Fabrication of High Impact Resistant Polyimide Nanocomposites with Outstanding Thermomechanical Properties

1. Introduction

The need to discover new materials with improved properties is the driving force for many scientific and technological ventures. In the past decade, the area of conducting polymers has attracted enormous attention from many branches of science: physics, chemistry and several engineering disciplines. The desirable electrical, optical, chemical and electrochemical properties of conducting polymers have made their applicability in many technological areas possible [1,2,3,4,5,6].

Much of the research on conducting polymers, particularly polyaniline, has been focused on understanding its applicability in coatings [7], biosensors [8], light emitting diodes [9], batteries [10] and microelectronics However, despite its desirable applicability, polyaniline is insoluble in many organic solvents such as N-methyl-pyrrolidone (NMP), thereby posing a problem as far as its processing is concerned. In order to circumvent the problem of solubility, previous researchers have explored polyaniline derivatives such as poly(o-anisidine) [12,13,14] and poly(N-ethyl-aniline) [15,16,17]. A tremendous research effort has also been dedicated to the study of polyaniline copolymers due to their solubility in common organic solvents such as N-methyl-pyrrolidone [18,19,20,21,22,23,24,25,26]. Longun et al. [18], synthesized poly(o-anisidine-co-aniline) copolymer via chemical oxidative polymerization and studied its chemical, optical and thermal properties. They reported extended UV-Vis absorption and distinctive chemical features of the copolymer compared to their respective homopolymers.

Previous researchers have studied polyimide/clay nanocomposites and reported increased tensile modulus and thermal stability as well as decreased permeability to gases [27,28,29,30,31,32]. In this study, montmorrilonite clay, Cloisite 30B was modified by poly(N-ethyl-aniline-co-aniline-2-sulfonic-acid) [SPNEAC] copolymer synthesized by chemical oxidative polymerization The resultant modified-clay was used a filler material in the condensation polymerization of polyimide to determine its effect on the imidization process. Thereafter, the effect of poly(N-ethyl-aniline-co-aniline-2-sulfonic-acid)-modified-clay [SPNEAC] on the optical, morphological, chemical, thermal, and dynamic mechanical properties of the polyimide nanocomposites was studied. This systematic study is geared towards understanding the properties of the newly created polyimide nanocomposite materials which combines desirable optical and electrical properties of polyaniline copolymers, and the superior dimensional and thermal stability of clay with the outstanding thermal stability and tensile modulus of polyimide. Incorporation of the polyaniline copolymer/clay filler into the polyimide matrix would also enable the matrix and the copolymer to interact synergistically as a result of their opposite polarity. Polyimide has been shown to be negatively charged [33,34,35] while polyaniline is positively charged [36] and the interaction between the two polymers can enable the polyimide matrix to play the role of polymeric dopant, thereby influencing the optical properties of the polyaniline copolymer. The suggested intimate interaction between polyimide and polyaniline copolymer, in the presence of clay, is also important in regard to the imidization of polyimide, which is thoroughly discussed in this study. It is believed that the intimate interaction between polyimide, clay and the polyaniline copolymer can play an important role in the cyclodehydration process which is essential for efficient and proper processing of polyimide.

2. Materials and Methods

2.1. Materials

The reagents used in this study are as follows: aniline-2-sulfonic acid (95% purity), N-ethyl-aniline (98% purity) and N-methyl-pyrrolidone (99% purity), purchased from Sigma-Aldrich Company. Oxalic acid (99% purity), Dodecyl-benzene sulfonic acid monohydrate (98% purity), 4,4-oxydianiline (99% purity), Pyromellitic dianhydride (99% purity) and Ammonium persulfate (99% purity) were also purchased from Sigma-Aldrich Company. Cloisite-30B clay was purchased from Southern Clay Inc. The Al-2024-T3 Q-panels were purchased from Q-Labs, Cleveland, Ohio. All the reagents listed above are of analytical grade; AR. Doubly distilled and de-ionized water were also used.

2.2. Synthesis of Poly(N-ethyl-aniline-co-aniline-2-sulfonic acid-clay( SPNEAC) copolymer

Synthesis of Poly(N-ethyl-aniline)-copolymer-clay was achieved via a single polymerization process. 5.41 g of aniline-2-sulfonic(A2-S), 3.94 of N-ethyl-aniline(NEA), 1.125 of Oxalic acid (OA), 4.36 of dodecyl-benzene-sulfonic acid monohydrate(DBSA) and 0.585 g of Cloisite 30B Clay were added to a beaker containing 150 ml of doubly distilled water followed by ultrasonication for 3-5 minutes. Stirring was continued for 30 minutes using a mechanical stirrer after which Cloisite-30B clay was added to the resultant mixture followed by stirring for 30 minutes. 4.36 g of ammonium persulfate (APS) was added drop-wise to the mixture followed by stirring for 24 hours. The product was dried in a vacuum oven at 120 ^˚C and then stored in a desiccator.

2.3. Synthesis of poly(amic acid)/Poly(N-ethyl-aniline)-co-aniline-2-sulfonic acid)-Clay(SPNEAC-PAA)

5.1608 g of 4, 4-oxydianiline (ODA was added to 100 ml of N-methyl-pyrrolidone (NMP) in a round-bottom flask and stirred for 30 minutes using a mechanical stirrer. Poly(N-ethyl-aniline-co-aniline-2-sulfonic acid)-clay [SPNEAC] (Figure 1) prepared in the previous step was added to the ODA solution followed by vigorous stirring. After 6 h of stirring, 5.6216 g of pyrromellitic dianhydride, PMDA was added to the mixture and stirring was continued for 15 h. Suspensions were prepared using 0.25, 0.5, 1, 2, 5 and 10 wt.% SPNEAC.

2.3. Preparation of polyimide/Poly(N-ethyl-aniline-co-aniline-2-sulfonic acid)-clay(SPNEAC-PI) nanocomposite films

Polyimide/poly(N-ethyl-aniline-co-aniline-2-sulfonic acid)-clay films were prepared by casting SPNEAC-PAA suspension on glass substrate. Films were prepared by solution casting on glass plate followed by drying in a vacuum oven at120^˚C for 2 h. Final curing was done in a vacuum oven at 200 ^˚C for 1h to produce polyimide nanocomposites (Figure 1 (bottom)) filled with varying amounts of the nanomaterials.

2.4. Characterization

Nicolet 6700 FT-IR instrument equipped with Smart Orbit ATR accessory with diamond crystal, was used to determine the chemical composition of the nanocomposite. ATR analysis was performed over a wavenumber range of 4000 cm^-1 to 400 cm^-1. UV-Visible spectroscopy was used to study the optical properties of polyimide and SPNEAC/polyimide nanocomposites. Measurements were performed by using a U-3000 series spectrometer with wavelength range of 190 nm to 900 nm. UV-Vis samples were prepared by dispersing 0.5 mg of the sample in 10 ml of N-methyl-pyrrolidone (NMP). Solid-state UV-Vis, was performed by using UV-Vis spectrophotometer. Single cell Peltier Accessories was used to measure the transparency of the SPNEAC-PI nanocomposite films. Dynamic Mechanical Spectroscopy (DMS) was used to study the viscoelastic behavior of the nanocomposites. Measurements were performed from 25 ^˚C to 550 ^˚C using EXSTAR6000, Seiko Instruments Inc., under tensile loading mode at a heating rate of 5^˚C/min and a frequency of 1 Hz. Thermal transitions were determined by using DSC 6200; Seiko Instruments Inc. Tests were performed at a heating rate of 10 ˚C/min under a nitrogen atmosphere. Cross-sectional morphology of the films was studied using Scanning Electron Microscopy, SEM Hitachi FEG Model.

SEM samples were prepared by immersion in liquid nitrogen and then fractured using a pair of tweezers to expose the cross-sectional area. A Polaron SC7640 sputter coater was used to sputter coat the samples with silver, Ag.

3. Results

3.1. Optical properties

Figure 3A shows the transmittance of PI and SPNEAC-PI nanocomposite films as a function of composition and wavelength. The percent transmittance of the SPNEAC-PI nanocomposite decreases with increasing copolymer/clay (SPNEAC) loading. Percent transmittance of 85, 81, 66 and 42 % were obtained in the Visible and Near-Infrared (VIS-NIR) region at1000 nm for PI nanocomposite membranes containing 0, 1, 2 and 5 wt. % SPNEAC, respectively. The decreased transmittance in the VIS-NIR region is believed to be due to the optical activity of poly(N-ethyl-aniline-co-aniline-2-sulfonic acid) in this region. The effect of composition of SPNEAC on the optical transmittance at two frequencies (λ) of 500 and 1000 nm is shown on Figure 2B and Table 1. It is shown that neat PI membrane has the highest transmittance of about 62 and 82 % at 500 and 1000 nm, respectively. Interestingly, the optical transmittance for the nanocomposites shows two regimes. The first regime occurs between 450 and 600 nm and it is due mainly to the matrix (PI), while the second regime occurring between 700 and 1000 nm, is believed to be due to the contribution from both SPNEAC and the Matrix. This observation is in agreement with the data shown in Figure 3 which indicates that the transmittance of SPNEAC decreases sharply with increasing wt.% SPNEAC at frequencies of 500 and 1000 nm (Figure 2).

3.1. Functional group analysis and degree of imidization

Figure 3a,b show the ATR spectra of poly(N-ethyl-aniline)-copolymer (SPNEA) and poly(N-ethyl-aniline)-copolymer-clay (SPNEAC) nanocomposite, respectively. The ATR spectrum (Figure 3b) of SPNEAC shows an absorption band at 3430 cm^-1, which is assigned to OH stretching of surface hydroxyl groups. The absorption bands at 2980 and 2850 cm^-1 are assigned to the asymmetric and symmetric stretching of the CH₂ groups in clay. The absorption bands at 3430, 2980 and 2850 cm^-1 confirm the presence of copolymer modified-clay. Absorption bands at 1570 and 1470 cm^-1 due to the quinoid and benzenoid structures, respectively, are shown.

Figure 4 shows the ATR spectra of 30B clay and PI-30B clay nanocomposites. Typical clay peaks are observed at 2850 and 2950 cm^-1 due to methylene and methyl group present in organoclay, and the absorption band at 1360 cm^-1 is due to C-N stretching of the tertiary amine. The 1470 cm^-1 absorption band due to the stretching of the phenyl ring is also observed. Figure 3 and Figure 4 confirm the presence of clay in SPNEAC and PI-clay nanocomposites.

Figure 5 shows the variation of the degree of imidization (DOI) at 150 ^˚C with composition for neat PI and the polyimide nanocomposites membranes. The extent of imidization at 150˚C increased with increasing wt.% nanofiller. Imidization is a very important phenomenon in polyimide and polyimide-based composite processing because it directly affects the thermal and dimensional stability of the resultant material. The degree of Imidization (DOI) was calculated using the area of the imide peak at 1778 cm^-1 with respect to the area of the phenyl peak, 1500 cm^-1 as shown in eq. 1, where “A” denotes peak area and (A₁₇₈₀/A₁₅₀₀)_100% refers to the peak area ratio for 100% imidized polyimide.

(1)

The imidization process, which is temperature dependent, involves the formation of the cyclic imide ring followed by removal of water. One of the important aspects of imidization is the effect of curing temperature and clay weight fraction on the degree of imidization (DOI). Tyan et al. [36] studied the effect of clay on the imidization process and they reported significant enhancement of imidization in the presence of organoclay. Tyan et al. [36] suggested that the enhancement of imidization in the presence of clay is due to increased surface area and availability of active sites for cyclization and dehydration processes provided by clay. As shown in Figure 5, the degree of imidization (DOI) is sensitive to concentration of clay and copolymer-modified clay. In the presence of clay, the degree of imidization (DOI), on average, increased from 53.5% at 0 wt. % to 88% at 5wt. % clay, which represents a 34.5 % improvement over the PI matrix. The degree of imidization for PI composites membranes containing copolymer/clay (60:40) [SPNEAC] is higher than of Clay/PI nanocomposites at all concentration of copolymer/clay. This significant improvement in DOI, which is due in-part to increased polarity and compatibility of the PI matrix and polyaniline copolymer, is attributed to favorable interaction between PI matrix, clay and the copolymer during imidization. The addition of the copolymer/clay (98:2) [SPNEAC2] to poly(amic acid) matrix has a similar effect on imidization as that of 30B clay especially at low concentration ( ≤ 1wt.%), however, at higher SPNEAC2 loading, the effect on imidization decreased slightly possibly due to the heterogenous nature of the nanocomposite.

3.1. Thermal transitions

Figure 6 shows the second scan of the DSC thermograms of neat-PI and SPNEAC-PI nanocomposite, respectively. Values of the transition temperatures were obtained. Neat-PI shows endothermic peaks at ~ 80, 250 and ~415 ^˚C, respectively. The endotherms at 80, 250 and 415 ^˚C are associated with solvent removal, imidization and chain relaxation (glass transition) of polyimide, respectively. SPNEAC-PI also show endotherms in similar regions as neat-PI.

Glass transition often occurs as a step in the DSC thermogram but in this case, it is not so distinct, we therefore obtained the T_g from the change in slope in the glass-rubber transition region and plotted as a function of composition in Figure 7. The T_g obtained using DSC is also compared to that obtained from the dynamic mechanic spectroscopy, DMS as shown in Figure 7. There is a good agreement in the results obtained from both techniques at low filler loadings of 0.25-1.0 wt.% SPNEAC.

The glass transition temperature of SPNEAC-PI nanocomposites in general, decreased with increasing SPNEAC wt.%. DMS results shows that initially at low wt.% SPNEAC < 0.5 wt.%, a slight increase in T_g is obtained over that for neat PI. However, at higher wt.% SPNEAC, ≥ 0.5%, the T_g of the SPNEAC-PI nanocomposites show a decreasing trend. Comparison of the T_g obtained by DSC and DMS shows similar trend of decreasing T_g with increasing SPNEAC wt.%. Better agreement is shown between 0.2 and 0.5 wt.% SPNEAC. Greater disparity in the T_g for neat PI and SPNEAC-PI occurred at filler loadings > 1wt.%.

Figure 8 shows a comparison of the variation of T_g for clay-30B/PI and SPNEAC2-PI with composition. It shows two distinct regimes of behavior of the nanocomposites. At low wt.% of nanofillers <1wt.% Clay-PI shows lower T_g than the neat PI, while SPNEAC2-PI show slightly higher or similar T_g as the neat PI. At higher wt.% nanofillers, ≥ 5 wt.%, clay 30B-PI nanocomposites show a remarkably higher T_g of up to 460˚C. The SPNEAC2-PI nanocomposites, however, show slightly decreasing T_g at higher nanofiller loading ≥1 wt.%. The differences in the chain relaxation behavior of these nanocomposites can be traced to the chemical structure and morphological differences between the two systems.

3.5. Dynamic mechanical analysis

3.5.1. Storage Modulus

The storage modulus for SPNEAC-PI nanocomposites was plotted between 25 ^˚C and 400 ^˚C as shown in Figure 9. The plot of Storage modulus versus SPNEAC loading (Figs. 9 and 10) show slightly increased modulus with respect to SPNEAC loading between 0.25 and 0.5 wt.% where a maximum modulus of 2.3 ± 0.025 GPa was obtained. The lowest modulus of the SPNEAC-PI films was 0.93± 0.020 GPa for 5 wt.% SPNEAC. In the rubbery plateau region, there is a sharp decrease in modulus and this is often associated with the increased chain flexibility and rubber elasticity. However, the rubbery plateau modulus for SPNEAC-PI nanocomposites containing 5 wt.% SPNEAC is significantly higher that that for the neat PI.

Figure 10 chows the variation of the glassy region storage modulus with SPNEAC-PI composition. At low wt.% SPNEAC between 0.25 and 0.5 wt.%, the nanocomposites show slightly higher storage modulus of 1.8 and 2.3 GPa, respectively, which is slightly higher that the value of 1.2 GPa obtained for the neat PI. Increasing the SPNEAC loading to 5 wt.%, resulted to slightly decreased glassy region storage and increased rubbery plateau modulus. The reported increase in the rubbery plateau modulus for SPNEAC-PI containing ≥5 wt.% is due to the presence of physical crosslinks and entangled chains.

3.5.2. Comparison of the dynamic mechanical properties of clay-30B-PI and SPNEAC2-PI nanocomposites

The temperature corresponding to α-transition peak (Figs. 9, 11) is the glass-transition temperature. The glass transition temperature (Figs. 9 and 11), generally decreases with increasing SPNEAC weight fraction. The decrease in T_g at higher SPNEAC loading could point to increased free-volume and chain flexibility. However, at lower SPNEAC and SPNEAC2 concentrations of 0.25 and 0.5 wt.%, a slight increase in the T_g of the nanocomposite over that for the neat PI was observed. It is also noteworthy that the damping ability of the nanocomposite containing 0.25 and 0.5 wt.% SPNEAC and SPNEAC2 are slightly higher than that for the neat PI. Figure 12 shows the variation of the storage modulus for clay 30B-PI and SPNEAC2-PI nanocomposites with composition. It is shown that increasing wt.% of the nanofillers resulted in increased storage modulus. Both nanocomposites show slight decrease in storage modulus, E’ for nanofiller wt.% below 2 wt.% nanofillers. However, beyond 1 wt.% 30B clay and 2 wt.% SPNEAC2, a remarkable trend of increasing storage modulus with increasing wt.% nanofillers is clearly demonstrated. SEM images of the nanocomposites (Figure 12) show an open morphology at high SPNEAC2 loading (≥ 0.5 wt %). For the nanocomposites containing 0.5 wt.% SPNEAC2, loosely bound nano-sized spheres are evident.

3.5.3. Damping behavior and impact energy

Damping is a result of viscoelastic molecular relaxation phenomenon which is more pronounced at the glass-transition temperature. The height of tan δ peak (Figs. 9 &11), which is a measure of damping ability, increases with increasing SPNEAC loading, except for 0.25wt.%-SPNEAC-PI. The height of tan δ peak increases from 116.5 for neat-PI to a maximum of 137.28 for 2wt%-SPNEAC-PI. This increase in damping at higher SPNEAC loading can be attributed to the open and connected morphology of the nanocomposites which is believed to impact the relaxation behavior of the nanocomposite. This phenomenon also associated with the observed decreasing T_g values.

The area under tan d versus temperature curve can be related to the impact energy in accordance with Wada and Kasahara, (equation 2) (38-39). The area under the a-transition peak was calculated for the SPNEAC nanocomposites and analyzed as shown on Figure 13 and tabulated in Table 2. The impact energy of the nanocomposites increased with increasing wt.% SPNEAC. At 10 wt.% SPNEAC the impact energy of the nanocomposites is about 4 times that for the neat PI.

$${\sigma }_{imp}\propto {\displaystyle \underset{T_o}{\overset{T_t}{\int }}\left(\tan \delta \right)}\partial T$$

(2)

3.6. SPNEAC2-PI Nanocomposite Morphology

Figure 14 shows the SEM images of neat-PI, and SPNEAC2 nanocomposites containing 0.25, 0.5 and 5wt%-fillers, respectively, at 50,000 x magnification. The SPNEAC2-PI nanocomposite film has a dense fiberlike columnar morphology at lower loading of SPNEAC2 (≤ 0.5wt %), however, the morphology becomes spheroidal with increasing SPNEAC loading. For the 0.5wt%SPNEAC2-PI nanocomposite, discrete and overlapping fibroids are visible throughout the morphology. These features are closely connected, thereby resulting in dense spiky columnar morphology. At 2 and 5 wt.% SPNEAC2, the morphology becomes more spherical and connected but uniformly distributed throughout the cross-sectional area. It is suggested that these particulates contribute to the enhanced storage modulus observed at 5wt%-SPNEAC2. For the neat-PI films, an amorphous-like diffuse morphology with no distinct features is observed.

The cross-sectional morphologies of 30B-PI and SPNEAC2-PI containing 10 wt.% nanofillers are shown in Figure 15. While the SPNEAC2-PI cross-sectional morphology shows close packed spherical particulate morphology with uniformly dispersed spheroidal, the cross-section morphology for 30B-PI shows a dense morphology with spiky edges of 30B clay sandwiching smooth continuous PI matrix.