- Email: [email protected]
Thermal stability of polypropylene-clay nanocomposites subjected to laser pulse heating
Polymer Degradation and Stability 98 (2013) 2497e2502
Contents lists available at ScienceDirect
Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab
Thermal stability of polypropylene-clay nanocomposites subjected to laser pulse heating Stephen F. Bartolucci a, *, Karen E. Supan b, Jeffrey S. Wiggins c, Lawrence LaBeaud c, Jeffrey M. Warrender a a b c
US Army Armaments Research Development and Engineering Center e Benét Laboratories, Watervliet, NY 12189, USA Norwich University, Department of Mechanical Engineering, Northfield, VT 05663, USA University of Southern Mississippi, School of Polymers and High Performance Materials, Hattiesburg, MS 39406, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 May 2013 Received in revised form 22 August 2013 Accepted 8 September 2013 Available online 21 September 2013
Polypropylene based nanocomposites filled with montmorillonite nanoclay prepared by twin screw extrusion have been studied for thermal stability at high heating rates. In contrast to traditional thermal stability and flammability studies of polymer nanocomposites using heating rates on the order of tens of degrees per minute, this study achieves heating rates that are six orders of magnitude higher. This was accomplished using laser pulse heating. The results show that the nanoclay increases thermal stability of the polymer, as measured by a decrease in the mass loss for a laser pulse at a given energy and intensity. Electron microscopy and various spectroscopic techniques show that a silicate-rich char layer may provide the mechanism for protection of the polymer and decreased degradation rates. The results of the study are compared to the typical results found in traditional thermal stability testing. Published by Elsevier Ltd.
Keywords: Nanoclay Nanocomposite Laser pulse heating Polypropylene Thermal stability Degradation
1. Introduction The addition of nanoscale material to polymer matrices for improved mechanical, thermal, and electrical properties has been studied extensively in recent years [1e3]. One of these nanoscale materials, montmorillonite (MMT), is a layered-aluminosilicate clay consisting of silicon dioxide tetrahedral layers with a metal oxide octahedral center layer [4]. The nanoclay has plate-like morphology with a thickness on the order of 1 nm and in-plane dimensions on the order of several hundred nanometers. Naturally hydrophilic, the nanoclay must be organically modified in order to be dispersed into a hydrophobic polymer. This is typically accomplished by substituting the sodium counter ions with, for example, an organic quaternary ammonium ion that contains alkyl chains. Van der Waals or hydrogen bonding is responsible for holding the clay platelets together, which allows the plates to easily expand or contract dependent upon the cation that is exchanged between the plates. The d-spacing of the clay platelets is easily measured by X-ray diffraction (XRD) and can give information on the intercalative state of the clay. Layered-silicate nanoclay can be dispersed * Corresponding author. Tel.: þ1 518 266 5189; fax: þ1 518 266 5161. E-mail address: [email protected] (S.F. Bartolucci). 0141-3910/$ e see front matter Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.polymdegradstab.2013.09.005
into polymers using melt-mixing techniques [5,6]. Well-dispersed and exfoliated clay can provide improved properties for the polymer, such as increased thermal stability. The thermal stability of a polymer is measured as the resistance to weight loss and chemical change, or degradation, when exposed to higher temperatures. The analytical technique conventionally used for characterizing thermal stability is thermogravimetric analysis (TGA), where the mass of the sample is monitored as the temperature increases. Thermal stability tests have traditionally been performed under relatively slow heating conditions, at heating rates on the order of 20 C/min or below, while some equipment is able to reach heating rates about one order of magnitude higher. Many studies have shown that the addition of nanoclay into polymer matrices generally increases the thermal stability of a polymer during TGA characterization [7e12]. Qin et al. demonstrated that a 5 wt% organically modified montmorillonite addition to polypropylene can increase the temperature at which the peak degradation rate occurs by about 50 C compared to the pure polymer when heated at 20 C/min in air [8]. In addition to thermal stability testing by TGA, several groups have conducted research on the flammability properties of polymer-nanoclay composites [13e 18]. In work by Gilman et al., polypropylene (PP)-layered silicate nanocomposites had a 75% lower peak heat release rate, HRR,
2498
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
compared to the pure polymer during cone calorimetry testing [13]. The enhanced thermal stability and flammability resistance of the nanoclay-polymer composites is attributed to the formation of a protective char layer in the nanocomposite, a multilayered carbonaceous-silicate structure that acts as an excellent mass transport barrier, slowing the flow of heat into the polymer and slowing the escape of the volatile decomposition products from the polymer [14]. While these testing methods provide information on the behavior of polymer nanocomposite during slow heating rate regimes, they are inadequate in describing how the material behaves during extremely high heating rates that may be of interest in applications such as ballistic environments [19]. The application of laser pulse heating (LPH) with a variable pulse duration millisecond laser opens a new regime of study on these materials at heating rates up to six orders of magnitude higher than those achievable with conventional methods. There is literature precedent for investigating the degradation of polymers (simulating rapid heating during severe fires) using laser heating [20–23]. Here, we attempt to extend this work to the case of polymer nanocomposites, using a millisecond pulsed laser. 2. Experimental 2.1. Preparation of the nanocomposites The clay nanocomposites were prepared by mixing in a twin screw extruder at 100 rpm. Zones 1 through 4 of the extruder were at 150 C, 190 C, 190 C, 190 C, respectively and the die was at 190 C. The extrudate was subsequently passed through an ice water bath before being pelletized. The pellets were stored under vacuum at elevated temperature (40 C) overnight to remove any residual moisture. Dried pellets were then fed into a single screw extruder (Zone 1e170 C, Zone 2e180 C and Zone 3e180 C) and extruded through a circular die for the sole purpose of forming a rod shape of a desired diameter unavailable to us with the twin screw dies. The rod was then machined into slices approximately 2 mm thick for laser pulse heating experiments. Pure isotactic polypropylene (Fortilene, Melt Flow Index of 2.8 g/10 min, at 230 C and 2.16 kg) was used to let-down the clay masterbatch. The clay masterbatch (Nanocor, NanoMax-PP) contained 50 wt% organically modified (quaternary ammonium) Montmorillonite (MMT) clay with compatibilizers, grafted maleic anhydride (PP-g-MA). The polypropylene used in the masterbatch has a Melt Flow Index of 20 g/10 min, according to the manufacturer. All material included 1 wt% carbon black (Cabot, Monarch 120) to give the materials uniform optical absorption during laser irradiation. Pure polypropylene (PP-0), PP with 5 wt% MMT (PP-5), 10 wt% MMT (PP-10), 25 wt% MMT (PP-25) and 50 wt% MMT (PP-50) samples were prepared for this study.
be approximately 8 0.2%, meaning most of the laser light was absorbed into the sample, as desired. 2.3. Characterization of nanocomposites The mass of the sample was weighed before and after irradiation on a microbalance with a readability of 0.001 mg (Sartorius ME-36 S). Ten samples for each condition were run for statistics. Scanning Electron Microscopy (SEM) was performed on a JEOL-840 electron microscope operating at 10 kV equipped with Energy Dispersive Spectroscopy (EDS) with a Thermo Electron 457A X-ray detector, and an FEI Helios Nanolab 600i Field Emission Electron Microscope (FE-SEM) operating at 5 kV. Samples were coated with a thin layer of gold in order to reduced charging effects. Fourier Transform Infrared Spectroscopy (FTIR) was performed on a PerkineElmer Spectrum One Spectrometer operating in Attenuated Total Reflectance mode (ATR) between 4000 cm1 to 515 cm1 at 4 cm1 resolution. X-ray Photoelectron Spectroscopy (XPS) was conducted on a PHI5701 LSci with a monochromated Al-Ka 1486.6 eV X-ray source (Evans Analytical Group). X-ray diffraction (XRD) was performed on a Bruker D8 Discover diffractometer with Cu-Ka radiation (l ¼ 1.505945 Å) from 5 to 25 2q with a 240 s accumulation time. The distance, between clay sheets, d, was calculated by Bragg’s Law, nl ¼ 2d sin q, using 2q peaks from the XRD spectra. Transmission Electron Microscopy (TEM) samples were prepared by carefully removing char material from the surface of a LPH region of the nanocomposite under a microscope using ultrafine tweezers. The material was deposited onto a carbon coated Cu TEM grid for viewing on an FEI Titan Environmental S/TEM operating at 300 kV. Profilometry of LPH samples was conducted on a Nanovea ST400 non-contact optical profilometer. The total ablated volume, as defined by the material loss below the initial sample, and the volume above the original surface, was determined by 3-D surface topography. 3. Results and discussion 3.1. Mass loss measurements Clay nanocomposites with varying clay content were subjected to conventional TGA and LPH. The TGA results confirmed the behavior expected from the literature, namely, increasing clay content leads to increased thermal stability for a fixed heating rate. The results of the TGA are similar to those published previously in the literature [7,8]. The mass loss for each clay nanocomposite during LPH is presented in Fig. 1. Error bars on the mass loss data
2.2. Laser pulse heating The Laser Pulse Heating (LPH) setup included an Nd:YAG laser (MegaWatt Lasers, Inc) that sends 1064 nm light into an optical fiber. The light from the fiber was passed through a 50 mm focal length collimating lens, followed by a 250 mm focal length objective lens, which produced a laser spot diameter of 3.0 mm (0.003 m). The incident laser energy was measured using a beam splitter (Newport) and an energy meter (Ophir Nova). The laser pulse energy and pulse duration can be adjusted, but all samples in this paper were irradiated at 3.1 J pulses lasting 5 ms, corresponding to a fluence of 44 J/cm2 and an intensity of 8.8 kW/cm2. The samples were irradiated in air to give thermo-oxidative conditions. The reflectance of the samples was measured in a UVevis spectrometer (PerkineElmer Lambda 40) at 1064 nm and found to
Fig. 1. Mass loss and volume loss of nanoclay composite material after laser pulse heating with a fluence of 44 J/cm2 (8.8 kW/cm2, 3.1 J, 5 ms) (average of 10 samples per data point).
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
correspond to the standard deviation of the 10 measurements for each composite material. As an increasing fraction of nanoclay is added to the composite, the mass loss decreases monotonically. In previous TGA and flammability studies, mass loss, and more specifically, mass loss rate, was seen to decrease with the addition of nanoclay to the polymer [13]. We present the volume loss obtained from two different methods. In the first, the volume loss is calculated from the mass loss and the weight fraction of clay in the nanocomposite, using 888 kg/m3 for the density of pure PP with 1% carbon black and 1345 kg/m3 for the density of the organically modified clay. For the second (open triangles), we directly measured the volume of the “crater” of removed material in the irradiated area using profilometry. These two methods give good agreement for samples with low clay content, but disagree increasingly with high clay content. At high clay content, the crater volume is negligible, and the profilometry measurements show features that extend above the original surface of the material (not included in calculated volume). We attribute this behavior to the formation of a char layer, and to the presence of trapped gaseous species underneath this layer. We adduce further support for this interpretation in Section 3.2.
3.2. Analysis of char formation 3.2.1. Electron microscopy and EDS Polypropylene is a non-charring thermoplastic polymer. When the polymer is pyrolyzed, it will completely degrade. When nanoclay is mixed into the polymer, it has been reported [24] that the clay acts as a promoter of char formation within the polymer when the polymer is exposed to degradation temperatures. The clay can slow down the degradation of the polymer and alter the degradation mechanisms of the polymer molecules and, hence, the kinetics of degradation [24,25]. While the clay itself can slow oxygen diffusion into the composite [26], the char layer further acts as a protective barrier for the remaining polymer as it slows heat and oxygen from entering the polymer and slows the escape of volatile decomposition products from the remaining polymer [27]. Electron microscopy in Fig. 2 reveals a different surface morphology for the various nanocomposites after LPH. The pure PP-0 material in (a) has a smooth, flat surface indicative of thermal ablation and volatile formation [28]. In contrast, the remaining materials show increased surface roughness with increasing clay content (b)e(e). The insets
2499
Fig. 3. Cross sectional SEM micrograph of PP-0 after LPH. A smooth ablation crater with no char layer is seen. The sample was scored and fractured through the LPH region at room temperature.
show a higher magnification top view of the LPH region of each sample. The lower-clay content nanocomposites show areas that are qualitatively similar to those of the PP-0, but also show areas with larger surface elevation differences, and some areas with different morphology. The higher-clay content nanocomposites have a dramatically different appearance compared to the PP0 sample. In order to study this further, the composites were fractured at room temperature through the LPH region and the cross sections were viewed in the SEM, with EDS mapping. Examples of the PP0 and PP-50 composites are shown in Figs. 3 and 4. In PP-50, a silicate-rich char layer is seen on the surface of the ablation crater. Underneath this layer appears to be a thin layer of polymer that has been exposed to high enough temperatures to cause melting and bubble formation from volatiles. Polymer exists below this melting and bubble region that does not appear to have been affected by high temperatures. The melting and bubble formation can cause ruptures in the char layer forcing the silicate-rich material to ablate or form structures perpendicular to the composite surface. An example of this phenomenon is seen in Fig. 5, in which we show a bubble that appears to have ruptured. EDS mapping shows the surface layer to be rich in Si, O, and Al, while being deficient in C, Na,
Fig. 2. SEM of nanoclay composites after Laser Pulse Heating with a fluence of 44 J/cm2 (8.8 kW/cm2, 3.1 J, 5 ms). a) 0%, b) 5%, c) 10%, d) 25% and e) 50% nanoclay.
2500
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
Fig. 4. Cross sectional SEM micrograph of PP-50 after LPH. EDS spectral map shows that the char layer is rich is Si, O and Al, and deficient in C.
heated by TGA at 10 C/min (a) with another that was heated by LPH (b). The micrographs of the two surfaces, taken at the same magnification, are qualitatively very different. In the slowly heated composite, the surface char is relatively intact with small surface holes scattered throughout. Trapped volatiles underneath the surface will slowly cause increased pressures, resulting in the eventual formation of holes in the surface as the volatiles breakthrough and pressure is relieved. In contrast, the LPH surface shows a disrupted surface char, presumably a result of the more drastic heating rate and ablation. Rapid volatile formation causes a more violent release of pressure resulting in the more disrupted composite char surfaces. EDS of the surfaces show a higher relative Si, O, Al, and Fe content and lower C content in this region of the LPH surface, as compared to the TGA heated surface.
Fig. 5. Silicate-rich feature that has been ruptured due to bubble expansion in the underlying material (PP-50 material after LPH).
and Ca (not shown in EDS map). The C deficiency is due to polymer degradation and volatilization. An interesting observation is that the exchangeable Na and Ca cations present in the clay appear to have decreased signal in the LPH region, unlike the Si and Al, which increase in signal. These cations may have been removed from between the aluminosilicate layers along with the organic modifiers that some of the cations were substituted with. Unlike the composite that contains the nanoclay, the PP-0 has a smooth ablation crater and no char layer formed. The polymer exposed to the heat has volatized, leaving more pure polymer exposed. This is comparable to the slow heating rate regime in which pure PP completely degrades and does not form a char residue when exposed to high temperatures under thermo-oxidative conditions. Fig. 6 compares the surface of a PP-50 composite that has been
3.2.2. Transmission electron microscopy TEM analysis was conducted on the surface char of PP-50, and the images of material scraped from the surface char post mortem are shown in Fig. 7. Clay tactoids are observed embedded in polymer, or carbonaceous material, at various locations in the material. The average d-spacing was measured to be 1.02 nm 0.1 nm, which is slightly less than the d-spacing (001 peak) of 1.16 nm calculated by XRD (2q ¼ 7.6 ) for the starting organically modified clay powder, close to being within the error of the measurements. A 001 peak that formed in the PP-50 starting composite at 2q ¼ 2.19 , corresponding to a d-spacing of 4.03 nm, is attributed to expansion of the clay platelets by intercalation of the polymer. The collapse of the d-spacing after exposure to high temperatures is expected [14], as polymer and/or organic modifiers are removed from the clay galleries. An EDS spectrum collected on this particle indicated the presence of Si, O, Al, Fe, and Mg, all constituents of the clay. Another particle seen in Fig. 8 is rich is silicon and oxygen as indicated in the EDS spectrum. The chemistry of this particle is consistent with the observations made with infrared spectroscopy, discussed in Section
Fig. 6. A comparison of the composite surface of a PP-50 sample after a) heating in a TGA at 10 C/min to 800 C and b) after LPH.
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
2501
Fig. 9. IR spectroscopy of PP-0% clay through PP-50% clay before and after LPH at 44 J/ cm2 (8.8 kW/cm2, 3.1 J, 5 ms). After laser heating, CeH bonding decreases and SieO bonding becomes more concentrated at the surface, as seen by a large increase in the SieO/CeH peak height ratio.
Fig. 7. TEM of the char after LPH for sample PP-50 showing clay tactoids in a polymer. The average d-spacing of the particle’s layers seen in the inset micrograph were measured at d ¼ 1.02 nm.
3.2.3. While the particles in these TEM images appear to be welldispersed, it is likely that larger aggregates occur in the bulk composite, particularly in the higher weight percent composites. The SEM and TEM analyses show that there are diverse and complex morphologies/particles evident on the surface of the nanocomposites after LPH. These include features that are rich in clay elements like silicon and oxygen and also features that contain these clay elements with larger amounts of carbon. Carbon black particles were seen in the TEM analysis of the surface char of the PP-50 sample. These particles could have survived the LPH, or they could have been inadvertently scraped from below the char during sample preparation. The rapid heating during LPH, to presumably very high temperatures, results in complex degradation reactions, melting, and ablation, which correspondingly induce complex surface features and chemistry. Measurement of the temperatures experienced during LPH for PP and PP-clay composites is ongoing work in our laboratory and will be the subject of another publication. 3.2.3. Infrared spectroscopy The FTIR spectroscopy results in Fig. 9 demonstrate how the chemical states of the nanocomposites differ before and after exposure to LPH. A pure PP material has strong characteristic absorption peaks for CeH bending at 1450 and 1375 cm1and CeH stretching at 2800e3000 cm1 from the CeH bonds along the
aliphatic polymer chains, as well as the side CH3 groups along these chains. As clay is added to the polymer, absorption peaks for SieO stretching become apparent at 1033 cm1. After LPH for any of the samples containing the montmorillonite, there is a decrease in the relative intensity of the CeH bending and stretching bands and an increase in the relative intensity of the SieO stretching bands observed in the FTIR spectra. In particular, the ratio of the SieO peak height (1033 cm1) to the CeH peak height (1375 cm1) for each clay composite increases significantly after LPH when compared to the non-irradiated composites. This is especially true for the PP-25 and PP-50 materials, which show a 6 and 9 increase in the ratio value, respectively. A clay-rich char layer forms in the PP-MMT composites, while the pure PP shows no char layer, only complete ablation of the polymer. This FTIR data is consistent with Qin et al. [8] who report similar FTIR data with a cone calorimeter combustion test of a PP-5wt% MMT composite. The peaks corresponding to the polymer CeH bonds nearly disappear for the PP-50 material after LPH, a consequence of the initially lower polymer content and subsequent degradation of the polymer molecules. Besides the obvious fact that SieO bonds will increase as the percentage clay added to the polymer increases, there are several phenomena that contribute to the decrease in CeH bonds found at the surface relative to SieO bonds after LPH. First, dehydrogenation of the polymer chains as a result of thermal degradation in air [29] will result in fewer CeH bonds relative to the number of SieO bonds. Secondly, ablation of polymer fragments that still contain CeH bonding will also result in a lower ratio. A third mechanism that can result in a higher relative SieO concentration at the surface is a percolation mechanism in the molten polymer [30,31]. We observe evidence of melting and gas bubbles in our materials during LPH and it is possible that clay particles can be propelled to the surface by this mechanism. 3.2.4. X-ray photoelectron spectroscopy XPS of the surface of the materials before and after LPH provide subtle composition changes induced by the laser heating. XPS was performed for several of the materials (PP-0, PP-5, and PP-10) before and after LPH as shown in Table 1. As expected, the PP-5 and PP-10 materials have higher atomic concentrations of Al, O, Table 1 Compositions determined by XPS.
Fig. 8. TEM image of Si and O rich particles. The arrow is pointing to the particle for which the EDS spectrum is included.
Sample
C
O
Al
Si
Fe
Si/Al
PP-0%, laser shot PP-0%, no laser shot PP-05%, laser shot PP-05%, no laser shot PP-10%, laser shot PP-10%, no laser shot
99.7 99.6 95.3 92.1 69.9 84.7
0.3 0.3 3.3 6.4 20.4 11.2
0.0 0.0 0.4 0.3 2.6 0.5
0.0 0.1 1.0 1.1 6.9 3.2
0.0 0.0 0.0 0.1 0.2 0.3
n/a n/a 2.6 3.3 2.7 6.1
2502
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
Si, and Fe, and those relative concentrations generally increase at the surface after LPH, with the exception of O, which decreases in the PP-5 after irradiation. The non-irradiated PP-0 contained a trace amount of Si, which we attribute to contamination during processing. These contaminants disappear after LPH. The higher clay content sample, PP-10, showed very small peaks for Na KLL and N1s which would be present in the organically modified clay. The Si 2p and Al 2p binding energies were higher on the laser irradiated samples. This could either be a charging-related phenomenon, or it could be an indication of a change in the clay structure/composition into a material composed of SiO2 and Al2O3. In any case, the XPS results are generally consistent with the EDS and FTIR data showing that Si, Al, and O become more predominant on the surface of nanocomposites that have been laser heated. 4. Conclusions The thermal degradation of PP-montmorillonite nanocomposite was studied using laser pulse heating in order to achieve heating rates that are many orders of magnitude larger than thermogravimetric analysis. The thermal stability of PP is increased with the addition of nanoclay as measured by the decreased mass loss of the composite for a given laser intensity and energy. In a mechanism similar to polymer-nanoclay composites subjected to typical heating rates in TGA and cone calorimetry testing, the nanocomposites exposed to LPH form a protective carbonaceous/silicate-rich layer on the surface. Several spectroscopic techniques confirmed that this layer is rich in Si, O and Al, while being deficient in C and the Ce H bonds normally observed in the polymer. The rapid heating experienced during LPH results in a variety of unique morphological features and chemistry on the surface of the nanocomposite as a result of altered degradation mechanisms of the composite, melting and bubble formation, and ablation. The composite surface shows a more disrupted char layer on the surface after LPH as compared to a composite surface heated in TGA under a traditional heating rate.
for assistant with TEM, Evans Analytical Group for XRD and XPS services, and Mr. Joe Carter for XRD assistance at Benet Laboratories.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]
Acknowledgments [30]
The authors acknowledge Dr. Tom Murray at the State University of New York, College of Nanoscale Science and Engineering
[31]
Luo J-J, Daniel IM. Comp Sci Tech 2003;63:1607e16. Han Z, Fina A. Prog Polym Sci 2011;36:914e44. Okamoto M, Morita S, Kotaka T. Polymer 2001;42(6):2685e8. Beall GW, Powell CE. Fundamentals of polymer-clay nanocomposites. Cambridge University Press; 2011. Vaia RA, Ishii H, Giannelis EP. Chem Mater 1993;5(12):1694e6. Giannelis E. Adv Mater 1996;8:29e35. Bertini F, Canetti M, Audisio G, Costa G, Falqui L. Polym Degrad Stab 2006;91: 600e5. Qin H, Zhang S, Zhao C, Feng M, Yang M, Shu Z, et al. Polym Degrad Stab 2004;85:807e13. Jang B-N, Wilkie CA. Polymer 2005;46:2933e42. Golebiewski J, Galeski A. Comp Sci Tech 2007;67:3442e7. Leszczynska A, Njuguna, Pielichowski K, Banerjee JR. Thermochimica Acta 2007;453:75e96. Leszczynska A, Njuguna, Pielichowski K, Banerjee JR. Thermochimica Acta 2007;454:1e22. Gilman JW, Jackson CL, Morgan AB, Harris RH, Manias E, Giannelis EP, et al. Chem Mater 2000;12:1866e73. Gilman JW, Harris RH, Shields JR, Kashiwagi T, Morgan AB. Polym Adv Tech 2006;17:263e71. Zhu J, Morgan AB, Lamelas FJ, Wilkie CA. Chem Mater 2001;13:3774e80. Kim S, Wilkie CA. Polym Adv Technol 2008;19:496e506. Wang J, Du J, Zhu J, Wilkie CA. Polym Degrad Stab 2002;77:249e52. Mittal V, editor. Thermally stable and flame retardant polymer nanocomposites. Cambridge University Press; 2011. Warrender JM, Mulligan CP, Underwood JH. Wear 2007;263:1540e4. Gao F, Price D, Milnes GJ, Eling B, Lindsay CI, McGrail PT. J Anal Appl Pyrol 1997;40-41:217e31. Price D, Milnes GJ, Gao F. Polym Degrad Stab 1996;54:235e40. Bodzay B, Marosfoi BB, Igricz T, Bocz K, Marosi G. J Anal Appl Pyrol 2009;85: 313e20. Price D, Gao F, John Milnes G, Eling B, Lindsay CI, McGrail PT. Polym Degrad Stab 1999;64:403e10. Bourbigot S, Gilman JW, Wilkie CA. Polym Degrad Stab 2004;84:483e92. Jang BN, Costache M, Wilkie CA. Polymer 2005;46:10678e87. Choudalakis G, Gotsis AD. Euro Polym J 2009;45:967e84. Gilman J, Kashiwagi T, Lichtenhan JD. In: Proc of the 6th European meeting on fire retardancy of polymeric materials. FRPM; 1997. p. 19e20. Lippert T, Dickinson JT. Chem Rev 2003;103:453e85. Zanetti M, Camino G, Reichert P, Mülhaupt R. Macromol Rapid Commun 2001;22:176e80. Okada A, Kawasumi M, Kurauchi T, Kamigaito O. Polym Preprints 1987;28: 447e8. Lewin M, Pearce EM, Levon K, Mey-Marom A, Zammarano M, Wilkie CA, et al. Polym Adv Tech 2006;17:226e34.
Contents lists available at ScienceDirect
Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab
Thermal stability of polypropylene-clay nanocomposites subjected to laser pulse heating Stephen F. Bartolucci a, *, Karen E. Supan b, Jeffrey S. Wiggins c, Lawrence LaBeaud c, Jeffrey M. Warrender a a b c
US Army Armaments Research Development and Engineering Center e Benét Laboratories, Watervliet, NY 12189, USA Norwich University, Department of Mechanical Engineering, Northfield, VT 05663, USA University of Southern Mississippi, School of Polymers and High Performance Materials, Hattiesburg, MS 39406, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 May 2013 Received in revised form 22 August 2013 Accepted 8 September 2013 Available online 21 September 2013
Polypropylene based nanocomposites filled with montmorillonite nanoclay prepared by twin screw extrusion have been studied for thermal stability at high heating rates. In contrast to traditional thermal stability and flammability studies of polymer nanocomposites using heating rates on the order of tens of degrees per minute, this study achieves heating rates that are six orders of magnitude higher. This was accomplished using laser pulse heating. The results show that the nanoclay increases thermal stability of the polymer, as measured by a decrease in the mass loss for a laser pulse at a given energy and intensity. Electron microscopy and various spectroscopic techniques show that a silicate-rich char layer may provide the mechanism for protection of the polymer and decreased degradation rates. The results of the study are compared to the typical results found in traditional thermal stability testing. Published by Elsevier Ltd.
Keywords: Nanoclay Nanocomposite Laser pulse heating Polypropylene Thermal stability Degradation
1. Introduction The addition of nanoscale material to polymer matrices for improved mechanical, thermal, and electrical properties has been studied extensively in recent years [1e3]. One of these nanoscale materials, montmorillonite (MMT), is a layered-aluminosilicate clay consisting of silicon dioxide tetrahedral layers with a metal oxide octahedral center layer [4]. The nanoclay has plate-like morphology with a thickness on the order of 1 nm and in-plane dimensions on the order of several hundred nanometers. Naturally hydrophilic, the nanoclay must be organically modified in order to be dispersed into a hydrophobic polymer. This is typically accomplished by substituting the sodium counter ions with, for example, an organic quaternary ammonium ion that contains alkyl chains. Van der Waals or hydrogen bonding is responsible for holding the clay platelets together, which allows the plates to easily expand or contract dependent upon the cation that is exchanged between the plates. The d-spacing of the clay platelets is easily measured by X-ray diffraction (XRD) and can give information on the intercalative state of the clay. Layered-silicate nanoclay can be dispersed * Corresponding author. Tel.: þ1 518 266 5189; fax: þ1 518 266 5161. E-mail address: [email protected] (S.F. Bartolucci). 0141-3910/$ e see front matter Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.polymdegradstab.2013.09.005
into polymers using melt-mixing techniques [5,6]. Well-dispersed and exfoliated clay can provide improved properties for the polymer, such as increased thermal stability. The thermal stability of a polymer is measured as the resistance to weight loss and chemical change, or degradation, when exposed to higher temperatures. The analytical technique conventionally used for characterizing thermal stability is thermogravimetric analysis (TGA), where the mass of the sample is monitored as the temperature increases. Thermal stability tests have traditionally been performed under relatively slow heating conditions, at heating rates on the order of 20 C/min or below, while some equipment is able to reach heating rates about one order of magnitude higher. Many studies have shown that the addition of nanoclay into polymer matrices generally increases the thermal stability of a polymer during TGA characterization [7e12]. Qin et al. demonstrated that a 5 wt% organically modified montmorillonite addition to polypropylene can increase the temperature at which the peak degradation rate occurs by about 50 C compared to the pure polymer when heated at 20 C/min in air [8]. In addition to thermal stability testing by TGA, several groups have conducted research on the flammability properties of polymer-nanoclay composites [13e 18]. In work by Gilman et al., polypropylene (PP)-layered silicate nanocomposites had a 75% lower peak heat release rate, HRR,
2498
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
compared to the pure polymer during cone calorimetry testing [13]. The enhanced thermal stability and flammability resistance of the nanoclay-polymer composites is attributed to the formation of a protective char layer in the nanocomposite, a multilayered carbonaceous-silicate structure that acts as an excellent mass transport barrier, slowing the flow of heat into the polymer and slowing the escape of the volatile decomposition products from the polymer [14]. While these testing methods provide information on the behavior of polymer nanocomposite during slow heating rate regimes, they are inadequate in describing how the material behaves during extremely high heating rates that may be of interest in applications such as ballistic environments [19]. The application of laser pulse heating (LPH) with a variable pulse duration millisecond laser opens a new regime of study on these materials at heating rates up to six orders of magnitude higher than those achievable with conventional methods. There is literature precedent for investigating the degradation of polymers (simulating rapid heating during severe fires) using laser heating [20–23]. Here, we attempt to extend this work to the case of polymer nanocomposites, using a millisecond pulsed laser. 2. Experimental 2.1. Preparation of the nanocomposites The clay nanocomposites were prepared by mixing in a twin screw extruder at 100 rpm. Zones 1 through 4 of the extruder were at 150 C, 190 C, 190 C, 190 C, respectively and the die was at 190 C. The extrudate was subsequently passed through an ice water bath before being pelletized. The pellets were stored under vacuum at elevated temperature (40 C) overnight to remove any residual moisture. Dried pellets were then fed into a single screw extruder (Zone 1e170 C, Zone 2e180 C and Zone 3e180 C) and extruded through a circular die for the sole purpose of forming a rod shape of a desired diameter unavailable to us with the twin screw dies. The rod was then machined into slices approximately 2 mm thick for laser pulse heating experiments. Pure isotactic polypropylene (Fortilene, Melt Flow Index of 2.8 g/10 min, at 230 C and 2.16 kg) was used to let-down the clay masterbatch. The clay masterbatch (Nanocor, NanoMax-PP) contained 50 wt% organically modified (quaternary ammonium) Montmorillonite (MMT) clay with compatibilizers, grafted maleic anhydride (PP-g-MA). The polypropylene used in the masterbatch has a Melt Flow Index of 20 g/10 min, according to the manufacturer. All material included 1 wt% carbon black (Cabot, Monarch 120) to give the materials uniform optical absorption during laser irradiation. Pure polypropylene (PP-0), PP with 5 wt% MMT (PP-5), 10 wt% MMT (PP-10), 25 wt% MMT (PP-25) and 50 wt% MMT (PP-50) samples were prepared for this study.
be approximately 8 0.2%, meaning most of the laser light was absorbed into the sample, as desired. 2.3. Characterization of nanocomposites The mass of the sample was weighed before and after irradiation on a microbalance with a readability of 0.001 mg (Sartorius ME-36 S). Ten samples for each condition were run for statistics. Scanning Electron Microscopy (SEM) was performed on a JEOL-840 electron microscope operating at 10 kV equipped with Energy Dispersive Spectroscopy (EDS) with a Thermo Electron 457A X-ray detector, and an FEI Helios Nanolab 600i Field Emission Electron Microscope (FE-SEM) operating at 5 kV. Samples were coated with a thin layer of gold in order to reduced charging effects. Fourier Transform Infrared Spectroscopy (FTIR) was performed on a PerkineElmer Spectrum One Spectrometer operating in Attenuated Total Reflectance mode (ATR) between 4000 cm1 to 515 cm1 at 4 cm1 resolution. X-ray Photoelectron Spectroscopy (XPS) was conducted on a PHI5701 LSci with a monochromated Al-Ka 1486.6 eV X-ray source (Evans Analytical Group). X-ray diffraction (XRD) was performed on a Bruker D8 Discover diffractometer with Cu-Ka radiation (l ¼ 1.505945 Å) from 5 to 25 2q with a 240 s accumulation time. The distance, between clay sheets, d, was calculated by Bragg’s Law, nl ¼ 2d sin q, using 2q peaks from the XRD spectra. Transmission Electron Microscopy (TEM) samples were prepared by carefully removing char material from the surface of a LPH region of the nanocomposite under a microscope using ultrafine tweezers. The material was deposited onto a carbon coated Cu TEM grid for viewing on an FEI Titan Environmental S/TEM operating at 300 kV. Profilometry of LPH samples was conducted on a Nanovea ST400 non-contact optical profilometer. The total ablated volume, as defined by the material loss below the initial sample, and the volume above the original surface, was determined by 3-D surface topography. 3. Results and discussion 3.1. Mass loss measurements Clay nanocomposites with varying clay content were subjected to conventional TGA and LPH. The TGA results confirmed the behavior expected from the literature, namely, increasing clay content leads to increased thermal stability for a fixed heating rate. The results of the TGA are similar to those published previously in the literature [7,8]. The mass loss for each clay nanocomposite during LPH is presented in Fig. 1. Error bars on the mass loss data
2.2. Laser pulse heating The Laser Pulse Heating (LPH) setup included an Nd:YAG laser (MegaWatt Lasers, Inc) that sends 1064 nm light into an optical fiber. The light from the fiber was passed through a 50 mm focal length collimating lens, followed by a 250 mm focal length objective lens, which produced a laser spot diameter of 3.0 mm (0.003 m). The incident laser energy was measured using a beam splitter (Newport) and an energy meter (Ophir Nova). The laser pulse energy and pulse duration can be adjusted, but all samples in this paper were irradiated at 3.1 J pulses lasting 5 ms, corresponding to a fluence of 44 J/cm2 and an intensity of 8.8 kW/cm2. The samples were irradiated in air to give thermo-oxidative conditions. The reflectance of the samples was measured in a UVevis spectrometer (PerkineElmer Lambda 40) at 1064 nm and found to
Fig. 1. Mass loss and volume loss of nanoclay composite material after laser pulse heating with a fluence of 44 J/cm2 (8.8 kW/cm2, 3.1 J, 5 ms) (average of 10 samples per data point).
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
correspond to the standard deviation of the 10 measurements for each composite material. As an increasing fraction of nanoclay is added to the composite, the mass loss decreases monotonically. In previous TGA and flammability studies, mass loss, and more specifically, mass loss rate, was seen to decrease with the addition of nanoclay to the polymer [13]. We present the volume loss obtained from two different methods. In the first, the volume loss is calculated from the mass loss and the weight fraction of clay in the nanocomposite, using 888 kg/m3 for the density of pure PP with 1% carbon black and 1345 kg/m3 for the density of the organically modified clay. For the second (open triangles), we directly measured the volume of the “crater” of removed material in the irradiated area using profilometry. These two methods give good agreement for samples with low clay content, but disagree increasingly with high clay content. At high clay content, the crater volume is negligible, and the profilometry measurements show features that extend above the original surface of the material (not included in calculated volume). We attribute this behavior to the formation of a char layer, and to the presence of trapped gaseous species underneath this layer. We adduce further support for this interpretation in Section 3.2.
3.2. Analysis of char formation 3.2.1. Electron microscopy and EDS Polypropylene is a non-charring thermoplastic polymer. When the polymer is pyrolyzed, it will completely degrade. When nanoclay is mixed into the polymer, it has been reported [24] that the clay acts as a promoter of char formation within the polymer when the polymer is exposed to degradation temperatures. The clay can slow down the degradation of the polymer and alter the degradation mechanisms of the polymer molecules and, hence, the kinetics of degradation [24,25]. While the clay itself can slow oxygen diffusion into the composite [26], the char layer further acts as a protective barrier for the remaining polymer as it slows heat and oxygen from entering the polymer and slows the escape of volatile decomposition products from the remaining polymer [27]. Electron microscopy in Fig. 2 reveals a different surface morphology for the various nanocomposites after LPH. The pure PP-0 material in (a) has a smooth, flat surface indicative of thermal ablation and volatile formation [28]. In contrast, the remaining materials show increased surface roughness with increasing clay content (b)e(e). The insets
2499
Fig. 3. Cross sectional SEM micrograph of PP-0 after LPH. A smooth ablation crater with no char layer is seen. The sample was scored and fractured through the LPH region at room temperature.
show a higher magnification top view of the LPH region of each sample. The lower-clay content nanocomposites show areas that are qualitatively similar to those of the PP-0, but also show areas with larger surface elevation differences, and some areas with different morphology. The higher-clay content nanocomposites have a dramatically different appearance compared to the PP0 sample. In order to study this further, the composites were fractured at room temperature through the LPH region and the cross sections were viewed in the SEM, with EDS mapping. Examples of the PP0 and PP-50 composites are shown in Figs. 3 and 4. In PP-50, a silicate-rich char layer is seen on the surface of the ablation crater. Underneath this layer appears to be a thin layer of polymer that has been exposed to high enough temperatures to cause melting and bubble formation from volatiles. Polymer exists below this melting and bubble region that does not appear to have been affected by high temperatures. The melting and bubble formation can cause ruptures in the char layer forcing the silicate-rich material to ablate or form structures perpendicular to the composite surface. An example of this phenomenon is seen in Fig. 5, in which we show a bubble that appears to have ruptured. EDS mapping shows the surface layer to be rich in Si, O, and Al, while being deficient in C, Na,
Fig. 2. SEM of nanoclay composites after Laser Pulse Heating with a fluence of 44 J/cm2 (8.8 kW/cm2, 3.1 J, 5 ms). a) 0%, b) 5%, c) 10%, d) 25% and e) 50% nanoclay.
2500
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
Fig. 4. Cross sectional SEM micrograph of PP-50 after LPH. EDS spectral map shows that the char layer is rich is Si, O and Al, and deficient in C.
heated by TGA at 10 C/min (a) with another that was heated by LPH (b). The micrographs of the two surfaces, taken at the same magnification, are qualitatively very different. In the slowly heated composite, the surface char is relatively intact with small surface holes scattered throughout. Trapped volatiles underneath the surface will slowly cause increased pressures, resulting in the eventual formation of holes in the surface as the volatiles breakthrough and pressure is relieved. In contrast, the LPH surface shows a disrupted surface char, presumably a result of the more drastic heating rate and ablation. Rapid volatile formation causes a more violent release of pressure resulting in the more disrupted composite char surfaces. EDS of the surfaces show a higher relative Si, O, Al, and Fe content and lower C content in this region of the LPH surface, as compared to the TGA heated surface.
Fig. 5. Silicate-rich feature that has been ruptured due to bubble expansion in the underlying material (PP-50 material after LPH).
and Ca (not shown in EDS map). The C deficiency is due to polymer degradation and volatilization. An interesting observation is that the exchangeable Na and Ca cations present in the clay appear to have decreased signal in the LPH region, unlike the Si and Al, which increase in signal. These cations may have been removed from between the aluminosilicate layers along with the organic modifiers that some of the cations were substituted with. Unlike the composite that contains the nanoclay, the PP-0 has a smooth ablation crater and no char layer formed. The polymer exposed to the heat has volatized, leaving more pure polymer exposed. This is comparable to the slow heating rate regime in which pure PP completely degrades and does not form a char residue when exposed to high temperatures under thermo-oxidative conditions. Fig. 6 compares the surface of a PP-50 composite that has been
3.2.2. Transmission electron microscopy TEM analysis was conducted on the surface char of PP-50, and the images of material scraped from the surface char post mortem are shown in Fig. 7. Clay tactoids are observed embedded in polymer, or carbonaceous material, at various locations in the material. The average d-spacing was measured to be 1.02 nm 0.1 nm, which is slightly less than the d-spacing (001 peak) of 1.16 nm calculated by XRD (2q ¼ 7.6 ) for the starting organically modified clay powder, close to being within the error of the measurements. A 001 peak that formed in the PP-50 starting composite at 2q ¼ 2.19 , corresponding to a d-spacing of 4.03 nm, is attributed to expansion of the clay platelets by intercalation of the polymer. The collapse of the d-spacing after exposure to high temperatures is expected [14], as polymer and/or organic modifiers are removed from the clay galleries. An EDS spectrum collected on this particle indicated the presence of Si, O, Al, Fe, and Mg, all constituents of the clay. Another particle seen in Fig. 8 is rich is silicon and oxygen as indicated in the EDS spectrum. The chemistry of this particle is consistent with the observations made with infrared spectroscopy, discussed in Section
Fig. 6. A comparison of the composite surface of a PP-50 sample after a) heating in a TGA at 10 C/min to 800 C and b) after LPH.
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
2501
Fig. 9. IR spectroscopy of PP-0% clay through PP-50% clay before and after LPH at 44 J/ cm2 (8.8 kW/cm2, 3.1 J, 5 ms). After laser heating, CeH bonding decreases and SieO bonding becomes more concentrated at the surface, as seen by a large increase in the SieO/CeH peak height ratio.
Fig. 7. TEM of the char after LPH for sample PP-50 showing clay tactoids in a polymer. The average d-spacing of the particle’s layers seen in the inset micrograph were measured at d ¼ 1.02 nm.
3.2.3. While the particles in these TEM images appear to be welldispersed, it is likely that larger aggregates occur in the bulk composite, particularly in the higher weight percent composites. The SEM and TEM analyses show that there are diverse and complex morphologies/particles evident on the surface of the nanocomposites after LPH. These include features that are rich in clay elements like silicon and oxygen and also features that contain these clay elements with larger amounts of carbon. Carbon black particles were seen in the TEM analysis of the surface char of the PP-50 sample. These particles could have survived the LPH, or they could have been inadvertently scraped from below the char during sample preparation. The rapid heating during LPH, to presumably very high temperatures, results in complex degradation reactions, melting, and ablation, which correspondingly induce complex surface features and chemistry. Measurement of the temperatures experienced during LPH for PP and PP-clay composites is ongoing work in our laboratory and will be the subject of another publication. 3.2.3. Infrared spectroscopy The FTIR spectroscopy results in Fig. 9 demonstrate how the chemical states of the nanocomposites differ before and after exposure to LPH. A pure PP material has strong characteristic absorption peaks for CeH bending at 1450 and 1375 cm1and CeH stretching at 2800e3000 cm1 from the CeH bonds along the
aliphatic polymer chains, as well as the side CH3 groups along these chains. As clay is added to the polymer, absorption peaks for SieO stretching become apparent at 1033 cm1. After LPH for any of the samples containing the montmorillonite, there is a decrease in the relative intensity of the CeH bending and stretching bands and an increase in the relative intensity of the SieO stretching bands observed in the FTIR spectra. In particular, the ratio of the SieO peak height (1033 cm1) to the CeH peak height (1375 cm1) for each clay composite increases significantly after LPH when compared to the non-irradiated composites. This is especially true for the PP-25 and PP-50 materials, which show a 6 and 9 increase in the ratio value, respectively. A clay-rich char layer forms in the PP-MMT composites, while the pure PP shows no char layer, only complete ablation of the polymer. This FTIR data is consistent with Qin et al. [8] who report similar FTIR data with a cone calorimeter combustion test of a PP-5wt% MMT composite. The peaks corresponding to the polymer CeH bonds nearly disappear for the PP-50 material after LPH, a consequence of the initially lower polymer content and subsequent degradation of the polymer molecules. Besides the obvious fact that SieO bonds will increase as the percentage clay added to the polymer increases, there are several phenomena that contribute to the decrease in CeH bonds found at the surface relative to SieO bonds after LPH. First, dehydrogenation of the polymer chains as a result of thermal degradation in air [29] will result in fewer CeH bonds relative to the number of SieO bonds. Secondly, ablation of polymer fragments that still contain CeH bonding will also result in a lower ratio. A third mechanism that can result in a higher relative SieO concentration at the surface is a percolation mechanism in the molten polymer [30,31]. We observe evidence of melting and gas bubbles in our materials during LPH and it is possible that clay particles can be propelled to the surface by this mechanism. 3.2.4. X-ray photoelectron spectroscopy XPS of the surface of the materials before and after LPH provide subtle composition changes induced by the laser heating. XPS was performed for several of the materials (PP-0, PP-5, and PP-10) before and after LPH as shown in Table 1. As expected, the PP-5 and PP-10 materials have higher atomic concentrations of Al, O, Table 1 Compositions determined by XPS.
Fig. 8. TEM image of Si and O rich particles. The arrow is pointing to the particle for which the EDS spectrum is included.
Sample
C
O
Al
Si
Fe
Si/Al
PP-0%, laser shot PP-0%, no laser shot PP-05%, laser shot PP-05%, no laser shot PP-10%, laser shot PP-10%, no laser shot
99.7 99.6 95.3 92.1 69.9 84.7
0.3 0.3 3.3 6.4 20.4 11.2
0.0 0.0 0.4 0.3 2.6 0.5
0.0 0.1 1.0 1.1 6.9 3.2
0.0 0.0 0.0 0.1 0.2 0.3
n/a n/a 2.6 3.3 2.7 6.1
2502
S.F. Bartolucci et al. / Polymer Degradation and Stability 98 (2013) 2497e2502
Si, and Fe, and those relative concentrations generally increase at the surface after LPH, with the exception of O, which decreases in the PP-5 after irradiation. The non-irradiated PP-0 contained a trace amount of Si, which we attribute to contamination during processing. These contaminants disappear after LPH. The higher clay content sample, PP-10, showed very small peaks for Na KLL and N1s which would be present in the organically modified clay. The Si 2p and Al 2p binding energies were higher on the laser irradiated samples. This could either be a charging-related phenomenon, or it could be an indication of a change in the clay structure/composition into a material composed of SiO2 and Al2O3. In any case, the XPS results are generally consistent with the EDS and FTIR data showing that Si, Al, and O become more predominant on the surface of nanocomposites that have been laser heated. 4. Conclusions The thermal degradation of PP-montmorillonite nanocomposite was studied using laser pulse heating in order to achieve heating rates that are many orders of magnitude larger than thermogravimetric analysis. The thermal stability of PP is increased with the addition of nanoclay as measured by the decreased mass loss of the composite for a given laser intensity and energy. In a mechanism similar to polymer-nanoclay composites subjected to typical heating rates in TGA and cone calorimetry testing, the nanocomposites exposed to LPH form a protective carbonaceous/silicate-rich layer on the surface. Several spectroscopic techniques confirmed that this layer is rich in Si, O and Al, while being deficient in C and the Ce H bonds normally observed in the polymer. The rapid heating experienced during LPH results in a variety of unique morphological features and chemistry on the surface of the nanocomposite as a result of altered degradation mechanisms of the composite, melting and bubble formation, and ablation. The composite surface shows a more disrupted char layer on the surface after LPH as compared to a composite surface heated in TGA under a traditional heating rate.
for assistant with TEM, Evans Analytical Group for XRD and XPS services, and Mr. Joe Carter for XRD assistance at Benet Laboratories.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]
Acknowledgments [30]
The authors acknowledge Dr. Tom Murray at the State University of New York, College of Nanoscale Science and Engineering
[31]
Luo J-J, Daniel IM. Comp Sci Tech 2003;63:1607e16. Han Z, Fina A. Prog Polym Sci 2011;36:914e44. Okamoto M, Morita S, Kotaka T. Polymer 2001;42(6):2685e8. Beall GW, Powell CE. Fundamentals of polymer-clay nanocomposites. Cambridge University Press; 2011. Vaia RA, Ishii H, Giannelis EP. Chem Mater 1993;5(12):1694e6. Giannelis E. Adv Mater 1996;8:29e35. Bertini F, Canetti M, Audisio G, Costa G, Falqui L. Polym Degrad Stab 2006;91: 600e5. Qin H, Zhang S, Zhao C, Feng M, Yang M, Shu Z, et al. Polym Degrad Stab 2004;85:807e13. Jang B-N, Wilkie CA. Polymer 2005;46:2933e42. Golebiewski J, Galeski A. Comp Sci Tech 2007;67:3442e7. Leszczynska A, Njuguna, Pielichowski K, Banerjee JR. Thermochimica Acta 2007;453:75e96. Leszczynska A, Njuguna, Pielichowski K, Banerjee JR. Thermochimica Acta 2007;454:1e22. Gilman JW, Jackson CL, Morgan AB, Harris RH, Manias E, Giannelis EP, et al. Chem Mater 2000;12:1866e73. Gilman JW, Harris RH, Shields JR, Kashiwagi T, Morgan AB. Polym Adv Tech 2006;17:263e71. Zhu J, Morgan AB, Lamelas FJ, Wilkie CA. Chem Mater 2001;13:3774e80. Kim S, Wilkie CA. Polym Adv Technol 2008;19:496e506. Wang J, Du J, Zhu J, Wilkie CA. Polym Degrad Stab 2002;77:249e52. Mittal V, editor. Thermally stable and flame retardant polymer nanocomposites. Cambridge University Press; 2011. Warrender JM, Mulligan CP, Underwood JH. Wear 2007;263:1540e4. Gao F, Price D, Milnes GJ, Eling B, Lindsay CI, McGrail PT. J Anal Appl Pyrol 1997;40-41:217e31. Price D, Milnes GJ, Gao F. Polym Degrad Stab 1996;54:235e40. Bodzay B, Marosfoi BB, Igricz T, Bocz K, Marosi G. J Anal Appl Pyrol 2009;85: 313e20. Price D, Gao F, John Milnes G, Eling B, Lindsay CI, McGrail PT. Polym Degrad Stab 1999;64:403e10. Bourbigot S, Gilman JW, Wilkie CA. Polym Degrad Stab 2004;84:483e92. Jang BN, Costache M, Wilkie CA. Polymer 2005;46:10678e87. Choudalakis G, Gotsis AD. Euro Polym J 2009;45:967e84. Gilman J, Kashiwagi T, Lichtenhan JD. In: Proc of the 6th European meeting on fire retardancy of polymeric materials. FRPM; 1997. p. 19e20. Lippert T, Dickinson JT. Chem Rev 2003;103:453e85. Zanetti M, Camino G, Reichert P, Mülhaupt R. Macromol Rapid Commun 2001;22:176e80. Okada A, Kawasumi M, Kurauchi T, Kamigaito O. Polym Preprints 1987;28: 447e8. Lewin M, Pearce EM, Levon K, Mey-Marom A, Zammarano M, Wilkie CA, et al. Polym Adv Tech 2006;17:226e34.