Performance of Clays, Carbon Nanofibers, Multi-walled Carbon Nanotubes, and Nano-alumina in Polyamide 11 Nanocomposites

Transcription

1 51st AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference<BR>18th April 2010, Orlando, Florida AIAA Performance of Clays, Carbon Nanofibers, Multi-walled Carbon Nanotubes, and Nano-alumina in Polyamide 11 Nanocomposites J. H. Koo 1, S. C. Lao 2, J. C. Lee 3, C. K. Lam 4, J. Lee 5, and T. J. Moon 6 The University of Texas at Austin, Austin, TX , USA L. A. Pilato 7, and G. E. Wissler 8 KAI, LLC, Austin, TX , USA P The objective of this research is to develop an improved polyamide 11 polymer with enhanced flame retardancy, thermal, and mechanical properties for selective laser sintering rapid manufacturing. In the present study, a nanophase was introduced into the polyamide 11 and combine with a conventional intumescent flame retardant (FR) additive via twin screw extrusion. Polyamide 11 polymer pellets were blended separately with chemically modified montmorillonite nanoclays (NCs), or carbon nanofibers (CNFs), or multi-walled carbon nanotubes (MWNTs), or nano-alumina (NA) and an intumescent FR additive to create a family of intumescent FR polyamide 11 nanocomposites. Transmission electron microscopy (TEM) analyses indicated good dispersion of individual nanoparticles and intumescent FR additives were achieved in the polymer matrix. Injection molded specimens were fabricated for physical, thermal, flammability, and mechanical properties characterization. Thermal stability of these polyamide 11 nanocomposites was examined by TGA. All systems of polyamide 11 containing nanoparticles and FR additive have higher decomposition temperatures than those compositions with solely nanoparticles or FR additive alone. None of the nanoparticles alone was able to pass UL 94 V-0 rating. Selective FR-PA11-nanoclay and FR-PA11-CNF nanocomposites passed the UL 94 V-0 requirement and suggest synergism between FR and selected nanoparticles. Some trends and relationships were observed between the results of TGA, UL 94, heat deflection temperature, and cone calorimetry. I. Introduction olymer nanocomposites (PNCs) are used in a variety of applications [1-3]. Because the nanoparticles do not simply provide enhancement for any one single polymer material property, thus in many cases multifunctional polymer material properties result. This encourages different material formulations to be explored in order to determine the optimal material for a specific application. Many of these PNCs are used in applications that require high strength, lightweight, and durability in high temperature environments. Flame retardant (FR) additives such as inorganic metal oxides/hydroxides or halogens with or without phosphorous and nitrogen containing materials are required in conventional methods to modify flammable thermoplastic materials as FR products [4]. Large amounts of FR additives (>30%) are necessary when using these methods to 1 Senior Research Scientist/Director of Solid Freeform Fabrication Consortium, Department of Mechanical Engineering, Texas Materials Institute, 1 University Station/C2200, and AIAA Associate Fellow. Corresponding author: jkoo@mail.utexas.edu 2 PhD Graduate Research Assistant, Materials Science and Engineering, Texas Materials Institute. 3 PhD Graduate Research Assistant, Department of Mechanical Engineering, and AIAA Student Member. 4 Undergraduate Research Student, Department of Mechanical Engineering, and AIAA Student Member. 5 Visiting Scholar, Department of Mechanical Engineering. 6 Professor, Department of Mechanical Engineering. 7 Chief Scientist. 8 Senior Scientist. Copyright 2010 by Copyright 2010 by J. H. Koo. Published by the, Inc., with permission.

2 make FR thermoplastics. In many cases a reduction of mechanical properties, such as toughness, melt flow, etc. and/or release of smoke and toxic emissions, occurs when the modified thermoplastic is burning. The incorporation of nanoparticles has been shown to be an effective approach for developing FR thermoplastic polymer by twin-screw extrusion (a melt blending process). Small amounts of nanoparticles (<7%) are required to prepare nanocomposites that exhibit similar enhanced flammability properties when compared with the modified thermoplastic processed by conventional methods [5-16]. However, unlike the conventional FR thermoplastics, the resulting nanocomposites exhibit enhanced mechanical properties such as high strength/modulus, moisture resistance, higher heat deflection temperature, etc. Therefore, nanotechnology can be used to develop novel FR thermoplastic structural components with improved, high performance characteristics. The major technical objective of this study is to investigate the feasibility of enhancing the fire retardancy and flammability properties of nylon (polyamide) thermoplastic material by combining it with selected nanoparticles. It involved the incorporation of nanoparticles into nylon (polyamide) 11 (PA11) to form a nylon (polyamide) 11 nanocomposite (PA11N). These PA11N materials are expected to exhibit reduced flammability and improved thermal properties as compared to the baseline PA11 material. The effect of nanoparticles on the resin will also be examined. Four types of nanoparticles such as clays [16,17], CNFs [16,17], MWNTs [18], and nano-alumina (nalumina or NA) [19] in different wt% levels were compounded separately, and selective formulations were combined with a conventional intumescent FR additive. In this study, four families of PA11/clay, PA11/CNF, PA11/MWNT, PA11/NA polymer blends were melt blended using twin-screw extrusion to form PA11N. The resulting pellets were injection molded into different test specimens. Transmission electron microscopy (TEM) was used to examine the extent of dispersion of nanoparticles. Thermal stability properties of the samples were analyzed using thermogravimetric analysis (TGA). Flammability properties were studied by conducting UL 94, cone calorimetry, and heat distortion temperature measurements. II. Description of Materials A. Polymer System Polyamides (PAs) are versatile thermoplastic engineering polymeric materials noted for outstanding properties such as high tensile strength, good resistance to flow under pressure (creep), excellent abrasion, chemical and heat resistance, and a low coefficient of friction. Polyamides such as PA6 and PA66 are high melting, moderately crystalline polymers (T m C) while PA11 and PA12 are intermediate melting materials (<200 C) and are less susceptible to moisture as compared to PA6 and PA66. Fornes and Paul [20] have examined the structure and properties of nanocomposites based on PA6, PA11, and PA12. Their studies were focused solely on nanoclays whereas our studies are directed to examining/comparing nanoclay and other nanoparticles to determine enhanced polymer characteristics such as flame retardancy and improved thermal/mechanical properties for the resulting PA11N. Arkema s RILSAN polyamide 11 (PA11) was selected for this study since it is an attractive polyamide used in a variety of applications. RILSAN PA11 thermoplastic is a polymer developed by Atofina Chemicals, Inc. (now known as Arkema, Inc.) in Derived from a series of complex chemical operations, RILSAN PA11 is one of the few polymers that are produced from green raw materials castor beans. RILSAN PA11 resin has earned a preferred material status in the most demanding applications due largely to their unique combination of thermal, physical, chemical, and mechanical properties. This results in an outstanding "cost performance ratio." Processing ease is another major benefit of RILSAN polyamide 11 resins. Supplied in powder or pellet form RILSAN PA11 resin can be processed by injection molding, extrusion, blown film extrusion, extrusion blow molding or rotomolding. The ease of processing of RILSAN has led designers to select them for industries as diverse as aerospace, offshore drilling, electrical cables, automotive, and pneumatic and hydraulic hose. B. Nanoparticles Four types of nanoparticles were used in this study. They are Southern Clay Products Cloisite 30B nanoclay, Arkema s Graphistrength C100 multiwall carbon nanotubes, Applied Sciences Pyrograf III PR-19-PS CNF, and Sasol s Dispal X-0 Needle, X-0 SR, and X-25 SR n-alumina materials. They are briefly described in the following. 1. Montmorillonite Nanoclays (NCs) Montmorillonite (MMT) nanoclay is a widely investigated nanoparticle, most commonly formed by in-situ

3 alteration of volcanic ash. Its sheet structure consists of layers containing the tetrahedral silicate layer and the octahedral alumina layer. Each sheet of MMT nanoclay has thickness of approximately 0.96 nm. MMT nanoclays are able to improve mechanical properties, barrier performance, and application processing when full exfoliation is achieved in a polymeric material. In order to achieve exfoliation in various continuous phases, surface treatment of the nanoclay is needed. The MMT nanoclay of choice in this project is the Cloisite 30B [2,21], which is a surfacetreated montmorillonite manufactured by Southern Clay Products. Cloisite additives allow complete dispersibility and miscibility with many different resin systems. Surface treatment is accomplished with ion exchange between inorganic alkali cations on the clay surface with the desired organic cation. In this case, quaternary ammonium ions are chosen, and they act as a compatibilizing agent between the interface of MMT nanoclay and the continuous phase of the polymer. Specific chemical pre-treatment modifier (MT2EtOH: methyl, tallow, bis-2-hydroxyethyl, quaternary ammonium) is used for the preparation of Cloisite 30B [2,21]. 2. Multiwall Carbon Nanotubes (MWNTs) Arkema produces multiwall carbon nanotubes (CNTs) under the trade name Graphistrength. Process is chemical vapor deposition at elevated temperature of ethylene on metal/ceramic catalyst. Arkema produces CNT s (Graphistrength C100) and CNT-containing products. Plain CNTs exist in bundles with median diameter approximately 10 to 20 µm. Typical dimensions are 10 to 15 nm in diameter corresponding to approximately 5 to 15 concentric tubes, with lengths approximately 1 to 10 µm. Graphistrength C100 is used in this study [2,22]. 3. Carbon Nanofibers (CNFs) Carbon nanofibers (CNFs) are a form of vapor-grown carbon fiber, which is a discontinuous graphitic filament produced in the gas phase from the pyrolysis of hydrocarbons. CNFs are able to combine many of the advantages of these other forms of carbon for reinforcement in engineered polymers. CNFs also have transport and mechanical properties that approach the theoretical values of single crystal graphite but can be made in high volumes at low cost. The CNFs are manufactured by Applied Sciences Inc./Pyrograf Products by pyrolytic decomposition of methane in the presence of iron-based catalyst particles at temperatures above 900 C. CNFs are produced only in a discontinuous form, where the length of the fiber can be varied from about 100μm to several cm, and the diameter is of the order of 100nm. The CNFs used in the study is PR-19-PS CNF. PR-19-PS CNF has fiber diameters of 100 to 200nm and fiber lengths of 30 to 100μm [2,23]. 4. Nano-alumina (NA) They are synthetic alumina nanopartilces which are high purity, highly dispersible, colloidal boehmite alumina powders manufactured by Sasol. These alumina powders come in platelet or needle shape and can be obtained in different particle size distributions. They are available in micron size but will become nano-sized in the dispersed phase. Three types of NA with different organic surface treatment and shape were used in this study. They were Sasol s Dispal X-0 Needle, X-0 SR, and X-25 SR [2,24]. 5. Fire Retardant Additive Exolit OP 1230 is an intumescent FR additive manufactured by Clariant and is based on metal phosphinate. It is a white powder with high phosphorus content (~24 wt%) is non-toxic, hydrophobic and exhibits good thermal stability [25]. PA66 products filled with ~24 wt% Exolit OP 1230 showed enhanced FR properties [26]. III. Experimental Approach A. Measurements 1. Morphological Microstructures Analysis Small portions of the injection molded specimens of the PA11 nanocomposites were investigated by TEM to examine the uniformity of nanoparticles dispersion within the PA11 polymer matrix. Uniform distribution of the nanoparticles within the polymer matrix is essential to yield the best enhancement of material properties of the polymer matrix. 2. Thermal Stability Testing Thermal stability of the PA11 baseline and PA11N were examined by thermogravimetric analysis (TGA) using the Perkin Elmer TGA 7. Weight changes in sample materials are measured as a function of temperature or time in TGA.

4 The sample is heated by a furnace with nitrogen while the loss or gain of sample weight is monitored by a sensitive balance. Weight, temperature, and furnace calibrations were carried out within the range of the TGA ( C) at scan rates of 10 C/min and 20 C/min. 3. Flammability Testing UL 94 It is a standard test for flammability of plastic materials in industry that serves as a preliminary indication of plastics acceptability for use as a component of a device or appliance with respect to its flammability behavior. UL 94 is not intended to reflect the hazards of a material under actual fire conditions but is considered as a preliminary step toward obtaining plastic recognition and subsequent listing in the Plastics Recognized Component Directory (formerly known as Yellow Cards ). The materials are tested in a vertical setting to determine the UL 94 V-0 rating. UL 94 Vertical burning test involves a 1/2 x 5 specimen which is held at one end in the vertical position. A burner flame is applied to the free end of the specimen for two 10 second intervals separated by the time it takes for flaming combustion to cease after the first application. Five specimens are tested for each formulation. Materials were tested as received with no conditioning/drying. The test was performed in our lab and the UL 94 testing requirements and procedures were followed as strictly as possible. The lab is not certified for UL 94 testing and the results serve only for screening purposes. The testing was performed in a fume hood with a preset airflow of ft/min. Fume hood sash was lowered as much as possible to prevent airflow from the outside environment. The erratic exhaust airflow from the fume hood and the surrounding made the reproducibility of the testing challenging. The burner is lit during the time of this experiment to keep the applied flame constant between each specimen. Cone calorimetry It is a standard test method for heat and visible smoke release rates for materials and products using an oxygen consumption calorimeter. Its premise is that, in most materials, the net heat of combustion is directly proportional to oxygen released per kilogram of oxygen consumed. Induced by a spark ignition source, test specimens burn in ambient air conditions subject to a pre-determined external heat flux (typically kw/m 2 ; 35 or 50 kw/m 2 in this study). Test results are expressed in terms of time to ignition; peak, average and total heat release rates; mass loss and mass loss rate; effective heat of combustion; visible smoke development; and release rates of carbon monoxide and carbon dioxide. In this study, samples were examined with no additional conditioning/drying prior to testing. Measurements were made on a FTT Dual Cone Calorimeter with an exhaust flow of 24L/s using the standardized cone calorimeter procedure (ASTM E d). All samples were tested without frame and grid with each sample s backside wrapped in aluminum foil. Data collected from all samples have an error of ±10% based on a specimen surface area of 100 cm 2. Three repetitions were conducted for each blend and specific attention paid to the heat release rate (HRR) and peak heat release rate (PHRR). Heat deflection temperature The temperature at which a standard test bar deflects a specified distance under a load is known as its heat deflection (or distortion) temperature (HDT). In structural applications, it is used to describe resistance to short-term heating; it distinguishes between materials able to sustain light loads at high temperatures and those that lose their rigidity over a narrow temperature range at elevated temperature. In this study, HDT was measured per ASTM D648 Standard Test Method for Deflection Temperature of Plastics under Flexural Load in the Edgewise Position using Atlas HDV2 DTUL/Vicat tester. For this test procedure, specimens are then lowered into a silicone oil bath where the temperature is raised at 2 C/minute until they deflect 0.25 mm. The temperature at the specified load and deflection is recorded. Three repetitions were conducted for each blend. IV. Discussion of Results A. Processing and Microstructural Characterization of the Polymer Nanocomposites A 30 mm Werner Pfleider corotating twin screw extruder which is configured for a wide variety of materials was used. The extruder L/D can be varied from 21 to 48, with options of multiple feeds and vents. The energy profile of the screw is adjusted to optimally meet the needs of the target product. Approximately 10 lbs of each PA11 nanocomposite formulations were produced. Separate volumetric feeders were used for the base resin and the nanoparticles. The PA11 was dried in a desiccant drier before compounding. Injection molded specimens of each polymer blend were prepared and examined by TEM. All the PA11 polymer with clays (Cloisite 30B in 2.5, 5, 7.5, and 10 wt%) [16], with CNFs (PR-19-PS in 2.5, 5, 7.5, and 10 wt%) [16], with MWNTs (Graphistrength C100 MWNTs in 1, 3, 5, and 7 wt%) [18], and with the three NA (in 2.5, 5, 7.5 wt%) [19] were compounded and tested. 1. Nanoclays and Carbon Nanofibers

5 Figure 1 shows TEM micrographs of PA11/nanoclay blend with 10% Cloisite 30B in 90% PA11. Good dispersion of these nanoclays in the polyamide 11 matrix is clearly shown in Figure 1. Figure 2 shows TEM micrographs of PA11/CNF blend with 10% CNF in 90% PA11 as a uniform dispersion of these CNFs in the polyamide 11 matrix. Figure 1. TEM micrographs of polyamide 11 with 10% Cloisite 30B nanoclay illustrating exfoliation and good dispersion. Figure 2. TEM micrographs of polyamide 11 with 5% CNF illustrating good dispersion. 2. Multiwall Carbon Nanotubes Figure 3 shows TEM micrographs of PA11/MWNT blend with 7% Graphistrength C100 in 93% PA11. Good dispersion of these MWNTs in the PA11 matrix is clearly observed in Figure 3. Figure 3. TEM micrographs of polyamide 11 with 7% MWNT illustrating good dispersion.

6 3. Nano-Alumina Figures 4 to 6 show TEM micrographs of PA11 with 5% X-0 Needle, 5% X0-SR, and 5% X25-SR, respectively. Good dispersion of these nanoparticles in the PA11 matrix is clearly shown in Figures 4 to 6. Figure 4. TEM micrographs of polyamide 11 with 5% X-0 Needle NA illustrating good dispersion. Figure 5. TEM micrographs of polyamide 11 with 5% X0-SR NA illustrating good dispersion. Figure 6. TEM micrographs of polyamide 11 with 5% X25-SR NA illustrating good dispersion. B. Thermal Stability of the Polymer Nanocomposites 1. Nanoclays and Carbon Nanofibers Thermogravimetric analysis (TGA) was performed on neat and FR/nanoparticle-reinforced PA11 under nitrogen using scan rates of 10 C/min and 20 C/min. The test could not be performed under air since the intumescent FR

7 additives would expand at high temperature under an air environment resulting in erratic results. Since lower and higher scan rates showed similar trend of curve shifts with the lower scan rate exhibiting a better separation of curves, only the TGA data obtained using the scan rate of 10 C/min was shown (Figures 7 and 8). From these data, the decomposition temperatures for both 10% and 50% mass loss, T 10% and T 50%, are summarized in Table Sample Mass (%) 60 PA11 neat 50 PA11 / 30%FR 40 PA11 / 2.5%NC2 PA11 / 5%NC2 30 PA11 / 7.5%NC2 20 PA11 / 20%FR / 7.5%NC2 PA11 / 20%FR / 5%NC2 10 PA11 / 25%FR / 5%NC Temperature ( ) Figure 7. Sample mass (%) of neat PA11 and PA11/NC2 nanocomposites with and without FR additive (TGA, scan rate 10 ºC/min) Sample Mass (%) 70 PA11 neat 60 PA11/30%FR 50 PA11/3%CNF 40 PA11/5%CNF 30 PA11/7%CNF 20 PA11/20%FR/5%CNF PA11/20%FR/7.5%CNF 10 PA11/25%FR/5%CNF Temperature ( ) Figure 8. Sample mass (%) of neat PA11 and PA11/CNF nanocomposites with and without FR additive (TGA, scan rate 10 ºC/min).

8 Table 1. Decomposition Temperature of Nanoparticle-Filled PA11 and PA11 with FR 1 Polymer FR Nanoparticle Decomp. Temp. at 10% mass loss Type Wt% Wt% Type Wt% Decomp. Temp. at 50% mass loss T ( C) ( C) T ( C) ( C) NC PA CNF NC PA CNF NC2 = Cloisite 30B CNF = PR-19-PS CNF FR = Clariant OP1230 FR 1 thermogravimetric analysis performed using heating rate of 10 C/min in nitrogen The decomposition temperatures of 70% PA11/25% FR/5% NC2 are higher than those without FR (Table 1). These indicate that the presence of FR materials can enhance the polymer s thermal stability. The data also show that NC2- only blends had the same or higher thermal stability than the FR-only blend, while FR/NC2 blends showed a much higher stability. These data confirm a plausible synergism between the FR and NC2 in the PA11 matrix. The sample with 20% FR/7.5% CNF had the highest decomposition temperatures while the PA11 baseline sample had the lowest decomposition temperatures (Table 1 and Figure 8). The data reveal that the increased loading of either CNF or FR enhances the material s thermal stability since the sample with 20% FR/5% CNF had lower decomposition temperatures than the other two samples which contained increased amounts of either CNF or FR additive. The CNF addition increased PA11 s thermal stability slightly, while the combination/synergy of FR and CNF yielded better FR enhancement, similar to NC2 formulations. Finally, for the formulations with both nanoparticles and FR additive, the compositions with highest decomposition temperatures in TGA had the lowest PHRR in cone calorimetry [16]. Thus combining TGA with UL 94 (as will be shown later) is potentially an effective screening tool for selecting candidates for cone calorimetry analysis. 2. Multiwall Carbon Nanotubes TGA was performed on all six formulations (neat and 1 to 20 wt% MWNTs) under nitrogen using scan rate of 10 C/min (Figure 9) [18]. A general trend of increased thermal stability is observed (higher decomposition temperatures) as the amount of MWNT is increased. Table 2 summarizes the 10% mass loss, onset, and 50% mass loss decomposition temperatures of the PA11/MWNT nanocomposites. The 10% mass loss, onset, and 50% mass loss decomposition temperatures of PA11/7%MWNT nanocomposite increase significantly from 418 o to 440 o C, 425 o to 457 o C, and 439 o to 475 o C, respectively.

9 Mass% 100% 90% 80% 70% 60% 50% 40% BMNO PA11 neat PA11 + 1%MWNT PA11 + 3%MWNT PA11 + 5%MWNT PA11 + 7%MWNT PA %MWNT 30% 20% 10% 0% Temperature (C) Figure 9. Thermogravimetric analysis of PA11/MWNT nanocomposites at 10 C/min in nitrogen. Table 2. Summary of Thermal Stability of PA11/MWNT Nanocomposites Samples Decomposition Temperatures ( C) Polymer Blend at 10% mass loss Onset at 50% mass loss PA11 neat PA11+1% MWNT PA11+3% MWNT PA11+5% MWNT PA11+7% MWNT PA11+20% MWNT Nano-alumina Three types of NA with different organic surface treatments, namely Dispal X-0 Needle, X-25SR, and X-0SR, were melt blended, separately, with RILSAN PA11 using twin-screw extrusion to form PA11N in three different weight loadings (2.5%, 5%, and 7.5%). A total of ten formulations (including the neat PA11) were produced. The resulting pellets were injection molded into different test specimens. Thermal properties of the samples were analyzed using thermogravimetric analysis (TGA). Flammability properties were studied by conducting UL 94. TGA was performed on the nine polymer blends and one control polymer under nitrogen using a scan rate of 10 C/min [19]. Table 3 shows the decomposition temperatures of the PA11/n-alumina nanocomposites as well as the neat nylon 11. All PA11/NA nanocomposites had substantial increases in decomposition temperatures (from 26 o to 46 o C) compared to the neat PA11. There was not a clear trend of thermal properties enhancement among different types of NA or different weight loading. However, the lowest weight loading of each type of NA appeared to provide the best improvement in decomposition temperature. Figures 10 to 12 show the TGA plots of polyamide 11 and the nanocomposities with 2.5%, 5%, and 7.5%, respectively, of each type of the NA.

10 Table 3. Decomposition Temperatures of PA11/NA Nanocomposites Samples Decomposition Temperature ( C) ΔT ( C) PA11 neat PA11/2.5% X PA11/5.0% X PA11/7.5% X PA11/2.5% X0-SR PA11/5.0% X0-SR PA11/7.5% X0-SR PA11/2.5% X25-SR PA11/5.0% X25-SR PA11/7.5% X25-SR PA11 neat 2.5% X-0 Needle 2.5% X-0 SR 2.5% X-25 SR 60 Weight % Temperature (C) Figure 10. TGA of polyamide 11 with 2.5% NA at 10 C/min in nitrogen PA11 neat 5% X-0 Needle 5% X-0 SR 5% X-25 SR 60 weight % Temperature (C) Figure 11. TGA of polyamide 11 with 5% NA at 10 C/min in nitrogen.

11 PA11 neat 7.5% X-0 Needle 7.5% X-0 SR 7.5% X-25 SR Weight % Temperature (C) Figure 12. TGA of polyamide 11 with 7.5% NA at 10 C/min in nitrogen. C. Flammability Properties of the Polymer Nanocomposites 1. Nanoclays and Carbon Nanofibers In order to improve the flame retardancy of polyamide 11 even further, a conventional intumescent FR was added to the nanoparticle-filled PA11 [16]. Again, due to the PA11/nanoparticle blends overall superiority with respect to flammability and mechanical properties [17], PA11/FR/nanoparticle blends were chosen for further cone calorimetry study (Table 4). To better simulate consumer-product plastics burning in a normal household fire, an external heat flux of 35kW/m 2, equivalent to a trash-can-fire heat flux rather than 50kW/m 2 was used for the cone calorimetry testing. UL 94 data are also shown in Table 4 but will be discussed later. Table 4. Flammability Properties of Nanoparticle-Filled PA11 with Intumescent FR PA11 FR Nanoparticle Peak Heat Release Rate 1 Tests for Flammability of Plastics Materials 2 Wt% Wt% Type Wt% kw/m 2 Δ(%) Averaged 1st burn flaming combustion duration (s) Averaged 2nd burn flaming combustion duration (s) Flaming drip UL 94 Rating N/A Yes failed No V NC No V No V No V CNF No V No V No V-0 1 cone calorimetry test (35 kw/m 2 ) was conducted in accordance with ASTM E test for flammability was conducted in accordance with Underwriter Laboratory Standard, UL 94 NC2 = Cloisite 30B CNF = PR-19-PS CNF FR = Clariant OP1230 FR

12 Two general trends are noteworthy: first, the PHRR of each FR/nanoparticle blend was lower than that of FR alone in spite of containing a lower FR wt%; and second, keeping FR wt% constant, PHRR decreased with increasing nanoparticle wt%. These results are consistent with the fact that higher nanoparticle wt% generally yields better FR properties. Moreover, these results demonstrate an anomaly of nanoparticle behavior. Usually when FR wt% increases with constant nanoparticle wt%, PHRR decreases. This is true for the PA11/FR/NC2 blends, but the opposite is true for the PA11/FR/CNF blends. While this anomaly may be due to some physical effects during the test, it is not currently understood and warrants further investigation. Representative HRRs during burning for neat and FR/nanoparticle-filled PA11 are plotted in Figure 13. The time-to- PHRR of all PA11/FR/nanoparticle blends was longer than that of neat PA11. However, the time-to-phrr of the FR/NC2 formulations was lower than for FR-only blend and the FR/CNF formulations exhibited the longest timeto-phrr. This may suggest that CNF has a better synergistic effect with the FR additive than NC2. Nevertheless, all PA11/FR/nanoparticle blends had lower PHRR than the FR-only blend and burned longer. HRR (kw/m 2 ) PA11 neat PA11 / 30%FR PA11 / 20%FR / 5%NC2 PA11 / 20%FR / 7.5%NC2 PA11 / 25%FR / 5%NC2 PA11 / 20%FR / 5%CNF PA11 / 20%FR / 7.5%CNF PA11 / 25%FR / 5%CNF Time (s) Figure 13. Heat release rate of neat and nanoparticle-filled PA11 with and without FR additive as a function of time (cone calorimetry, 35 kw/m 2 ). Table 4 summarizes the UL 94 test results. Only two neat PA11 specimens were examined because its flammability classification was anticipated. Both specimens burned for more than 10 seconds after first flame application. Moreover, drips from the specimens ignited the cotton below the flame. Therefore, it was safe to conclude that neat PA11 did not pass the UL 94 V-0 test. The addition of FR and/or the nanoparticles greatly reduced the polymer combustion after the first burner flame application and also reduced dripping. The PA11/FR/CNF blends and 70% PA11/30% FR had the shortest combustion times. The 70% PA11/30% FR blend performed best. Of three NC2 blends, only one (72.5% PA11/20% FR/7.5% NC2) passed the requirement of V-0 rating, while all three of the PA11/FR/CNF blends passed the V-0 rating. The UL 94 results clearly indicate the enhancement of PA11 with FR and NC2 fillers, albeit the results varied with the actual composition of the PA11/FR/NC2 blend. In terms of total flaming time for five specimens of each formulation: the 72.5% PA11/20% FR/7.5% NC2 passed the V-0 requirement; the 70% PA11/25% FR/5% NC2 failed by a small margin; and the 75% PA11/20% FR/5% NC2 clearly failed. By comparing the results for 72.5% PA11/20% FR/7.5% NC2 and 70% PA11/25% FR/5% NC2, it is clear that NC2 out-performed the FR in improving fire retardancy: the slight NC2 increase (2.5 wt%) and corresponding FR reduction (5 wt%) led to better results. Another formulation, specifically 70% PA11/20% FR/10% NC2, should be tested. This similar trend was observed with the CNF blends: higher CNF wt% improved fire retardancy, albeit the difference was not as distinct as in NC2 blends. Both 72.5% PA11/20% FR/7.5% nanoparticle blends passed the V-0 test requirements. The CNF blend had no combustion at the first flame application and a lower total combustion time than the NC2 blend. Therefore, the FR/CNF blend out-performed its NC2 counterpart.

13 Reassuringly, the UL 94 test results were consistent with the cone calorimetry data: those blends that passed the V-0 requirement had the lowest PHRR with one exception, 70% PA11/25% FR/5% NC2. It had a slightly lower PHRR than the 72.5% PA11/20% FR/7.5% NC2 but passed with only a V-1 rating. However, as noted earlier, it only failed the V-0 rating by a small margin. Three of the five repetitions passed the V-0 rating and the remaining two burned for less than a few seconds longer than the requirement. These variations may reflect limits on the specimen-tospecimen reproducibility due to the apparatus or operator. Six formulations (neat PA11, PA11/20% FR/5% NC2, PA11/20% FR/7.5% NC2, PA11/20% FR/5% CNF, PA11/25% FR/5% CNF and PA11/20% FR/7.5% CNF) were chosen for heat deflection temperature (HDT) measurement [16]. Those NC2 specimens had the same weight loading of FR additive (20 wt%) but with different amounts of NC2. Figure 14 illustrates that PA11 with additive exhibited substantial increase in HDT compared to neat PA11; moreover, higher NC2 wt% resulted in higher HDT and agreed with the prediction based on the TGA results. More experiments are needed to analyze the effect of the FR additives and to compare these results to those of PA11/NC2-only composites since their HDTs were not measured due to economical constraints. Figure 14. Heat deflection temperature of neat and nanoparticle/fr-filled PA11 as a function of nanoparticle wt%. All three FR/CNF formulations were chosen for HDT testing because they exhibited the lowest PHRR. Consistent with the flammability data, all PA11/FR/CNF blends had higher HDT values than the baseline PA11, as well as the PA11/FR/NC2 formulations. Similar to the previous PA11/FR/NC2 blends, higher CNF wt% yielded higher HDT, as the FR wt% remained constant. Higher FR wt% yielded better HDT as the CNF wt% remained the same. B. Multiwall Carbon Nanotubes Figure 15 shows photos of PA11/MWNT specimens before and after the test while Figure 16 shows a burning specimen during the test [18]. All formulation specimens with MWNTs burned longer than 100 seconds after the first flame was applied and continued to burn. Therefore, none of the formulations passed any rating of the UL 94 test. However, it should be noted that the PA11/MWNT nanocomposites did not exhibit any dripping when they were burned, while the neat PA11 dripped copiously during the test. None of these MWNT formulations passed the UL 94 V-0 test.

14 Figure 15. UL 94 specimens (7wt% MWNT) before (left) and after (right) the burning test. Figure 16. Flame burned all the way to the clamp in the UL 94 test. C. Nano-alumina As shown in Table 5, the addition of nano-alumina (NA) to the PA11 material increased the FR properties of the nylon 11 by reducing dripping and flaming combustion time [19]. Though all formulations of the PA11N failed the V-0 and V-1 ratings, the X-0 Needle formulations performed the best in terms of average burn time and resistance to dripping. The X-0 SR performed the worst, with several samples completely burned, and the X-25 SR was the middle performer.

15 Table 5. UL 94 Results of PA11/NA Nanocomposites Samples 1 st Burn Drip 2 nd Burn Drip Drip Rank (1 = best) UL 94 Rating PA11 Yes Yes 3 Failed PA11/2.5% X-0 Yes Yes 1 V-2 PA11/5.0% X-0 Yes Yes 1 V-2 PA11/7.5% X-0 Yes Yes 1 V-2 PA11/2.5% X-0 SR Yes Yes 3 Failed PA11/5.0% X-0 SR Yes Yes 3 Failed PA11/7.5% X-0 SR Yes Yes 3 Failed PA11/2.5% X-25 SR Yes Yes 2 V-2 PA11/5.0% X-25 SR Yes Yes 2 Failed PA11/7.5% X-25 SR Yes Yes 2 Failed Note: To obtain a UL 94 V-0 rating, one of the criteria is that the afterflame time for each individual specimen t 1 (afterflame time after first flame application) or t 2 (afterflame time after second flame application) has to be 10s. For V-1 rating the afterflame time for each individual specimen t 1 or t 2 has to be 30s. X-0 Needle Overall these samples exhibited the highest resistance to dripping compared to the other PA11N formulations. On average, they also had the lowest total flaming combustion time. The 5% weight loading achieved the lowest total flaming combustion time. All three weight loadings achieved a V-2 rating. X-25 SR These samples exhibited less resistance to dripping than the X-0 Needle, but performed comparably in terms of total flaming combustion time. Increasing the weight loading resulted in higher total flaming combustion times. This is because the drip resistance of the PA11N was increased, but allowed for flame propagation up the material. For the lower weight loadings, the flaming portion of the sample was allowed to drip and fall, not allowing enough time for the flame to propagate. X-0 SR The PA11 samples with X-0 SR exhibited high total combustion times and several of the samples burned completely. Increasing the weight loading increased the total combustion times, and several samples did not withstand the first 10 second flame application. V. Summary and Conclusions Thirty polymer blends of different combinations of PA11, MMT nanoclay, CNF, and intumescent FR additives were compounded separately via twin screw extrusion. Injection molded specimens were fabricated for thermal and flammability properties measurements. This study demonstrated that good dispersion of the NCs, CNFs, and FR additives in the PA11 polymer was achieved. Thermal stability of the PA11 was enhanced by NC2 while the CNFs had little or no effect. All FR/nanoparticlereinforced PA11 blends had higher decomposition temperatures than those reinforced with FR additive only. The PHRR of PA11 blends under cone calorimetry decreased as the nanoparticle wt% increased. The PHRR of the PA11/FR/NC2 or PA11/FR/CNF decreased as the NC2 or CNF wt% increased; the PHRR of these polymers was lower than that of PA11/FR composite. For the PA11/FR/NC2 blends, only the 20% FR/7.5% NC2 sample passed the UL 94 V-0 test; for the PA11/FR/CNF blends, all three formulations passed UL 94 V-0 test. The combined use of TGA and UL 94 testing can facilitate the selection of candidates for cone calorimetry analysis. The HDT increased substantially by the addition of nanoparticles and FR additive. More data are needed to investigate the effect of the amount of intumescent FR on the HDT of the polymer systems. Synergism was established in terms of thermal and FR properties that occur between the nanoparticles (NC2 and CNF) and the intumescent FR additive. A total of five formulations of PA11/MWNT nanocomposites were compounded separately via twin screw extrusion and injection molded to specimens. TEM showed that MWNTs were well dispersed into the PA11 matrix although it is possible that better dispersion can be achievable by using a more aggressive screw design. MWNTs enhanced the thermal stability of the PA11. However, all PA11/MWNT nanocomposites failed the UL 94 test. To make them

16 truly fire retardant, a second additive such as conventional intumescent fire retardant additives to act as a synergist are proposed. It has been demonstrated that thermal and flammability can be enhanced substantially as long as synergism is established between the nanoparticles and conventional fire retardant additive in the polyamide 11 matrix. A total of ten formulations of PA11/NA nanocomposites, including the neat PA11, were melt-compounded and injection molded to specimens. TEM demonstrated that NA surface treated particles are well dispersed in the PA11 matrix. TGA showed that NA nanoparticles improve the thermal stability of the PA11 substantially. The UL 94 test showed that all formulations did not pass the ideal V-0 rating. All of them had flammable dripping that ignited the cotton under the specimens during the test. The X-0 Needle formulations performed the best, followed by X-25 SR, and the X-0 SR performed the worst in performance. The addition of selective fire retardant additives or antidripping agents will be incorporated into the system together with the NA for enhanced flammability properties. Insight and physical knowledge on the roles of nanoclays, carbon nanofibers, multi-walled nanotubes, and n-alumina in polyamide 11 nanocomposites were obtained in this study. This information would provide useful guidance for the development of a new generation of polymer nanocomposites. Besides the obvious contrasting differences in the flammability properties of different nanoparticles, the incorporating of various fillers has both advantages and disadvantages from the viewpoint of flame retardancy. References 1. Pinnavaia, T.J., and Beall, G.W. (eds.), Polymer-Clay Nanocomposites, Wiley, New York, Koo, J.H., Polymer Nanocomposites: Processing, Characterization, and Applications, McGraw-Hill, New York, Morgan, A.B., and Wilkie, C.A. (eds.), Flame Retardant Polymer Nanocomposites, Wiley, New York, Nelson, G.N., and Wilke, C.A. (eds.), Fire and Polymers III. Materials and Solutions for Hazard Prevention, American Chemical Society, Washington, D.C., Giannelis, E., Adv. Mater. 1996, 8, Gilman, J.W., and Kashiwagi, T., SAMPE J., 1997, 33(4), Gilman, J.W., Appl. Clay. Sci., 1999, 15, Zhu, J. et al., Chem. Mater., 2001, 13, Zanetti, M., Camino, G., and Mulhaupt, R., Polym. Degrad. Stabil., 2001, 74, Gilman, J.W. et al., Chem. Mater., 2000, 12, Kashiwagi, T., Grulke, E., Hilding, J., Harris, R., Awad, W. and Douglas, J., Macromol. Rapid Commun., 2002, 23, Koo, J.H. et al., Proc SAMPE ISSE, SAMPE, Covina, CA, 2003, p Yang, F., Yngard, R.A., and Nelson, G.L., Nanocomposites 2002, San Diego, CA, Sept Kashiwagi, T. et al., Nanocomposites 2002, San Diego, CA, Sept Pool, A.D., and Hahn, H.T., Proc SAMPE ISSE, SAMPE, Covina, CA, Lao, S.C., Wu, C., Moon, T.J., Koo, J.H., Morgan, A., Pilato, L., and Wissler, G., Flame-retardant Polyamide 11 and 12 Nanocomposites: Thermal and Flammability Properties, J. of Composite Materials, 2009, 43 (17), Lao, S.C., Wu, C., Moon, T.J., Koo, J.H., Pilato, L., and Wissler, G., Flame-retardant Polyamide 11 and 12 Nanocomposites: Processing and Mechanical Properties, J. of Composite Materials, 2010, in press. 18. Lao, S.C., Koo, J.H., Yong, W., Lam, C., Zhou, J., Moon, T., Piccione, P.M., Wissler, G., Pilato, L., and Luo, Z.P., Polyamide 11-Carbon Nanotubes Nanocomposites: Preliminary Investigation, Proc Solid Freeform Fabrication Symposium, The University of Texas at Austin, Austin, TX, Aug Lao, S.C., Koo, J.H., Moon, T., Hadisuyoto, B., Yong, W., Wissler, G., and Pilato, L., Flammability and Thermal Properties of Polyamide 11-Nano-alumina Nanocomposites, Proc Solid Freeform Fabrication Symposium, The University of Texas at Austin, Austin, TX, Aug Fornes, T.D., and Paul, D. R., Macromolecules, 2004, 37(20), Cloisite 30B technical datasheet, Southern Clay Products, Gonzales, TX. 22. McAndrew, T.P, Laurent, P., Havel, M., and Roger, C., Arkema Graphistrength Multi-Walled Carbon Nanotubes, Report from Arkema Inc., Research Center, King of Prussia, PA. 23. PR-19-PS CNF technical datasheet, Applied Sciences Inc., Cedarville, OH. 24. Private communication, Chang, Y., Sasol, NA, Lake Charles, LA.

17 25. Exolit OP 1230 technical data sheet, Clariant, Germany. 26. Dietz, M., Wanzke, W., Flame Retarded Thermoplastic and Thermoset Materials Based on Phosphorus Chemistry, Proc SAMPE ISTC, SAMPE, Covina, CA, Acknowledgements The authors would like to acknowledge NSF and KAI, LLC for financial support. The authors thank the following individuals for supporting our research: Dr. P. Piccone of Arkema, Dr. Z.P. Luo of Texas A&M University, Dr. D. Hunter of Southern Clay Products, Dr. Y. Chang of Sasol North Americal, and Mr. Tim Reiley of Clariant.