Reinforcement of nylon 6 with functionalized silica nanoparticles for enhanced tensile strength
and modulus
This article h...
Nanotechnology 19 (2008) 445702 (7pp) doi:10.1088/0957-4484/19/44/445702
Reinforcement of ny...
Nanotechnology 19 (2008) 445702 H Mahfuz et al
Figure 1. (a) Silane coupling agent and functional groups. It has two class...
Nanotechnology 19 (2008) 445702 H Mahfuz et al
Figure 2. Tensile tests of single filaments. The test was carried out
on a Z...
Nanotechnology 19 (2008) 445702 H Mahfuz et al
Figure 3. SEM images of cross-sections of filament samples:
(a) neat nylon 6...
Nanotechnology 19 (2008) 445702 H Mahfuz et al
Figure 4. SEM micrographs of fractured surfaces: (a) neat nylon 6, (b) nylo...
Nanotechnology 19 (2008) 445702 H Mahfuz et al
Figure 5. FTIR spectroscopy—(a) neat nylon 6, (b) nylon 6 with 1 wt% silica...
Nanotechnology 19 (2008) 445702 H Mahfuz et al
The area under the FTIR curves is also a qualitative
measure of the concent...
of 8


Published on: Mar 3, 2016

Transcripts - NanoTechnology-Paper-Mahfuz-Hasan-Vinod

  • 1. Reinforcement of nylon 6 with functionalized silica nanoparticles for enhanced tensile strength and modulus This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2008 Nanotechnology 19 445702 ( Download details: IP Address: The article was downloaded on 02/10/2008 at 14:31 Please note that terms and conditions apply. The Table of Contents and more related content is available HOME | SEARCH | PACS & MSC | JOURNALS | ABOUT | CONTACT US
  • 2. IOP PUBLISHING NANOTECHNOLOGY Nanotechnology 19 (2008) 445702 (7pp) doi:10.1088/0957-4484/19/44/445702 Reinforcement of nylon 6 with functionalized silica nanoparticles for enhanced tensile strength and modulus Hassan Mahfuz1,8 , Mohammad Hasan2 , Vinod Dhanak3 , Graham Beamson4 , Justin Stewart1 , Vijaya Rangari5 , Xin Wei6 , Valery Khabashesku7 and Shaik Jeelani5 1 Department of Ocean Engineering, Florida Atlantic University, Boca Raton, FL 33431, USA 2 Department of Mechanical and Industrial Engineering, University of Toronto, ON, M5S 3G8, Canada 3 Physics Department, University of Liverpool and Daresbury Laboratory, Warrington WA4 4AD, UK 4 National Centre for Electron Spectroscopy and Surface Analysis (NCESS), STFC Daresbury Laboratory, Warrington WA4 4AD, UK 5 Tuskegee University’s Center for Advanced Materials (T-CAM), Tuskegee, AL 36088, USA 6 Department of Chemistry, Texas Southern University, Houston, TX 77004, USA 7 Smalley Institute for Nanoscale Science and Technology, Rice University, Houston, TX 77005, USA E-mail: Received 11 June 2008, in final form 29 August 2008 Published 30 September 2008 Online at Abstract Pristine and functionalized silica (SiO2) nanoparticles were dispersed into nylon 6 and drawn into filaments through melt extrusion. The loading fraction of particles in both cases was 1.0 wt%. Fourier transform infrared (FTIR) studies revealed that reinforcement of pristine silica nanoparticles enhances the bond strength of each of the three basic bonds of nylon 6 namely, hydroxyl, amide, and carbonyl. As a result, the improvement over neat nylon in strength and modulus was 36% and 28% respectively, without any loss of fracture strain (80%). A silane coupling agent was then used through wet chemical treatment to functionalize silica nanoparticles. Functionalization induced an additional covalent Si–O–Si (siloxane) bond between silica particles and nylon backbone polymer while the enhancement in the basic bonds was retained. FTIR and x-ray photoelectron spectroscopy (XPS) studies confirmed the formation of the siloxane bond. This added chemical bond resulted in 76% and 55% improvement in tensile strength and modulus, and still retained 30% fracture strain. Calculation of the upper bound on Young’s modulus indicates that one can reach within 5% of the bound with pristine silica particles, but it is exceeded by 15% when particles are functionalized. (Some figures in this article are in colour only in the electronic version) 1. Introduction Nylon’s toughness, low coefficient of friction, and good abrasion resistance make it an ideal replacement for a wide variety of applications replacing metal and rubber. The amide groups of nylon are very polar, and can hydrogen bond with 8 Author to whom any correspondence should be addressed. each other. Because of these, and because the nylon backbone is so regular and symmetrical, nylons are partially crystalline, and they make very good fiber [1–4]. Nylons are made from a monomer, such as a cyclic amide called lactam. For example, Caprolactam is a lactam with six carbon atoms and nylon made from Caprolactam is called nylon 6. When nylon is spun into fibers, the long chain-like macromolecules line up parallel to 0957-4484/08/445702+07$30.00 © 2008 IOP Publishing Ltd Printed in the UK1
  • 3. Nanotechnology 19 (2008) 445702 H Mahfuz et al Figure 1. (a) Silane coupling agent and functional groups. It has two classes of functionality, X is a hydrolyzable group typically alkoxy, acyloxy, halogen, or amine. Following hydrolysis, a reactive silanol group is formed. (b) Formation of a Si–O–Si (siloxane) bond on the substrate surface. Newly formed silanol can condense with other silanol groups available at the silica surface to form a siloxane linkage. each other. The amide groups on adjacent chains then form strong bonds with each other called hydrogen bonds. These hydrogen bonds hold the adjacent chains together, making nylon yarn strong. When nylon 6 polymerizes, the amide link present in Caprolactam (starting monomer for nylon 6) opens up and the molecules join up in a continuous chain, providing an ideal mechanism for interacting with nanoparticles. On the other hand, silica particles are formed by strong covalent bonds between silicon and oxygen atoms by sharing their electron pairs at the p orbitals. In addition, the surface bound OH groups on silica surfaces offer an opportunity to form stable bonds with nylon or any functional group during polymerization. In order to utilize the extraordinary strength and stiffness of carbon nanotubes in bulk materials, several researchers [5–11] have recently infused CNTs into textile polymeric precursors and attempted to align the acicular particles along the length of the drawn filament. Infusion was carried out either through a liquid route using sonication, or a dry route followed by melt mixing in an extruder. Alignment of CNTs in the filament was enforced by extrusion or spinning followed by stretching. The resulting composites either in consolidated or filament form have no doubt demonstrated improved mechanical and thermal properties. While it was encouraging to see phenomenal improvements in strength and stiffness [12], it was also observed that the improvement was at the cost of sacrificing a significant amount of fracture strain which is not attractive for nylon. In an attempt to improve upon the fracture strain, spherical silica nanoparticles, in place of CNTs, were chosen in the present work. However, the improvements in mechanical properties with spherical silica particles were somewhat less than what we found with CNTs. Functionalization of silica particles was then introduced and it proved to be very effective in enhancing the modulus and strength. 2. Functionalization of silica nanoparticles We used organosilanes to modify the surface of silica nanoparticles. Organosilanes have the ability to incorporate both organic- and inorganic-compatible functionality within the same molecule. The inorganic compatibility comes from the alkoxy groups attached to the silicon atom. This bond is hydrolytically unstable and in the presence of moisture hydrolyzes to an intermediate Si–OH bond which then condenses with surface bound OH groups on inorganic surfaces to form stable Si–O–Si bonds. The molecular structure of organosilane and its interaction with silica substrates is depicted in figure 1. The final result of reacting an organosilane with a substrate is to utilize the substrate to catalyze chemical transformations at the heterogeneous interface, ordering the interface region, and modifying its partition characteristics. Most importantly, it includes the ability to affect a covalent bond between nylon and SiO2 particles. The silane coupling agent used in this investiga- tion was procured from Gelest (Gelest Inc., East Steel Road, Morrisville, PA 19067) and it was a diamine functional Silane–Trialkoxy (SIA05910), N-(2-aminoethyl)-3- aminopropyltriethoxysilane. Functionalization of silica particles was performed in the following manner. The weight (g) of silica nanoparticles that would be necessary for dispersion into nylon was first estimated. The total surface area of silica surface was then calculated from the known specific surface area of silica (440 m2 g−1 ) and the estimated weight. Since one gram of silane was capable of modifying approximately 358 m2 of inorganic surface, we could easily estimate the amount of silane needed for functionalization. Silane was then added through mechanical stirring into a starting mixture of 95% ethanol and 5% water to yield a 2% final concentration of silane. Mixing was allowed for a few minutes for hydrolysis and silanol formation. In the next step, nanoparticles were added by stirring them into the silane solution for about 2– 3 min. After the particles settled at the bottom, the solution was decanted. Ethanol and water were dried out by heating the particles in a furnace at around 110 ◦ C. If particles were agglomerated, a ball mill was used to bring them back to powder form. Silica nanoparticles were now ready to be dispersed into nylon. 2
  • 4. Nanotechnology 19 (2008) 445702 H Mahfuz et al Figure 2. Tensile tests of single filaments. The test was carried out on a Zwick-Roel TC-FR 2.5 TN Materials Testing Machine using a 50 N load cell. An individual filament was attached to the crossheads by special 8201 wedge grips. These grips eliminated slippage and stress concentration at the grips. The gage length for the filaments was 75 mm. The tests were performed at a constant crosshead speed of 1.27 mm min−1 . The diameter of each length of filament was measured using scanning electron microscopy (SEM) at several points along the length since it was not constant for any of the materials, and varied by as much as 5%. An average area estimated from several locations was used in the stress calculations. The diameter of the filament in each category also varied but was around 170 µm. Machine compliance was taken into consideration per ASTM D3379-75 while calculating the modulus. About 15 individual filaments were tested for each category of materials. 3. Filament extrusion In the present investigation, we dry mixed SiO2 nanoparticles (pristine/functionalized) with nylon 6 powder. The amount of nanoparticle loading was 1 wt%. The dry-mixed powder was then melted in a single screw extruder which was followed by distributive mixing, extrusion, stretching, and heat stabilization to continuously draw SiO2-reinforced filaments. In parallel, control filaments were also extruded using identical procedures. The dry mixing of nylon powder with SiO2 was performed in a mechanical blender for 3 h. The SiO2 nanoparticles were procured from MTI Corporation (2700 Rydin road, unit D, Richmond, CA 94804). The nanoparticles were 15–20 nm in diameter. Commercial grade nylon 6 was procured from UBE Industries, Ltd (UBE building 3- 11 Higashi-shinagawwa 2-chome, Shinagawa-Ku, Tokyo 140, Japan). A mechanical crusher was used to produce micron- sized powders of nylon which were then used in the mechanical blender for mixing with SiO2 nanoparticles. The density of nylon 6 was 1.07 g cm−3 with a melting point of 215 ◦ C. In order to eliminate moisture, the nylon and SiO2 mixture was placed into a cylindrical drying chamber. Hot air was supplied to the chamber through an insulated flexible tube using a vortex blower. The dryer was operated for 24 h with the temperature set at 90 ◦ C. Prolonged heating accompanied with a vortex flow broke up large agglomerates of Table 1. Tensile moduli and strength from single filament tests. E-modulus Gain/ Tensile Gain/ Material (GPa) loss (%) strength (MPa) loss (%) Neat nylon 6 1.10 ± 0.10 — 210 ± 10 — SiO2-nylon 6 1.41 ± 0.15 +28 285 ± 8 +36 1% SiO2-silane 1.70 ± 0.17 +55 370 ± 25 +76 Nylon 6 silica particles if any were left after the mechanical blending. Once the mixture was dried, it was extruded through a Wayne Yellow Label Table Top Extruder. Five thermostatically controlled heating zones were used to melt the admixture prior to extrusion. The die zone consisted of a circular plate, a 10 cm long steel tubing with an inner diameter of 4 mm, and the die itself. A distributive mixing of the silica nanoparticles with nylon was enforced through the use of a circular plate with multiple orifices [9, 11]. A specially designed die was used in the process. The die configuration generated two distinct flow regimes that significantly affected the distribution of the particles. After extrusion, filaments were solidified by passing them through chilled water maintained at approximately 10 ◦ C. In the next step, filaments were stretched using a tension- adjuster (Godet), and heat stabilized using the Wayne Yellow Jacket Stabilizing unit. The heater temperature was set at 110 ◦ C, and the filament travel per minute (FPM) was adjusted in the Godet stations to allow continuous drawing of filaments. Finally, the filaments were wound on a spool using a filament winder at a winding speed of about 70 rpm. Several of these spools were produced. 4. Tensile response Representative stress–strain responses from filament tests are shown in figure 2. Improvements in strength and stiffness are listed in table 1. It is observed that the gains in tensile strength and Young’s modulus are around 36% and 28%, respectively with pristine silica infusion. It is also noticed that the fracture strain of the filament in this case is still around 80% and the load–deformation behavior of the nanophased system is very similar to that of the neat system. If one considers 0.2% yield strength, the gain is almost in the identical range. When silica particles are silated (functionalized), the enhancement in strength and modulus increases by 76% and 55%, respectively as seen in table 1 and figure 2. It is also noticed that there is a loss of ultimate strain but it is still around 30% which is considerably higher than for other fibers. If we now calculate the upper bound on Young’s modulus based on micromechanical theories [13], we find it to be 1.48 GPa. The experimental value of 1.41 GPa as shown in table 1 for silica-reinforced nylon is within 5% of the upper bound. When silica particles are functionalized, the modulus increases to 1.7 GPa, exceeding the upper bound by 15%. A similar pattern in the case of strength was observed in an earlier work by Sumita et al [14]. They predicted the strength of thin films made from nylon 6 reinforced with silica and glass fillers using empirical formulations which would give the upper bound of the composite strength as the strength of the unfilled matrix. 3
  • 5. Nanotechnology 19 (2008) 445702 H Mahfuz et al Figure 3. SEM images of cross-sections of filament samples: (a) neat nylon 6, (b) nylon 6 with silica, (c) nylon 6 with functionalized silica. But they found that the strength of the composite when filled with 40 nm silica particles exceeded the neat nylon 6 strength, i.e. the upper bound, by about 16%. The loading of fillers was, however, 20.0 wt%—quite different from 1.0 wt% used in the present investigation. Enhancement in stiffness and strength can be visualized from various aspects. Crystallinity of nylon 6 as determined by differential scanning calorimetry (DSC) was around 24% [13]. With the addition of functionalized silica particles crystallinity increased to 29%. We have also noticed a corresponding shift in the crystalline temperature from 189 to 196 ◦ C. Similar changes in crystalline temperature have been reported for nylon 66 when acid dyes were added [15]. Since changes in per cent crystallinity in our case were small, their effect on the mechanical properties was modest. Similar observations have been made in [16] where changes in the crystallinity of nanocomposites were not very evident with the addition of different types of silica particles. It was also indicated that improvement in mechanical properties was mainly due to the effect of nanofillers. The increase in stiffness is no doubt contributed by the higher stiffness (nylon 1.1 GPa, silica 72.3 GPa) of the silica reinforcement. One order higher stiffness of silica would surely influence the stiffness of the composites simply from the rule of mixture concept. Although the volume fraction of silica is small, its contribution to stiffness is indeed substantial. The other reason for an increase in stiffness may be explained from the viscoelastic approach as the morphological structure of polymer is changed due to inclusion [17]. As the filler particles are surrounded by the shell of matrix there will be a relative increase in the particle diameter which effectively increases the particle volume fraction in the composite, which in turn influences the stiffness. It is also known [18] that an increase in particle diameter can be related to an increase in loss modulus of the composite which allows determination of the composite modulus [13]. The key to such improvement will, however, depend on the dispersion of particles within the body of the matrix. The SEM images of the surface morphologies of the filament cross-section for neat nylon 6 and nylon 6 filled with silica and functionalized silica nanoparticles are shown in figure 3. It is seen in figure 3 that in comparison with a quite rough surface of neat nylon 6 (figure 3(a)) the surface of the silica-filled sample (figure 3(b)) looks smooth, showing the embedded SiO2 nanoparticles (outlined by arrows). Their average size is quite large (about 200 nm) which indicates clustering. In contrast, the image of nylon 6, filled with functionalized silica particles, shows the dispersion of the latter in the form of 20–40 nm size individual nanoparticles, as shown by several arrows in figure 3(c). This improved dispersion can be attributed to aminosilane treatment of the silica surface and interaction of terminal amino groups on the functionalized silica surface with nylon 6 functional groups. The improvement in strength can be explained from a different perspective as it is dictated by the fracture process rather than by the deformation behavior. This fracture process is mostly governed by the interfacial adhesion between the matrix and the filler particles which essentially controls the load transfer to the filler. 5. Filament fracture process The fracture process in nylon filament has three distinct features. First is the formation of a V-notch, sometimes called a razor notch, at the beginning of the fracture process [19–21]. As the stress on the filament is increased, this is followed by a segment of slow crack growth region characterized by ductile cleaving across a straight front from the edge of the V-notch, as seen in figure 4. The third stage is a fast crack growth region featured by a large structural difference of the surface, giving a melted and rough appearance (figures 4(a) and (b)). This rough portion or the third stage fracture surface is associated with the catastrophic failure of the filament. A gaping hole is noticed in figure 4(a) which has evidently grown from a material flaw originated during the melt extrusion process. The three regions are identifiable with distinct boundaries especially in figures 4(b) and (c). These boundaries are somewhat merged 4
  • 6. Nanotechnology 19 (2008) 445702 H Mahfuz et al Figure 4. SEM micrographs of fractured surfaces: (a) neat nylon 6, (b) nylon 6 with silica, and (c) nylon 6 with functionalized silica. The fractured samples were collected after tensile tests of individual filaments in each category. SEM analysis was carried out in a JEOL JSM 5800 scanning electron microscope. in all cases, samples were placed on a carbon double-sided tape and coated with gold to prevent charge build-up during the electron absorption of the specimen. Three to ten kilovolt accelerating voltage was applied to get to the desired magnification. due to the presence of multiple V-notches in the case of neat nylon (figure 4(a)). These V-notches are formed due to transverse cracks originating from the surface craze as seen in huge numbers in figure 4(a). The number of such surface craze reduces significantly when we look at the silated sample. It is also noticed that the size of the V-notch decreases and the notch surface gets smoother as one moves from neat to nanophased samples. The size of the V-notch is usually determined by the speed of the propagating transverse crack and the relaxation rate of the fractured part. If crack propagation is higher than the relaxation rate of the material, it would force an undeveloped V-notch. In that sense i.e. from the size of the V- notch, it seems that crack propagation was highest in the silated samples. Also, the smoother surface is another indication of faster crack growth. Then what caused the higher strength of the nanophased samples? One explanation comes from the surface area of the third region. At the instant before failure the entire breaking load is supported by the third segment which has considerably larger area in the case of figures 4(b) and (c) as compared to figure 4(a). A larger area would require a larger force to break. The other reason is that in the nanophased samples, the initiation of the fracture process (at the first stage) is delayed. This delay can occur due to increased bonding between the particle and polymer as we will show in the next section, and due to a smoother outer surface with much fewer surface crazes, as seen in figures 4(b) and (c). The bonding characteristics of neat and nanophased nylon were next characterized by FTIR and XPS studies. 6. FTIR characterization In order to determine the development of various functional bonds during polymerization, FTIR spectroscopy was per- formed. FTIR is most useful for identifying chemical bonds 5
  • 7. Nanotechnology 19 (2008) 445702 H Mahfuz et al Figure 5. FTIR spectroscopy—(a) neat nylon 6, (b) nylon 6 with 1 wt% silica, and (c) nylon 6 with 1 wt% functionalized silica. FTIR peaks of all the samples were obtained using a Thermo-Nicollet Nexus 670 FTIR spectrometer. Samples were cut into small pieces and directly placed on the attenuated total reflectance (ATR) accessory’s diamond cell. The spectrometer was run at 4 cm−1 resolution step for 128 scans. Before the actual run, the machine was run without the sample to eliminate the environmental effects. Data were acquired and analyzed by using OMNIC software. (a) (b) Figure 6. XPS spectra of pristine and silated particles. A thin layer of each of the powder samples was mounted on a standard stainless steel sample stub using double-sided adhesive tape. Spectra were recorded at an electron take-off angle of 45◦ . Charge compensation was achieved using a VG Scienta FG300 low energy electron flood gun with the gun settings adjusted for optimum spectral resolution. The x-ray source power was 2.8 kW. For survey spectra the spectrometer pass energy was 150 eV and the entrance slit width of the hemispherical analyzer was 1.9 mm, whereas for region spectra values of 150 eV and 0.8 mm were used. Under these conditions the overall spectrometer resolution was 0.55 and 0.35 eV, respectively. that are either organic or inorganic, and its absorption spectrum is almost like a molecular fingerprint. An FTIR experiment with nylon 6 and with functionalized silica nanoparticles is shown in figure 5. Three curves are shown in figure 5; (a) neat nylon, (b) nylon with 1 wt% SiO2, and (c) nylon with 1 wt% functionalized (silated) SiO2. Three basic bonds of nylon 6, i.e. amide N–H at 3297 cm−1 , hydroxyl O–H at 2800– 3000 cm−1 , and carbonyl C=O at 1637 cm−1 were of primary interest. It is seen in figure 5(b) that IR absorbance for each of the three basic bonds has increased significantly, characterized by their sharper and higher peaks. Higher peak and larger area under the curve corresponds to higher absorption of light energy required for excitation. In other words, the IR absorbance is a direct measure of bond strength, indicating that SiO2 reinforcement into nylon was responsible for such an increase. On the other hand, after functionalization of SiO2 particles, i.e. in figure 5(c), it is seen that three basic bond strengths are maintained, while in addition, a siloxane Si–O– Si bond at 1090 cm−1 is formed which was not seen with figures 5(a) or (b). This is what we expected from functionalization; establishing a continuous covalent linkage across the particle (silica) and polymer (nylon 6) interface. 6
  • 8. Nanotechnology 19 (2008) 445702 H Mahfuz et al The area under the FTIR curves is also a qualitative measure of the concentration (number) of the respective bonds. Since there is no change in the frequency, the individual bond strength has not changed, but their number has increased in the nanophased samples. This higher concentration is due to the increase in the number of nucleation sites caused by the presence of nanoparticles during the polymerization process. 7. XPS studies XPS spectra of the pristine and treated silica nanoparticles were recorded using a VG Scienta ESCA300 spectrome- ter [22–24]. The survey spectrum of the pristine nanoparticles (figure 6(a)) shows Si, O, and a minor amount of C, as expected for pure silica. Whereas the spectrum of the treated nanoparticles (figure 6(b)) shows Si, O, and significant amounts of C and N, and thus demonstrates the success of the functionalization reaction. The elemental quantification data are: pristine nanoparticles, Si:O:C = 22.3:72.7:5.0 at.%; treated nanoparticles, Si:O:C:N = 13.4:28.0:51.5:7.2 at.%. The core line binding energies, used as reference are: pristine nanoparticles, Si 2p = 104.0, O 1s = 533.5, C 1s = 285.0; treated nanoparticles, Si 2p = 103.0, O 1s = 532.5, C 1s = 285.0, N 1s = 399.4. The Si 2p and O 1s binding energies of the pristine nanoparticles are consistent with the range of literature values measured for SiO2 [25] and the reduction in the Si 2p binding energy upon functionalization is consistent with the reduced O coordination number in a siloxane type environment. The N 1s binding energy of the functionalized nanoparticles is consistent with N in an amine type environment [22]. 8. Summary It is demonstrated that infusion of pristine silica nanoparticles can modestly increase the strength and modulus of nylon without any loss of fracture strain. FTIR analysis has shown that the mere presence of silica nanoparticles can significantly influence the three primary bonds of nylon 6 resulting in a 28% and 36% increase in modulus and strength, respectively. Once the silica particles are functionalized, the enhancement in primary bonds is still maintained, but a new silane bond is formed which improves the modulus and strength even further; by 55% and 76%, respectively. The source of the improvement is traced to: (i) an increase in the concentration of basic bonds nucleated by the presence of nanoparticles, (ii) formation of siloxane bond, and (iii) changes in the fracture process induced by the infused silica particles. Fracture studies have also revealed that after functionalization nylon 6 seems to exhibit brittle failure modes, but still retains high strength with 30% ultimate strain. Acknowledgments The authors would like to thank the Office of Naval Research, ONR (grant No. N00014-06-1-0696) and the National Science Foundation, NSF (grant No. HRD-976871) for supporting this research. References [1] Kohan M I (ed) 1973 Nylon Plastics (New York: Wiley) pp 3–4 [2] Fornes T D and Paul D R 2003 Polymer 44 3945 [3] Kim S and Young C 2002 Mater. Res. Soc. Symp. Proc. 740 441 [4] Zhang W, Phang I, Shen L, Chow S and Liu T 2004 Macromol. Rapid Commun. 25 1860 [5] Jin L, Bower C and Zhou O 1998 Appl. Phys. Lett. 73 1197 [6] Cooper C A, Ravich D, Lips D, Mayer J and Wagner H D 2002 Compos. Sci. Technol. 62 1105 [7] Thostenson E T and Chou T W 2002 J. Phys. D: Appl. Phys. 35 L77 [8] Gao J, Itkis M E, Yu A, Bekyarova E, Zhao B and Haddon R C 2005 J. Am. Chem. Soc. 127 3847 [9] Mahfuz H, Adnan A, Rangari V K and Jeelani S 2005 Int. J. Nanosci. 4 55 [10] Haggenmueller R, Gommans H H, Rinzler A G, Fisher J E and Winey K I 2002 Chem. Phys. Lett. 330 219 [11] Mahfuz H, Adnan A, Rangari V K and Jeelani S 2004 Composites A 35 519 [12] Mahfuz H, Adnan A, Rangari V, Hasan M M, Jeelani S, Wright W J and DeTeresa S J 2006 Appl. Phys. Lett. 88 083119 [13] Mahfuz H, Hasan M, Rangari V and Jeelani S 2007 Macromol. Mater. Eng. 292 437 [14] Sumita M, Shizuma T, Miyasaka K and Ishikawa K 1983 J. Macromol. Sci. Phys. B 22 601 [15] Lin W and Gowayed Y 1999 J. Appl. Polym. Sci. 74 2386 [16] Xu X, Li B, Lu H, Zhang Z and Wang H 2008 J. Appl. Polym. Sci. 107 2007 [17] Lipitov Y S 1994 Polymer Reinforcement (Toronto: ChemTec) [18] Kerner E 1956 Proc. R. Soc. B 69 808 [19] Lamb G E 1982 J. Polym. Sci. 20 297 [20] Hasan M M, Zhou Y, Mahfuz H and Jeelani S 2006 Mater. Sci. Eng. A 429 181 [21] Ogata N, Dougasaki S and Yoshida K 1979 J. Appl. Polym. Sci. 24 837 [22] Beamson G et al 1990 Surf. Interface Anal. 15 541 [23] Gelius U et al 1990 J. Electron Spectrosc. Relat. Phenom. 52 747 [24] Beamson G and Briggs D 1992 High resolution XPS of organic polymers The Scienta ESCA300 Database (New York: Wiley) [25] NIST XPS database 7

Related Documents