48 hours before testing. angle peak at 4.8° is significantly higher than the other LOI tests were carried out using a lab made LOI two, it is considered as the main crystalline d-spacing apparatus in accordance with ASTM D2863 . Two digital of the nanoclay used in this study. After twin screw flow meters control the flow rate of oxygen and nitrogen extrusion, the reflection peaks of nanoclay disappeared as they are mixed in the glass tube. Samples were ignited for all nanoclay formulations. The disappearance of with a lighter and the burning time was recorded. nanoclay reflection peaks suggests that high level exfoliation of nanoclay platelets was achieved. This observation is further confirmed by TEM results which 3.5 Tensile Properties will be discussed in the following section. Moreover, it is

Tensile tests were carried out in accordance to ASTM D638 using an Instron 5966. The crosshead speed was 5 mm/min and the distance between grips was 115 mm. All samples were conditioned at 25°C and 50% relative humidity for at least 48 hours before testing. Average values and error bars were calculated based on 3 tests on each formulation.

3.6 Izod Impact Test

Izod impact strength of selected samples was tested using a Shanta Engineering Izod/Charpy Impact Tester in accordance with ASTM D256. Five specimens were tested for each formulation. All specimens were conditioned at 25°C and 50 % relative humidity for at least 48 h before testing. Figure 1. Schematic of UL-94 vertical burning test.

3.7 Char Morphology Characterization

High resolution SEM characterization of char morphology was carried out using a Hitachi S-5500 STEM equipped with Energy-dispersive X-ray spectroscopy (EDS) detector. Char residues from MCC tests were transferred to the SEM sample holder and directly put into the sample chamber without coating. The accelerating voltage for imaging was 5kV and in order to get a better signal to noise ratio, 30kV was used for EDS. Bruker ESPRIT software was used for image acquisition and data analysis.

4 Results and Discussion

4.1 XRD results

Figure 2. XRD patterns of neat nanoclay powders and PA6/FR/elas- Figure 2 shows XRD patterns of neat nanoclay powders tomer/nanoclay blend containing 15 wt% FR, 5 wt% elastomer, and and a FR PA6 formulation containing 5 wt% nanoclay. 5 wt% nanoclay. Self-Extinguishing and Non-Drip Flame Retardant Polyamide 6 Nanocomposite 5 noted that the incorporation of nanoclay had no influence Neat PA6 has tensile strength and Young’s modulus of on the crystalline structure of the PA6: γ-form crystal is 58.7 and 2175 MPa, respectively. As anticipated, addition still the dominant crystal structure as characterized by its of FR particles results in major losses in tensile strength 2θ peak at 10.9° and 21.3° correspond to (020) and (002) and elongation (Figures 4 and 6) and in increased Young’s planes of the γ phase crystal of PA6 [22]. The distinct peaks Modulus (Figure 5), signaling a brittle behavior. When of FR additives at 9.1°, 14.7°, 18.6°, and 26.3° suggest that comparing formulations with same amount of FR and FR crystals are mixed within the PA6 without changing its elastomer, higher nanoclay loadings appear to yield crystalline structure. higher tensile modulus (Figure 5). On the other hand, it is apparent that higher nanoclay loadings negatively impact the elongation at break (Figure 6). With the 15wt% 4.2 TEM results elastomer loading, sample 7 exhibits recovery of the highest elongation at break of 76%. Figure 3 shows high magnification TEM images of sample #5 containing 15wt% flame retardant, 10wt% elastomer, and 5wt% nanoclay. Although intercalated nanoclay platelets exist within the sample, most nanoclay structures are exfoliated platelets uniformly dispersed throughout the polymer matrix. Flame retardant particles embedded in the polymer matrix can also be seen in Figure 3. The distribution of the flame retardant particles and the elastomers are uniform throughout the sample which is similar to the rubber toughened flame retardant PA6 formulations without nano-additives which have been published in [4].

4.3 Tensile properties

Table 2 summarizes the mechanical properties of neat PA6 and all formulations containing flame retardant additives. Figure 4. Tensile strength of FR PA6 formulations.

Figure 3. TEM of FR PA6 nanocomposite. 6 H. Wu, et al.

Table 2. Tensile properties of FR PA6 formulations

Sample # Composition Tensile strength (MPa) Young’s Modulus (MPa) Elongation at break (%)

1 Neat PA6 58.7±0.5 2175±71.1 205.4±12.6 2 20FR 47.8±0.3 2980±46 4.3±0.9 3 15FR_5KR_2.5NC 45.1±0.5 2420±40 16.9±1.0 4 15FR_5KR_5NC 48.3±0.5 2730±29 6.8±0.7 5 15FR_10KR_2.5NC 41.0±0.3 2190±29 27.6±3.7 6 15FR_10KR_5NC 42.7±0.2 2450±25 12.0±0.6 7 15FR_15KR_2.5NC 39.8±0.6 1770±65 76.0±2.7 8 15FR_15KR_5NC 37.9±0.3 2012±22 20.7±1.9

Figure 5. Young’s modulus of FR PA6 formulations. Figure 6. Elongation at break of FR PA6 formulations.

4.4 Izod Impact Strength

Izod impact strength of neat polyamide 6 and its nanocomposites are compared in Figure 7. Addition of 20wt% flame retardant significantly affected the toughness of the compound causing a more than 25% decrease in impact strength from 11.4 J/m2 to 8.8 J/m2. With increasing amount of elastomer, the impact strength is recovered and at 15wt% elastomer loading, the impact strength is 8% higher than the neat polyamide 6. On the other hand, higher nanoclay loadings tend to have lower impact strength except for nanocomposite samples containing 15wt% FR and 10wt% elastomer. Such deviation may not be statically significant due to the large error bars. Figure 7. Izod impact strength of polyamide 6 and its nanocomposites. 4.5 Thermal properties Table 3. The 10wt% mass loss temperature for neat nylon 6 is 402.1°C; after adding 20wt% of flame retardant, the Figure 8 shows that thermal degradation curves of 10wt% loss temperature dropped 14-15°C. The lower onset all formulations. The decomposition temperatures degradation temperature is common in polymer systems calculated from the thermographs are summarized in Self-Extinguishing and Non-Drip Flame Retardant Polyamide 6 Nanocomposite 7

Table 3. Decomposition temperatures of FR PA 6 formulations

Sample Tdec at 10wt% mass loss (°C) Tdec at 50wt% mass loss (°C) Residue at 1000°C (wt%)

1. neat PA6 402.1 438.5 0.51 2. 20FR 387.1 426.4 7.11 3. 15FR_5KR_2.5NC 392.1 431.8 8.16 4. 15FR_5KR_5NC 396.9 432.6 11.29 5. 15FR_10KR_2.5NC 396.1 431.8 10.13 6. 15FR_10KR_5NC 389.6 435.2 11.08 7. 15FR_15KR_2.5NC 400.3 436.1 7.59 8. 15FR_15KR_5NC 397.6 435.3 10.14

Figure 8. TGA curves of FR PA 6 formulations. Figure 9. DrTGA curves of FR PA 6 formulations. containing intumescent compounds [2, 23] and can be the combustion results which will be discussed in the next explained by the formation of phosphoric esters through section. the reaction between phosphate in the flame retardant Figure 10 shows the DSC first heating endotherms of and hydrocarbons in the nylon 6 [24]. samples taken from injection molded neat PA6 and five After heating the materials to 1000°C, neat nylon different FR formulations. The melting peak temperature

6 samples have only 0.5% of char residue left. The (Tm) and percent of crystallinity calculated from the char residue increased significantly to ~7-8wt% with heating scans are summarized in Table 4. The endotherms all formulations containing flame retardant. The from the first heating allows for examination of the concentration of elastomer appears to have minimum impact of different components in the FR formulation effect on the amount of char formations. on the crystallization behavior of the PA6 matrix. Neat Figure 9 shows that neat nylon 6 has a single mass PA6 has two distinct endothermic peaks at 220.1°C and loss peak at 442.5°C. Again because the flame retardants 225.1°C which correspond to the melting of γ crystals start to degrade at lower temperature, the mass loss peak and α crystals, respectively. The small sloped shoulder for all flame-retardant-modified formulations has shifted before the melting peaks may be caused by the presence to a lower temperature around 400°C. It can also be found of γ crystals. Note that only γ crystals are detected from from Figure 9 that adding elastomer slightly increased XRD results, this may be caused by the nature of the core- the peak mass loss rate while the overall mass loss rates skin structure of the injection molded materials [25]. The are still lower than neat nylon 6. This slight increase in skin part undergoes faster cooling and higher stress; mass loss rate could be caused by rapid degradation of the therefore, γ crystal is dominant. The core experiences elastomers; however, the same trend was not observed in slower cooling and less stress; as a result, a fair amount 8 H. Wu, et al. of α crystal is formed [25]. From the heating scans, neat PA6 has relatively low crystallinity which could be due to the lack of nucleus or fast cooling during the injection molding process. Adding 20wt% of FR to the neat polymer diminished the separation of the two crystal melting peaks leaving only one major peak at 223.1°C, which corresponds to the melting of α crystal. The lower melting peaks may also be attributed to the decreased crystallite thickness of PA6 [25]. Elastomer addition seems to have no influence on the crystallization behavior of PA6 as no significant difference is observed in the exotherms. On the other hand, formulations containing nanoclay show a notable lower temperature peak corresponding to the γ crystal. Figure 11 compares the cooling scans for the same set of samples and Table 4 summarizes the peak crystallization temperature denoted as Tc. Unlike the heating scans, all cooling scans have only one exothermic peak. The neat Figure 10. First DSC heating scans of different injection molded FR PA6 has a crystallization peak at 182.7°C which is slightly PA6 formulations (exotherms up). Samples were heated at 10°C/ lower than its FR formulations. Under a controlled min. Scans are shifted for clarity. cooling rate of 10°C/min, all formulations yielded lower crystallinity than the neat PA6. Addition of 20wt% FR increased the crystallization temperature to 185°C which suggests that the FR particles could lower the free energy barrier to crystallization by facilitating the heterogeneous nucleation within PA6. Samples containing elastomer have slightly lower crystallization temperature around 183.9°C this could be due to the finely dispersed elastomer phases hindering the chain mobility of PA6 which result in slower crystallization process. Adding nanoclay to the blend led to a small increase in crystallization temperature as compared to the PA6/FR/elastomer blend. This may be because of the well-known nucleation effect for PA6/clay nanocomposites [25, 26].

4.6 Flammability

Figure 11. First DSC cooling scans of different injection molded FR Rubber toughening of the FR PA6 formulations has PA6 formulations (exotherms up). Samples were cooled at 10°C/ shown promising results characterized by the MCC [4]. min. Scans are shifted for clarity.

Table 4. Summary of percent of DSC data for different injection molded FR PA6 formulations. (Xc,1 is the percent crystallinity calculated by integration of the first heating peak, Xc,2 is the percent crystallinity calculated by integration of the first cooling peak) Sample Tm (°C) Tc (°C) Χc,1* (%) Χc,2* (%)

neat PA6 222.1/225.1 182.7 21.8 25.1 20FR 223.1 185.0 24 24.1 20_5 223.0 183.8 22.8 24.2 20_10 222.2 183.9 24.1 23 15_5_2.5 213.1/222.6 184.2 21.3 24.5 15_5_5 211.4/221.2 184.3 20.8 23.5 Self-Extinguishing and Non-Drip Flame Retardant Polyamide 6 Nanocomposite 9

Typical MCC heat release curves of PA6/FR/elastomer formulations are shown in Figure 12. Because of the nature of the intumescent FR additive [27], all FR formulations have lower onset degradation temperature than the neat PA6. Addition of the elastomers does not change the degradation temperature, but a small decrease in peak heat release rate is observed. The impact of nanoclay addition is depicted in Figure 13, and Table 5 showing Heat Release Capacity (HRC) results, and heat release curves of neat PA6 and seven different flame retardant formulations. With 15wt% FR loadings in all nanocomposite samples, lower HRC than the original 20FR sample was achieved in all formulations, which is indicative of the enhancement effect between FR and nanoclay. At 10wt% and 15wt% elastomer level, higher concentration of nanoclay clearly leads to lower HRC. This is due to the barrier effect of a clay reinforced char layer which will be discussed in the char morphology section. However, this trend does not Figure 12. Typical heat release curves of PA6 and its apply to samples with low elastomer concentration, i.e. nanocomposites. 5wt%. All formulations combining FR, elastomer, and nanoclay achieved V-0 rating except the one with lowest elastomer and nanoclay loading as shown in Table 6. Pictures of post UL-94 test specimens of neat nylon 6 and 20FR are shown in Figure 14. Although neat nylon 6 self– extinguished after removal of fire, melted nylon 6 dripped and ignited the cotton indicator underneath it; therefore, only a V2 rating was attributed. After adding 20wt% flame retardant, the dripping phenomenon was effectively suppressed and combustion ceased almost immediately after flame removal thus it is rated as V-0. Table 6 summarizes the UL-94 ratings of all nanocomposite formulations. Figure 15 shows the post UL-94 test specimens of the nanocomposite formulations. After removal of the flame source, samples with 15wt% FR, 5wt% elastomer and 2.5wt% nanoclay kept burning Figure 13. Heat release capacity of PA6 and its nanocomposites.

Table 5. Summary of flame retardant properties of PA6 and its nanocomposites

Sample Heat Release Capacity (J/g-K) Peak HRR (W/g) Total HR (kJ/g)

Neat 604±4.3 705.1±4.9 29.9±0.1 20FR 407.6±6.5 472.7±7.9 26.7±0.1 15FR_5KR_2.5NC 383.6±1.7 537.8±2.6 28.5±0.1 15FR_5KR_5NC 391.8±3.5 552.8±4.8 28.0±0.1 15FR_10KR_2.5NC 394.2±1.1 557.0±1.4 29.3±0.0 15FR_10KR_5NC 368.2±4.2 517.5±5.7 28.1±0.1 15FR_15KR_2.5NC 393.8±3.0 554.6±4.0 29.2±0.1 15FR_15KR_5NC 368.6±3.2 520.9±4.3 28.4±0.1 10 H. Wu, et al.

Figure 14. UL-94 posttest specimens of (a) neat and (b) 20FR

Table 6. Summary of UL-94 ratings of FR PA6 formulations Sample UL-94 rating LOI

Neat PA6 V2 26.6

20FR V0 30.5 15FR_5KR_2.5NC V1 29.5 15FR_5KR_5NC V0 32.2 15FR_10KR_2.5NC V0 28.9 15FR_10KR_5NC V0 31.7 15FR_15KR_2.5NC V0 29.0 15FR_15KR_5NC V0 29.3

for prolonged time, therefore only V-1 was assigned. coherent thick char layer covers most of the surface with Meanwhile all other nanocomposite formulations self- the exception of some air bubbles. extinguished immediately and thus achieved a V-0 rating. Cross sectional views of the sample in Figure 18 show No dripping was observed with any of the nanocomposite a char layer with thickness greater than 20 µm. Such FR samples. a thick char layer may act as a better thermal shield, LOI results of the rubber toughened nanocomposite protecting virgin materials underneath the pyrolysis zone. FR PA6 samples are shown in Figure 16 and Table 7. Even Higher magnification images show closely packed char though neat PA6 has an LOI of 26.6 which is higher than structures of clay layers. It needs to be noted that such the oxygen concentration in air, the melt dripping remains char does not only exist on the surface of the specimen; a problem. All FR formulations have LOI higher than the it was also observed beneath the surface forming a 3D neat PA6 indicating better flame retardant properties. cellular structure throughout the char thickness. Higher loadings of elastomer additives tend to result in lower LOI values. However, this reduction in LOI appears compensated by the incorporation of nanoclay. Indeed, 5 Conclusion when comparing formulations with same amount of FR Elastomer toughened FR nylon 6 nanocomposite systems and elastomer, higher loadings of nanoclay yield higher with varying elastomer and nanoclay concentrations LOI values. were prepared by twin screw melt blending. The average Figure 17 below shows the post UL-94 char surface elastomer particle size is around 100nm as revealed by of nanocomposite sample 15_5_5 containing 15wt% TEM micrographs. Addition of the elastomer not only flame retardant, 15wt% elastomer, and 5wt% nanoclay. A significantly improved the Izod impact strength, it also Self-Extinguishing and Non-Drip Flame Retardant Polyamide 6 Nanocomposite 11

Figure 15. UL-94 post-test specimens of all nanocomposite formulations: (a) 15FR_5KR_2.5NC, (b) 15FR_5KR_5NC, (c) 15FR_10KR_2.5NC, (d) 15FR_10KR_5NC, (e) 15FR_15KR_2.5NC, and (f) 15FR_15KR_5NC.

helped recover ductility lost caused by the presence of the heat release capacity. In the case of UL-94 tests, all large FR particles. Both flame retardant additives and nanocomposite formulations achieved UL-94 V-0 except elastomers degrade the tensile strength of the neat nylon for the one with lowest elastomer and nanoclay loading. 6. However, flame retardant enhanced the modulus, and Incorporation of nanoclays also led to significant increase elastomers effectively recovered elongation at break. in LOI values through all formulations. In conclusion, Addition of flame retardants and elastomer PA6/FR/elastomer/nanoclay nanocomposite blends changed the melting and re-crystallization temperature could provide effective flame retardant performance with characterized by DSC. Micro-scale combustion calorimetry improved toughness and preserved ductility for the fiber tests show that adding nanoclay could effectively lower spinning process. 12 H. Wu, et al.

Figure 16. LOI comparison of rubber toughened FR PA6 nanocomposites.

Figure 17. Top view of post UL-94 char morphology of sample #12 containing 15wt% flame retardant, 15wt%Kraton elastomer and 5wt% nanoclay.

Figure 18. Cross sectional view of post UL-94 char morphology of sample #12 containing 15wt% flame retardant, 15wt%Kraton elastomer, and 5wt% nanoclay. Self-Extinguishing and Non-Drip Flame Retardant Polyamide 6 Nanocomposite 13

posites: thermal and flammability properties, J. Compos. References Mater. 43(17) (2009) 1803-1818. [15] Yin X., Wu H., Krifa M., Londa M., Koo J.H., Flame retardant [1] Weil E.D., Levchik S., Current Practice and Recent Commercial polyamide 6-MWNT nanocomposites: Characterization and Developments in Flame Retardancy of , J. Fire Sci. processing via electrospinning, SAMPE TECH 2011, Fort Worth, 22(3) (2004) 251-264. TX, 2011. [2] Wu H., Krifa M., Koo J.H., Flame retardant polyamide 6/ [16] Yin X., Krifa M., Koo J.H., Flame-retardant polyamide 6/carbon nanoclay/intumescent nanocomposite fibers through electro- nanotube nanofibers: Processing and characterization, J. Eng. spinning, Textile Res. J. 84(10) (2014) 1106-1118. Fiber. Fabr. 10(3) (2015). [3] H. Vahabi, J.M. Lopez-Cuesta, C. Chivas-Joly, 6 - [17] Y. Shi, T. Kashiwagi, R.N. Walters, J.W. Gilman, R.E. Lyon, D.Y. High-performance fire-retardant polyamide materials A2 Sogah, Ethylene vinyl acetate/layered silicate nanocom- - Wang, De-Yi, Novel Fire Retardant Polymers and Composite posites prepared by a surfactant-free method: Enhanced flame Materials, Woodhead Publishing, 2017, pp. 147-170. retardant and mechanical properties, Polymer 50(15) (2009) [4] Wu, H., Krifa M. and Koo J.H., 2017. Rubber (SEBS-g-MA) 3478-3487. Toughened Flame Retardant Polyamide 6: Microstructure, [18] J. Alongi, M. Ciobanu, G. Malucelli, Novel flame retardant Combustion, Extension, and Izod Impact Behavior. Polymer- finishing systems for cotton fabrics based on phosphorus- Technology and Engineering,2017, 1-13. containing compounds and silica derived from sol–gel [5] Paul D.R., Robeson L.M., Polymer nanotechnology: Nanocom- processes, Carbohydrate Polymers 85(3) (2011) 599-608. posites, Polymer 49(15) (2008) 3187-3204. [19] Ahn Y.-C., Paul D.R., Rubber toughening of nylon 6 nanocom- [6] Laoutid F., Bonnaud L., Alexandre M., Lopez-Cuesta J.M., posites, Polymer 47(8) (2006) 2830-2838. Dubois P., New prospects in flame retardant polymer materials: [20] S.C. Lao, W. Yong, K. Nguyen, T.J. Moon, J.H. Koo, L. Pilato, From fundamentals to nanocomposites, Mat Sci Eng R 63(3) G. Wissler, Flame-retardant polyamide 11 and 12 nanocom- (2009) 100-125. posites: processing, morphology, and mechanical properties, J. [7] Gilman J.W., Kashiwagi T., Lichtenhan J.D., Nanocomposites: Compos. Mater. 44(25) (2010) 2933-2951. a revolutionary new flame retardant approach, SAMPE Journal [21] Lao S.C., Koo J.H., Moon T.J., Londa M., Ibeh C.C., Wissler G.E., 33(4) (1997) 40-46. Pilato L.A., Flame-retardant polyamide 11 nanocomposites: [8] Gilman J.W., Flammability and thermal stability studies of further thermal and flammability studies, J. Fire Sci. 29(6) polymer layered-silicate (clay) nanocomposites, Applied Clay (2011) 479-498. Science 15(1–2) (1999) 31-49. [22] G. Gururajan, S.P. Sullivan, T.P. Beebe, D.B. Chase, J.F. Rabolt, [9] Wu H., Yin X., Krifa M., Londa M., Koo J.H., Fabrication and Continuous electrospinning of polymer nanofibers of Nylon-6 Characterization of Flame Retardant Polyamide 6 Nanocom- using an atomic force microscope tip, Nanoscale 3(8) (2011) posites via Electrospinning, SAMPE TECH 2011, Fort Worth, TX, 3300-3308. 2011. [23] S. Bourbigot, S. Duquesne, Fire retardant polymers: recent [10] T. Kashiwagi, R.H. Harris, X. Zhang, R.M. Briber, B.H. Cipriano, developments and opportunities, Journal of Materials S.R. Raghavan, W.H. Awad, J.R. Shields, Flame retardant Chemistry 17(22) (2007) 2283-2300. mechanism of polyamide 6-clay nanocomposites, Polymer [24] H.-Q. Peng, Q. Zhou, D.-Y. Wang, L. Chen, Y.-Z. Wang, A novel 45(3) (2004) 881-891. charring agent containing caged bicyclic phosphate and its [11] P. Kiliaris, C.D. Papaspyrides, Polymer/layered silicate (clay) application in intumescent flame retardant polypropylene nanocomposites: An overview of flame retardancy, Progress in systems, Journal of Industrial and Engineering Chemistry 14(5) Polymer Science 35(7) (2010) 902-958. (2008) 589-595. [12] Bourbigot S., Duquesne S., Jama C., Polymer Nanocomposites: [25] Fornes T.D., Paul D.R., Crystallization behavior of nylon 6 How to Reach Low Flammability?, Macromol. Symp. 233(1) nanocomposites, Polymer 44(14) (2003) 3945-3961. (2006) 180-190. [26] Wu T.-M., Chen E.-C., Liao C.-S., Polymorphic behavior of nylon [13] Bourbigot S., Duquesne S., Fontaine G., Bellayer S., Turf T., 6/saponite and nylon 6/montmorillonite nanocomposites, Samyn F., Characterization and Reaction to Fire of Polymer Polymer Engineering & Science 42(6) (2002) 1141-1150. Nanocomposites with and without Conventional Flame [27] U. Braun, H. Bahr, B. Schartel, Fire retardancy effect of Retardants, Molecular Crystals and Liquid Crystals 486(1) aluminium phosphinate and melamine polyphosphate in glass (2008) 325-339. fibre reinforced polyamide 6, e-Polymers (2010). [14] Lao S.C., Wu C., Moon T.J., Koo J.H., Morgan A., Pilato L., Wissler G., Flame-retardant polyamide 11 and 12 nanocom-